7.38, Sum: LP and typed feature logics

Wed Jan 10 04:53:14 UTC 1996

---------------------------------------------------------------------------
LINGUIST List:  Vol-7-38. Tue Jan 9 1996. ISSN: 1068-4875. Lines:  128

Subject: 7.38, Sum: LP and typed feature logics

Moderators: Anthony Rodrigues Aristar: Texas A&M U. <aristar at tam2000.tamu.edu>
            Helen Dry: Eastern Michigan U. <hdry at emunix.emich.edu>
            T. Daniel Seely: Eastern Michigan U. <dseely at emunix.emich.edu>

Associate Editor:  Ljuba Veselinova <lveselin at emunix.emich.edu>
Assistant Editors: Ron Reck <rreck at emunix.emich.edu>
                   Ann Dizdar <dizdar at tam2000.tamu.edu>
                   Annemarie Valdez <avaldez at emunix.emich.edu>

Software development: John H. Remmers <remmers at emunix.emich.edu>

Editor for this issue: aristar at tam2000.tamu.edu (Anthony M. Aristar)

---------------------------------Directory-----------------------------------
1)
Date:  Tue, 09 Jan 1996 22:22:47 CST
From:  fcosws at prairienet.org (Steven Schaufele)
Subject:  LP and typed feature logics: summary

---------------------------------Messages------------------------------------
1)
Date:  Tue, 09 Jan 1996 22:22:47 CST
From:  fcosws at prairienet.org (Steven Schaufele)
Subject:  LP and typed feature logics: summary

Four months ago (LINGUIST 6-1218) i posted the following query:

> For some time now, I've been excited about the development, within
> such frameworks as HPSG, of the ability to describe
> dominance/constituency relations by means of formal feature logics.
> I've been wondering about the possibility of also describing linear
> order/precedence by similar means.  Has anybody out there done any
> work in this direction, or know of anyone who has?

I would like to thank the following scholars for literature suggestions:

Mark Baltin <baltin at acfcluster.nyu.edu>
Claire Gardent <claire at coli.uni-sb.de>
Kevin Russell <krussll at cc.umanitoba.ca>
Frank Richter <fr at sfs.nphil.uni-tuebingen.de>
Manfred Sailer <mf at sfs.nphil.uni-tuebingen.de>
Richard Clay Taylor <rctst3+ at pitt.edu>
Koen Versmissen <Koen.Versmissen at let.ruu.nl>

Baltin and Taylor both referred me to Richard Kayne's recent LI monograph
The Antisymmetry of Syntax.  I sent them the following message in reply,
which i am reproducing here in the interests of clarification:

> Yes, i'm acquainted with Kayne's Antisymmetry hypothesis; perhaps i
> should have mentioned it in my initial posting.  Unfortunately, while
> i think it's a truly valiant attempt, Kayne's results are so contrary
> to everything i know from long experience with SOV languages that i
> regard his theory as very much in need of huge amounts of supporting
> evidence to reach even the level of plausibility.  Frankly, i think
> the attempt to account for the `horizontal' dimension (linear order)
> of linguistic structure by exactly the same sort of feature systems
> that have proven so fruitful in describing the `vertical' dimension
> (dominance) is misguided.  If a feature-logic description of linear
> precedence turns out to be feasible i think that would be great, but i
> strongly suspect it will not be of exactly the same sort as the sort
> of thing used in unification grammars to handle dominance relations.
>
> What i'm really looking for is not a formal logic that says `because
> of premises A, B, and C UG can only generate orders M, N, and O' but
> rather a feature-logic system that would allow me to describe the
> variety of constituent-order typological possibilities actually
> attested in the languages of the world, including discontinuous
> constituents.  And i'm afraid Kayne's approach is too much of the
> first kind and therefore has little if any ability to fulfill the
> second.
>
> In fact, a certain amount of my own research in the past year and a
> half has been devoted to poking holes in Kayne's hypothesis.

A further clarificatory statement that i sent to Kevin Russell:

> i should perhaps make it clear that i am motivated by concern not
> only for concatenation processes but for wrapping and/or other
> linearization processes involving node-deletion to account for
> discontinuous constituency (cf. my paper in the 1993 ESCOL
> Proceedings).  What i would really like is a feature-logic formalism
> that not only permits the exhaustive description of all attested
> `normal' concatenation processes but, for instance, would permit a
> statement to the effect that `in this grammar only NP nodes can be
> deleted/erased'.

Apart from Kayne's book, other, more promising works suggested to me were:

Bird, Steven.  1990.  Constraint-Based Phonology.  Diss., Edinburgh.

Manandhar, Suresh.  1995.  Deterministic consistency checking of LP
constraints.  In Proceedings of the Second Conference of the European
Chapter of the Association for Computational Linguistics, Dublin.

Richter, Frank & Manfred Sailer.  1995.  Remarks on Linearization:
Reflections on the Treatment of LP-Rules in HPSG in a Typed Feature Logic.
Magisterarbeit, Tuebingen.

(available via WWW from Richter's homepage:
http://garside.sfs.nphil.uni-tuebingen.de/~fr)

Koen Versmissen also has a paper on the subject available via WWW:
ftp://ftp.let.ruu.nl/pub/users/versmissen/dyana.ps.gz

Again, thanks to all who responded.  Obviously, i'm not the first or the
only linguist to get intrigued by this possibility, and i'm glad some
people have been working on it who know more than i about typed feature
logics!

Best,
Steven
- -------------------
Dr. Steven Schaufele
712 West Washington
Urbana, IL  61801
217-344-8240
fcosws at prairienet.org

**** O syntagmata linguarum liberemini humanarum! ***
*** Nihil vestris privari nisi obicibus potestis! ***
------------------------------------------------------------------------
LINGUIST List: Vol-7-38.