Xref: utzoo comp.ai:1282 sci.lang:1795
Path: utzoo!mnetor!uunet!kddlab!icot32!nttlab!gama!etlcom!hasida
From: hasida@etlcom.etl.JUNET (Koiti Hasida)
Newsgroups: comp.ai,sci.lang
Subject: Re: words order in English and Japanese
Message-ID: <5494@etlcom.etl.JUNET>
Date: 20 Jan 88 04:49:45 GMT
Reply-To: hasida%etl.jp@relay.cs.net
Organization: Electrotechnical Laboratory, Tukuba Science City
Lines: 43


In article <1671@russell.STANFORD.EDU> nakashim@russel.stanford.edu write:
>To avoid this kind of real-time misunderstanding, English must
>transfer essential information first, refining it later.
>
>In farming, on the other hand, there are lots of time.
>So, in Japanese, you can specify lots
>of objects first and then combine them together at the end with
>several modifications added further.

A standard comment by linguists (especially syntacticians) would be
that this kind of global word-order variation is largely accounted for
in more syntactic terms. According to Chomskyan parameter-setting
approach, for instance, the word-order variation between head-initial
languages (such as English and French) and head-final languages (such
as Japanese and Korean) is attributed to the value of a single binary
parameter associated with X-bar component of syntax. This parameter is
turned on in one class of languages, and off in the other. The order
between verb and its object, the choice between preposition and
postposition, etc. follow from this single decision.

Whether or not this parameter is innate is irrelevant here.  I would
rather like to reduce this parameter to more fundamental computational
terms, instead of postulating it to be preprogrammed. But such an
account of mine would be as syntactic as is Chomskyan approach.  The
point is that the set of syntactic constraints has some internal
dependence in its own right without recourse to semantics or
pragmatics, and thus a small decision on a piece of syntactic
constraint influences a lot of other part of syntax.  Even if your
pragmatic theory were basically right, it is imperfect; a more
syntactic aspect such as mentioned above should be taken into account
as well. For instance, your theory would fail to explain why English
employs prepositions rather than postpositions despite the fact that
in a prepositional phrase the object noun phrase tend to convey more
information than the preposition does.

A fatal defect of your theory is that you only refer to modern
English. Old English and its antecedent languages exhibit word-order
variations different from that of modern English. Pragmatic
requirement of hunting situation thus seems less relevant to
word-order than you suppose.

HASIDA, Koiti	Electrotechnical Lab.	hasida%etl.jp@relay.cs.net