Re: R: Re: Greenberg's universals

"SMITH,MARCUS ANTHONY" wrote:

> On Wed, 13 Sep 2000, J Matthew Pearson wrote:
>
> > "SMITH,MARCUS ANTHONY" wrote:
> >
> > > On Wed, 13 Sep 2000, J Matthew Pearson wrote:
> > >
> > > I think this stems from a misunderstanding of a previous conversation. I
> > > mentioned that SVO languages tend to have *definite* articles (ie,
> > > distinct from demonstratives), while the others often have no articles or
> > > use the demonstratives for this role.
> >
> > If by "the others" you mean all other word order types, I don't think that's a valid
> > tendency.  The language groups I mentioned in my previous post, with verb-initial
> > order, all have definite articles lexically distinct from demonstratives.  Moreover,
> > it's not hard to think of languages (Indonesian) and language groups (Slavic,
> > Chinese, Bantu, perhaps Algonquian) with verb-medial order and no articles.
>
> But that's typology for you.  No such thing as a solid universal -- that's
> why I used the words "tend" and "often".
I understand that.  But even by the loose standards of typology, I find no evidence for a
one-to-one correlation between SVO order and definite articles.  Even "tend" and "often"
are way too strong, I think.


> What I stated was a
> generalization that was given to me in a typology course.  I have very
> little exprience with VSO languages, and the one's I have looked at do, in
> fact, lack definite articles: Eyak, Haida, Tlingit, and I can't recall
> seeing a definite article in any Athabaskan languages either.
Kwakwala (a.k.a. Kwakiutl), a Wakashan language spoken in the same general area as Haida
and Tlingit, is verb-initial *with* articles, as are the nearby Salishan languages.  The
only Athapaskan language I know anything about is Navajo, which is SOV.  Which ones did
you have in mind?

Hmm, I just remembered the Oto-Manguean languages, which are mostly VSO and lack definite
articles.  But other than those, all the verb-initial languages I'm familiar with have
them.


> > The only generalization I would hazard is that verb-final languages tend not to have
> > articles--and even that is just a guess on my part, based on casual observation.
>
> I'd have to agree with that -- based on casual observation as well.
Matt.