[Prev][Next][Index]

CSLI Calendar, 15 February, vol. 5:17


       C S L I   C A L E N D A R   O F   P U B L I C   E V E N T S
-----------------------------------------------------------------------------
15 February 1990                    Stanford                   Vol. 5, No. 17
_____________________________________________________________________________

    A weekly publication of the Center for the Study of Language and
Information (CSLI), Ventura Hall, Stanford University, Stanford, CA 94305-4115
			     ____________

          CSLI ACTIVITIES FOR THIS THURSDAY, 15 FEBRUARY 1990

12:00 noon		TINLunch
      Cordura 100	Reflexives and Subject Antecedents
			Annie Zaenen
			(zaenen.pa@xerox.com)
			Abstract in last week's Calendar

 2:15 p.m.		CSLI Seminar
      Cordura 100	Characteristics of Different Neurogenic
			Communication Disorders
			Terry Wertz
			Chief, Audiology and Speech Pathology
			VA Medical Center
			Abstract in last week's Calendar

 3:30 p.m.		TEA
      Cordura 117
      (2d lounge)

 4:15 p.m.		CSLI Colloquium
      Ventura 17	Some Restructuring Effects in German
			Jaklin Kornfilt
			University of Syracuse
			Abstract below
			     ____________

			     ANNOUNCEMENT

There will be no TINLunch, CSLI Seminar, or CSLI Colloquium next
Thursday, 22 February.  TINLunch will resume on Thursday, 1 March, and
will be announced in next week's Calendar.
			     ____________

		     THIS WEEK'S CSLI COLLOQUIUM
		 Some Restructuring Effects in German
			   Jaklin Kornfilt

Some recent work (e.g., Kayne, den Besten) has suggested that
restructuring (or its equivalents, like multidimensional
representations) can be avoided by using mechanisms such as
head-to-head movement, scrambling, etc.  This paper looks at three
construction types in German, all of which involve apparent locality
violations, and argues that these constructions cannot be accounted
for by the proposed alternatives to restructuring, in turn triggered
by verb raising.  A treatment of restructuring in terms of
morphologically derived and syntactically conflated categories (cf.
also recent work by Eric Reuland) will be explored.

The three construction types are illustrated by the following
examples:
 
A.  Clitic climbing:

(1) dass uns [1] der Hans vergessen hat [PRO e[1] sein Auto zu zeigen]
    "that Hans has forgotten to show us his car"

B.  "Long" passives:

(2) dass das Auto [1] vergessen wurde [(PRO) e[1] zu reparieren]
    "that it was forgotten to repair the car"

C.  Preposed verb clusters

(3) [zu reparieren vergessen] hat der Hans das Auto nicht
    "Hans hasn't forgotten to repair the car"

Finally, the question is addressed of why subject control verbs as
well as some dative control verbs allow restructuring, while this is
never possible with accusative control verbs.
			     ____________

		 SEMINAR ON ISSUES IN LOGICAL THEORY
			    Philosophy 396
		   How Many Real Numbers Are There?
			   Paul C. Gilmore
		    Department of Computer Science
		    University of British Columbia
		   Thursday, 15 February, 3:45 p.m.
			     Cordura 100

In this lecture, a natural deduction-based set theory NaDSet will be
used to provide a formal framework for logical foundations of category
theory admitting genuine instances of self-membership.  However, usual
diagonal arguments leading to inconsistency are blocked, including
Cantor's argument for uncountability of the set of real numbers.

The following week, 22 February, David Israel will present material on
proof theory and meaning, inspired by chapter III.8 of the _Handbook of
Philosophical Logic_.
			     ____________

			SYMBOLIC SYSTEMS FORUM
	      Why are Computer Programs so Complicated?
		       John Lamping, Xerox PARC
			(lamping.pa@xerox.com)
		   Thursday, 15 February, 4:15 p.m.
			Building 60, Room 61G
         
One reason for the complexity of computer programs is that current
programming languages don't allow programmers to say what they mean.
Rather than letting the programmer say what a program is supposed to
accomplish, the languages force the programmer to give the computer a
set of instructions on what to do.  The intent of the program is
relegated to the comments.

Functional programming and logic programming take a step in the right
direction, but a much smaller step than it might first appear.  While
both approaches putatively let the programmer specify what the program
should compute, the catch is that the way the programmer writes down
the specification completely determines how (and thus how quickly) the
computer will compute the result.  To get acceptable performance on
anything but toy programs, the description of the result usually must
be contorted into something that looks suspiciously like instructions
on how to compute the result.  The intent of the program is still
relegated to the comments.

What if the description of what to compute could be decoupled from how
the result would be computed?  It's a bit much to expect a system that
automatically figures out a good way to compute a result, but that job
can be left to the programmer.  The programmer would give a high-level
description of what the result should be, and annotate that
description with instructions on how it should be implemented.

In this talk, I'll give some examples of what this kind of programming
might look like and suggest how to build a programming system that
supports it.  In such a system, the programmer would talk explicitly
about the representation relationships within the program and between
the program and its implementation.  The key to designing such a
system seems to be a good theoretical understanding of such
representation relationships.
         
The following week, 22 February, Philip Cohen, SRI International and
Stanford Linguistics, will talk.  Title: Intention, Commitment, and
Communication.
         		     ____________

		   PHILOSOPHY DEPARTMENT COLLOQUIUM
		      Spinoza's Theory of Error
			    Yirmiahu Yovel
		     Hebrew University, Jerusalem
		    Friday, 16 February, 3:15 p.m.
			Building 90, Room 92Q

No abstract available.
			     ____________

		  LINGUISTICS DEPARTMENT COLLOQUIUM

Contrary to earlier announcements, there will be no Linguistics
Department Colloquium this Friday, 16 February.  John Nerbonne will
talk next Friday, 22 February, and the second meeting of the linking
symposium will be on Friday, 9 March.
			     ____________

		 LANGUAGE ACQUISITION INTEREST GROUP
     Temporality in Untutored Adult Second Language Acquisition:
		 Functional Approach to Data Analysis
			 Marya Teutsch-Dwyer
		   Tuesday, 20 February, 12:00 noon
		     Building 100, Greenberg Room

No abstract available.
			     ____________