nl-kr-request@CS.ROCHESTER.EDU.UUCP (06/04/88)
NL-KR Digest (6/03/88 17:12:13) Volume 4 Number 56
Today's Topics:
Language Learning: BBS Call for Commentators
BBN AI Seminar: Josh Tennenberg
CFP - Information in Text
From CSLI Calendar, May 26, 3:30
From CSLI Calendar, June 2, 3:31
Re: Language Learnability: BBS Call for Commentators
Submissions: NL-KR@CS.ROCHESTER.EDU
Requests, policy: NL-KR-REQUEST@CS.ROCHESTER.EDU
----------------------------------------------------------------------
Date: Wed, 25 May 88 00:23 EDT
Date: Wed, 25 May 88 00:23 EDT
From: Stevan Harnad <harnad@mind.UUCP>
Subject: Language Learning: BBS Call for Commentators
Below is the abstract of a forthcoming target article to appear in
Behavioral and Brain Sciences (BBS), an international journal of "open
peer commentary" in the biobehavioral and cognitive sciences, published
by Cambridge University Press. For information on how to serve as a
commentator or to nominate qualified professionals in these fields as
commentators, please send email to: harnad@mind.princeton.edu
or write to: BBS, 20 Nassau Street, #240, Princeton NJ 08542
[tel: 609-921-7771]
______________________________________________________________________
The Child's Trigger Experience: "Degree-0" Learnability
David Lightfoot
Linguistics Department
University of Maryland
A selective model of human language capacities holds that people come
to know more than they experience. The discrepancy between experience
and eventual capacity is bridged by genetically provided information.
Hence any hypothesis about the linguistic genotype (or "Universal
Grammar," UG) has consequences for what experience is needed and what
form people's mature capacities (or "grammars") will take. This BBS
target article discusses the "trigger experience," i.e., the experience
that actually affects a child's linguistic development. It is argued
that this must be a subset of a child's total linguistic experience
and hence that much of what a child hears has no consequence for the
form of the eventual grammar. UG filters experience and provides an
upper bound on what constitutes the triggering experience. This filtering
effect can often be seen in the way linguistic capacity can change between
generations. Children only need access to robust structures of minimal
("degree-0") complexity. Everything can be learned from simple, unembedded
"domains" (a grammatical concept involved in defining an expression's
logical form). Children do not need access to more complex structures.
--
Stevan Harnad ARPANET: harnad@mind.princeton.edu or
harnad%princeton.mind.edu@princeton.edu UUCP: princeton!mind!harnad
CSNET: harnad%mind.princeton.edu@relay.cs.net
BITNET: harnad%mind.princeton.edu@pucc.bitnet
------------------------------
Date: Thu, 26 May 88 14:21 EDT
From: Marc Vilain <MVILAIN@G.BBN.COM>
Subject: BBN AI Seminar: Josh Tennenberg
BBN Science Development Program
AI Seminar Series Lecture
ABSTRACTION IN SYMBOLIC PLANNING
Josh Tennenberg
University of Rochester
(josh@cs.rochester.edu)
BBN Labs
10 Moulton Street
2nd floor large conference room
10:30 am, Tuesday May 31
The use of abstraction in planning is explored in order to simplify the
task of reasoning about the effects of an agent's actions within a complex
world. Two representational issues emerge which form the basis of this
research. First, the abstract views must sanction plan construction for
frequently occurring problems, yet never sanction the deduction of
contradictory assertions. Second, a correspondence between the abstract
and concrete views must be maintained so that abstract solutions bear a
precise relationship to the concrete level solutions derived from them.
These issues are explored within two different settings. In the first, an
abstraction hierarchy is induced by relaxing some of the constraints on the
application of actions. In the second, a predicate mapping function is
defined which extends the notion of inheritance from object types to
arbitrary relations and actions.
------------------------------
Date: Fri, 27 May 88 07:38 EDT
From: Robert Amsler <amsler@flash.bellcore.com>
Subject: CFP - Information in Text
UNIVERSITY OF WATERLOO
CENTRE FOR THE NEW OXFORD ENGLISH DICTIONARY
4TH ANNUAL CONFERENCE
CALL FOR PAPERS - CALL FOR PANELISTS
INFORMATION IN TEXT
October 27-28, 1988
Waterloo, Canada
This year's conference will focus on ways that text stored as electronic
data allows information to be restructured and extracted in response to
individualized needs. For example, text databases can be used to:
- expand the information potential of existing text
- create and maintain new information resources
- generate new print information
Papers presenting original research on theoretical and applied aspects of
this theme are being sought. Typical but not exclusive areas of interest
include computational lexicology, computational linguistics, syntactic
and semantic analysis, lexicography, grammar defined databases, lexical
databases and machine-readable dictionaries and reference works.
Submissions will be refereed by a program committee. Authors should send
seven copies of a detailed abstract (5 to 10 double-spaced pages) by
June 10, 1988 to the Committee Chairman, Dr. Gaston Gonnet, at:
UW Centre for the New OED
University of Waterloo
Waterloo, Ontario
Canada, N2L 3G1
Late submissions risk rejection without consideration. Authors will be
notified of acceptance or rejection by July 22, 1988. A working draft
of the paper, not exceeding 15 pages, will be due by September 6, 1988
for inclusion in proceedings which will be made available at the
conference.
One conference session will be devoted to a panel discussion entitled
MEDIUM AND MESSAGE: THE FUTURE OF THE ELECTRONIC BOOK. The Centre invites
individuals who are interested in participating as panel members to submit
a brief statement (approximately 150 words) expressing their major
position on this topic. Please submit statements not later than
June 10, 1988 to the Administrative Director, Donna Lee Berg, at the above
address. Selection of panel members will be made by July 22, 1988.
The Centre is interested in specialists or generalists in both academic and
professional fields (including editors, publishers, software designers and
distributors) who have strongly held views on the information potential of
the electronic book.
PROGRAM COMMITTEE
Roy Byrd (IBM Corporation) Michael Lesk (Bell Communications Research)
Reinhard Hartmann (Univ. of Exeter) Beth Levin (Northwestern University)
Ian Lancashire (Univ. of Toronto) Richard Venezky (Univ. of Delaware)
Chairman: Gaston Gonnet (Univ. of Waterloo)
------------------------------
Date: Thu, 26 May 88 13:26 EDT
From: Emma Pease <emma@csli.stanford.edu>
Subject: From CSLI Calendar, May 26, 3:30
THIS WEEK'S CSLI COLLOQUIUM
Representation versus Interpretation
J. E. Fenstad
University of Oslo, Norway
Cordura Conference Room, 4:15, May 26
One basic assumption of the Montague approach is the compositionality
principle, i.e., the existence of a homomorphism from the "syntactic"
algebra to the "semantic" algebra. But various problematic aspects of
the "pull-back" from interpretation to linguistic forms argue for an
independent representational level. Another problematic aspect of the
Montague model is the extreme "constructionalism" of the approach,
i.e., everything is constructed by abstraction from individuals and
truth-values. In the talk I will give a survey of some recent work in
Oslo related to these problems.
------------------------------
Date: Wed, 1 Jun 88 20:36 EDT
From: Emma Pease <emma@csli.stanford.edu>
Subject: From CSLI Calendar, June 2, 3:31
ANNOUNCEMENT
This Thursday is the last day for the CSLI seminar and CSLI
Calendar. The next CSLI Calendar will appear in late September.
--------------
THIS WEEK'S CSLI SEMINAR
A Report from the Situated Inference Engine
Project on Pidgin, the Language of Interaction
Susan Stucky
(stucky.pa@xerox.com)
June 2
In this seminar, we will report the design of Pidgin--the language of
interaction for the Situated Inference Engine, which is an ongoing
project of the Situated Language Program at CSLI. Pidgin is unlike
any other designed or "formal" language its designers are aware of,
being based not only on explicit theories of syntax and semantics, but
also on discourse structure (including a nesting of "levels" of
conversation depth), dialogue structure (based on a simple but
cross-cutting notion of turn taking), simple speech acts and mood, and
a notion of subject and topic. In each case, the language is designed
to wear on its sleeve distilled versions of what current
natural-language research takes to be essential mechanisms underlying
natural human communication.
(A note to linguists: while the design of languages is familiar in
both computer science and artificial intelligence, it is less so in
linguistics. One way to think about this is to regard the design and
(soon-to-happen-we-hope) implementation of Pidgin as an example of
`experimental linguistics', in contrast to the more familiar
empirical linguistics. Though it's too soon to make any great claims,
language design is proving to be revealing about language, its
structure and embedding in the context of use and in the agent in ways
that seem different from more traditional approaches.)
------------------------------
Date: Thu, 26 May 88 19:44 EDT
From: Greg Lee <lee@uhccux.UHCC.HAWAII.EDU>
Subject: Re: Language Learnability: BBS Call for Commentators
From article <2498@mind.UUCP>, by harnad@mind.UUCP (Stevan Harnad):
" ...
" The Child's Trigger Experience: "Degree-0" Learnability
"
" David Lightfoot
" ...
" form people's mature capacities (or "grammars") will take. This BBS
" target article discusses the "trigger experience," i.e., the experience
" that actually affects a child's linguistic development. It is argued
" that this must be a subset of a child's total linguistic experience
Perhaps a commentator might point out to D. Lightfoot that this
hardly requires argument.
" and hence that much of what a child hears has no consequence for the
.... and that the "hence" introduces a non sequitur.
" form of the eventual grammar. UG filters experience and provides an
" upper bound on what constitutes the triggering experience. This filtering
" ...
" ("degree-0") complexity. Everything can be learned from simple, unembedded
""domains" (a grammatical concept involved in defining an expression's
" logical form). Children do not need access to more complex structures.
Unembedded grammatical concepts as opposed to the embedded ones, previously
thought to play a crucial role in acquisition, that is, I guess.
Nonlinguist participants in this newsgroup should take note of the
level of sophistication found in this exposition. This is how
the professionals do it.
Greg, lee@uhccux.uhcc.hawaii.edu
------------------------------
Date: Fri, 27 May 88 16:31 EDT
From: doug-merritt@cup.portal.com
Subject: Re: Language Learnability: BBS Call for Commentators
Greg Lee writes:
>Nonlinguist participants in this newsgroup should take note of the
>level of sophistication found in this exposition. This is how
>the professionals do it.
Neither the author (David Lightfoot) nor the editor (Stevan Harnad)
participate in this group, so I don't see the point...that was simply
a copy of something that will appear in a journal. Stevan isn't going
to respond to postings in sci.lang.
If you want a high level of sophistication in this newsgroup, I suggest
that you post things with such a high level, instead of complaining
about the lack thereof. I didn't appreciate that pompous letter you sent
me, nor do I appreciate the attitude in the above quote. I just posted a
reference to "Cross Cultural Universals of Affective Meaning", which
I would consider to be a "must read". Neither you nor anyone else
commented; who's going to judge which of us are the "sophisticated
linguists" and which are not?
There IS a lot of junk posted here. Just ignore it! By all means feel
free to make valuable contributions. Complaining about the lack of same
from *other* people is akin to throwing stones in a glass house.
Doug
--
Doug Merritt ucbvax!sun.com!cup.portal.com!doug-merritt
or ucbvax!eris!doug (doug@eris.berkeley.edu)
or ucbvax!unisoft!certes!doug
------------------------------
Date: Tue, 31 May 88 11:39 EDT
From: Greg Lee <lee@uhccux.uhcc.hawaii.edu>
Subject: Re: Language Learnability: BBS Call for Commentators
From article <5990@cup.portal.com>, by doug-merritt@cup.portal.com:
" Greg Lee writes:
" >Nonlinguist participants in this newsgroup should take note of the
" >level of sophistication found in this exposition. This is how
" >the professionals do it.
"
" Neither the author (David Lightfoot) nor the editor (Stevan Harnad)
" participate in this group, so I don't see the point...that was simply
" ...
" If you want a high level of sophistication in this newsgroup, I suggest
" that you post things with such a high level, instead of complaining
" about the lack thereof. I didn't appreciate that pompous letter you sent
" me, nor do I appreciate the attitude in the above quote. I just posted a
Prior to seeing this posting from Mr. Merritt, I never sent Mr. Merritt any
mail. And he has misconstrued what I wrote above. I have no complaint
about the sophistication of participants in this newsgroup. As Mr.
Merritt appears to realize, David Lightfoot is not a participant.
I was complaining about Professor Lightfoot's abstract -- I would
contrast it with the postings from participants, which I follow with
great interest.
" reference to "Cross Cultural Universals of Affective Meaning", which
" I would consider to be a "must read". Neither you nor anyone else
" commented; who's going to judge which of us are the "sophisticated
" linguists" and which are not?
Doubtless someone who is more sophisticated will comment. I'm sure
it's very interesting.
" ...
So that this will not be entirely a "meta"-posting, let me change the
subject and ask a question. What do people do when they
concatenate forms? Or: What is the nature of the mental
representation of a pronounced phrase, some pieces of which are,
or correspond to, memorized pronunciations of morphemes or words.
I ask because I'm working on phonology-syntax interaction, and
have come to suspect that phonological rule application and
syntactic concatenation are somehow the same. Yet in the ways
we usually think about them, they seem very different.
For instance, suppose memorized pronunciations are something
like a linked list of sound-segments. And when several such
lists are concatenated, we insert a temporary link at the
end of the first pointing to the second, and at the end of the
second pointing to the third, and so on. This scheme would
allow a given memorized form, say a morpheme, to occur only
once per phrase. Although it's easy to think of counter-examples,
there are lots of one-of-a-kind constraints known ...
I'd be grateful for any ideas or references.
Greg, lee@uhccux.uhcc.hawaii.edu
------------------------------
End of NL-KR Digest
*******************