[Lingtyp] 'basic word order' in the MP

Kilu von Prince watasenia at gmail.com
Thu Mar 24 08:30:53 UTC 2016


Hi Fritz,

I think you are right, the MP does not really tell us much about the order
of heads relative to their complements, but I think it does make some
predictions about word order in general. For example, an order of VSO,
where the verb and the object are separated by another constituent should
have to involve more movement than an order of, for example, VOS and should
therefore be more marked and typologically less frequent. This prediction
is not borne out by the typological data in WALS and I'm not aware of any
commentary on this.

Having said that, the MP is by no means the favourite framework among
generative syntactitians. Principles and Parameters are still quite popular
and my impression is that many generative syntactitians who care about
typology at all favour the asymmetry hypothesis, which is based on G&B. For
recent contributions in that area, see Hubert Haider's work on head
directionality. This approach was essentially inspired by typological
word-order distributions and does a decent job of addressing them.

Best,
Kilu

On Wed, Mar 23, 2016 at 9:56 PM, David Beck <dbeck at ualberta.ca> wrote:

> Hi, Fritz
>
> I haven’t really been tracking the theory much since the late ‘90s, but
> I’m not sure that it is true that linearization does “take place” at PF,
> since a tremendous amount of argumentation (pretty much all of it) in this
> syntactic framework still depends on moving things around in the syntactic
> tree (which I always understood to be pre-PF) to get the surface order. So
> even though X-bar is offically gone, most of the papers I see still use the
> same assumptions about binary branching and head-complement ordering at
> D-structure, and use the same kind of movement rules (I guess it’s only one
> kind these days, but it does pretty much the same job). These papers
> generally argue for a particular phrase being moved or not in the syntactic
> tree based on where it is linearized in the sentence.
>
> But even if the final tree-structure before Spell-Out reflects the linear
> order of the sentence, I think that your point is correct, ‘basic order’
> isn’t going to be a very coherent or clearly defined concept, since
> surface-order is going to be determined on a construction-specific basis
> (depending on which heads have to move to check which phi-features, what
> barriers there are in a given construction to the movements, which landing
> sites are available, etc.). So Irish is VSO in a certain type of clause
> because the requirements for feature-checking move V ahead of S, and that
> movement might be blocked in another type of clause where the conditions
> are different, or other conditions might move-S ahead of V in another
> construction, etc. I would guess that the only way to postulate a “basic”
> order for the language would be to talk in terms of the order that requires
> the fewest movements, or requires the fewest “special” conditions
> regulating movement (which is kind of the way everyone else does it, too).
>
> David
>
>
> > On Mar 23, 2016, at 1:47 PM, fjn at uw.edu wrote:
> >
> > Dear Lingtyp,
> >
> > I’m not sure how many formal linguists read this list, but some of you
> might have thought about this issue or have seen a reference. Do you know
> if anybody has taken on the question of 'basic word order' (VSO, OSV, etc.)
> within the Minimalist Program? It is not obvious that it is still a
> coherent notion. Given the abandonment of the level of D-structure and of
> X-bar theory and given that it is generally assumed (right?) that
> linearization takes place in PF, there is no obvious level or stage in the
> grammar that would identify Irish as VSO, Japanese as SOV, etc. While the
> mechanics of PF-syntax (at least to me) are quite murky, it seems like you
> would have to introduce something like extrinsic ordering in that component
> to postulate a 'basic order' for a particular language. For a language with
> only one ordering of S, O and V there is no issue, but every language, I
> think, allows more than one ordering.
> >
> > Any thoughts on all of this?
> >
> > Thanks,
> >
> > —fritz
> >
> > Frederick J. Newmeyer
> > Professor Emeritus, University of Washington
> > Adjunct Professor, U of British Columbia and Simon Fraser U
> >
> > _______________________________________________
> > Lingtyp mailing list
> > Lingtyp at listserv.linguistlist.org
> > http://listserv.linguistlist.org/mailman/listinfo/lingtyp
>
> _______________________________________________
> Lingtyp mailing list
> Lingtyp at listserv.linguistlist.org
> http://listserv.linguistlist.org/mailman/listinfo/lingtyp
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/lingtyp/attachments/20160324/3a340fb8/attachment.htm>


More information about the Lingtyp mailing list