LINGUIST List 2.447

Wed 28 Aug 1991

Calls: Lexical Workshop

Editor for this issue: <>


  1. , summary announcement

Message 1: summary announcement

Date: Tue, 27 Aug 91 17:30 GMT
From: <>
Subject: summary announcement
 Call for Papers
 Research Institute for Language and Speech
 University of Utrecht
 December 9 - 11, 1991
In December a workshop will be held in Utrecht on the issues of what type of
lexical specifications we need in a generative grammar and by what principles
this information is projected onto syntactic configurations, i.e. how lexical
insertion is executed. A number of invited speakers have agreed to present
their ideas on topics related to these questions. They include: Hans Bennis
(Leiden), Joe Emonds (Washington-Seattle), Jane Grimshaw (Brandeis), Hubert
Haider (Stuttgart), Lars Hellan (Trondheim), Rita Manzini (UC London), Bozena
Rozwadowska (Wroclaw), Margaret Speas (Amherst), Edwin Williams (Princeton).
The workshop will have a limited number of slots available for people who want
to present 35-minute papers on issues covered by the topic description of the
workshop (see below). Those who want to present a paper should submit an
abstract (max. 2 pages) to the organizers no later than October 20, 1991.
Notice of acceptance will be given very shortly afterwards.
Organizing committee: Peter Coopmans, Martin Everaert, Jane Grimshaw
For further information, please contact
Research Institute
for Language and Speech phone: 31-30-392006
University of Utrecht fax: 31-30-333380
Trans 10, 3512 JK Utrecht email:
The Netherlands otshutruu59.bitnet
 Workshop on Lexical Specification and Lexical Insertion
This workshop addresses the issues of what type of lexical specifications
we need in a generative grammar and by what principles this information is
projected onto configurations, i.e. how lexical insertion is executed. The
workshop will particularly focus on the question of what the syntactic
consequences are of choices that are made with respect to the lexical
specifications of heads. The general assumption is that 'lexical insertion'
means replacing a X-o position in a phrase-marker by a lexical formative.
Simultaneous insertion of more than one X-o, as in the case of phrasal idioms,
is probably exceptional. Lexical formatives contain inherent and contextual
features. Through the projection of the contextual features of a X-o (= head)
the structural properties of phrases are determined. Although the Projection
Principle is one of the core principles of UG, very little explicit discussion
has been devoted to determining what contextual features we have and what
'projection' actually means and how it is executed.
 Since Chomsky's Aspects several forms of lexical encoding have been proposed
in the literature, such as 'subcategorization', 'selectional restrictions',
'theta grid/argument structure', 'predicate argument structure', 'grammatical
function assignment', 'lexical conceptual structure', and more recently 'event
structure'. It is clear that some of these notions are mutually incompatible,
others are not, and that, furthermore, there is redundancy. Despite the work
that has been done, it is still very much open to debate to what extent the
lexical specifications that have been offered in the literature are necessary
and sufficient to fully specify the structural configurations in which a head
can appear.
 Specific questions that come to mind are:
- How are theta roles mapped onto syntactic configurations? Through specific
 realization rules, universal mapping principles, or through mediation of
 aspectual marking?
- To what extent are the types of lexical specifications autonomous, or are
 they linked? For instance is theta-specification linked to aspectual
 information or to specifications of Case information?
- How is,in general, the projection part of the Projection Principle executed?
- Should we distinguish inherent theta-grids and non-inherent theta-grids for
 semi-adjuncts? How is this encoded, and what syntactic consequences can be
 derived from this?
- Do we need a lexical specification of aspectual information, and if so what
 would such a specification look like?
- Is there a distinct level of lexical representation on which 'lexical
 collocational restrictions' are encoded? And if so, in what way are they
 expressed grammatically?
- To what extent can subcategorization be reduced to other selectional
 properties of predicates?
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue