Object-initial languages

[Bittor Hidalgo]

Thanks for your answer. I didn't know my message had circulated in the list
as I desired (I haven't received it as I expected).

Anyway, my hypothesis wouldn't divide languages between OV and SV languages
as you suggest, but between XV and VX languages (Vennemann). Sure, I didn't
explain myself clearly enough.

The grammaticalization of subject position in XV languages gives SV. And
this S, at least when pronominal, deslexicalized, functions just as a
morpheme of the verbal complex. Just as it happens in Basque where the
grammaticalization process has been far completed:

NOA (I go) = N (short of NI(nee) = I) +  OA (radical of the verb JOAN = GO)
HOA (you go) = H (short of HI(hee) = you) + OA
GOAZ (we go) = G (short of GU = we) + OA + Z (plural marker) // ...

If "I GO" functions as an inseparable verbal complex (just as NOA in
Basque), it constitutes a new V(SV) that again will tend to go proclitic, or
enclitic, with respect to a new X complement (inside a minimal
intonation -or accentual, or phonological- unit). English  makes VX [(I GO)
HOME], that is [V(SV) + X], and not normally "HOME I GO", because English
now works as a VX language. Meanwhile, Basque makes XV "ETXERA NOA" (I go
home), and not normally "NOA ETXERA", except if the verb NOA goes enclitic
with respect to a new X on its left "MIKELEKIN NOA ETXERA" (I go home with
Michel).

On the other side, there would be the languages that always have been VX. If
this languages have grammaticalized the S + V complex, it will have been in
VS position (not SV like English or French, former XV languages), and this
new V complex will go again proclitic with respect to a new X: [V(V+S) + X].

Changes could happen. For example, languages in contact could change because
of mutual influence from being XV to VX, or vice versa. But I think that
when a language becomes a culture language, when it is written, and long
sentences are commoner, even XV or VX, the normal position for long
complements (except specifically topical ones) is after the XV or VX
complex. And this option maight facilitate the conversion of XV languages in
VX languages (with intermediate steps -English, French, and in a diferrent
level Spanish, Italian,...-), and on the contrary, difficult the conversion
of VX languages in XV.

Rigid verb final culture languages (do they exist?) are a very serious
counterexample for this hypothesis. I don't know exactly what happens in
languages like Japanese or Korean, when a common sentence must have a lot of
complements and subordinate sentences inside (let's say, in a newspaper,
where language must be fluid). Basque purist syntacticians, who argued in
favor of a rigid verb final position, ask writers to use only short
sentences, because if not everybody knows that sentences become, if not
incomprehensible, at least very difficult to process, exhausting and
fatiguing.

You can see that my interest in word order question is really big.

Thanks

Bittor



---------------------------------------------
Victor Hidalgo Eizagirre
k/ Buztintxulo, 72, behea
20015 Donostia
Tfnoa eta faxa: 943-282192
posta-e: v.hidalgo at euskalnet.net
----- Original Message -----
From: "Brian MacWhinney" <macw at cmu.edu>
To: "Bittor Hidalgo" <v.hidalgo at EUSKALNET.NET>
Sent: Wednesday, October 30, 2002 12:34 AM
Subject: Re: Object-initial languages


Dear Viktor,
  I think you are basically right.  I think Theo Vennemann tried to make
this point many years ago, although he wandered into other issues too.  I
particularly like this idea from the viewpoint of child language.  To
summarize, you are saying that languages tend to go for either OV or SV and
that this is the fundamental divider.  Yes, that makes a lot of sense.
Hungarian is, of course, a great example of a language stuck between the
two, but choosing OV for objects without articles and going for SV in all
other cases.  The motivation for OV is usually thought to be the
incorporation of the object into the verb.

--Brian MacWhinney