LINGUIST List 11.1475

Wed Jul 5 2000

Calls: Adjacency/GLOW 2001, Cognitive Science/ICCS2001

Editor for this issue: Jody Huellmantel <jodylinguistlist.org>

Directory

1. Joao Costa, Adjacency (GLOW 2001)
2. HASIDA Koiti, Cognitive Science (ICCS2001)

24th GLOW Colloquium 2001
April 8-10, 2001
Portugal, Braga
Organization: Associacao Portuguesa de Linguistica, Universidade do Minho &
Universidade Nova de Lisboa

Call for Papers
(This year's Colloquium will have a parallel session on phonology!)

ADJACENCY

Obligatory adjacency between particular syntactic constituents is a
fact described in many constructions in many languages. Typically,
cases of adjacency between two constituents are understood as
instances of head-complement or Spec,head relations. A number of
studies in the eighties have proposed that adjacency plays a key role
in constraining certain syntactic processes: Stowell's (l981) Case
assignment under adjacency or Marantz's (l989) Morphological Merger
under adjacency. Chomsky (l995) explicitly claims that the general
framework of the Minimalist Program has no natural place for a
condition on adjacency. Yet Lasnik l999 still retains this condition
as a PF process, Halle and Marantz (l993) as well as as Bobaljik l995
assign to it a major role in morphosyntax, and current accounts of
second position effects rely on it in one form or the other (see
Anderson's l993 treatment of V2 and second position clitics and
Boskovic l995 and references cited there on second position
cliticization).

In Optimality Theory, adjacency phenomena may be described in terms of
interaction between alignment constraints and faithfulness constraints
(McCarthy and Prince (1993,1994), Prince and Smolensky (l993),
Grimshaw (1999), Anderson (l996)). Recent hypotheses on phrase
structure also presuppose a different view on adjacency. If Kayne's
(1994) hypothesis on phrase structure is correct, several phenomena
understood as instances of Spec,Head or V-complement must be
reinterpreted.

Furthermore, there are certain well-known restrictions on what kind of
elements can intervene between particular syntactic constituents. 
Adverbs or topics in particular may not often intervene between
certain syntactic constituents while other kinds of elements
(typically pronominal or adverbial clitics and negation) can. Under
some analyses, these intervention effects follow from interpretability
conditions imposed on base adjunction. This explanation, however,
relies on assumptions about adjunction that are unavailable under
current theories of phrase structure such as Kayne's antisymmetry
hypothesis. In this spirit, Rizzi (l997) proposes to capture a wide
range of adjacency and anti- adjacency effects in terms of the notion
"government", in interaction with a highly articulated layer of
functional projections.

In particular, the following questions arise: does adjacency play any
role in the grammar? Does it constrain syntactic processes or is its
role confined to PF or morphological structure? Does the interface
prosody-syntax relate with adjacency? How is PF-adjacency to be
determined in a model such as Chomsky's l998, l999, where SPELL OUT
is cyclic, at the phase level? Are strong phases (CP or vP)
"opaque" for the purposes of PF-adjacency? How is adjacency to
be explained? Is it a consequence of syntactic structure or is it a
matter of linearization? What is the analysis of V2-like phenomena, in
which there appears to be an adjacency requirement between a sentence-
initial constituent and the verb? To what extent is V2 related to
other second position effects, such as second position cliticization?
Can adjacency phenomena be taken as reliable criteria to evaluate
proposals regarding the functional structure of the clause? Is there a
way for accounting for lack of adjacency effects without a
proliferation of functional categories? Are adjacency phenomena the
effect of surface constraints formalizable within an OT-framework? How
are cases of adjacency between heads to be analyzed (pronominal or
adverbial clitics and verbs, restructuring contexts, etc) and how do
these bear on phrase structure theory?

With the advent of constraint-based frameworks of phonology, some
scholars have abandoned the rather abstract notions of locality that
were entertained in the heydays of autosegmental and metrical
phonology. The burden of the explanation now is on (the interaction
of) phonetically grounded constraints rather than on simple operations
on tridimensional representations.

Locality of spreading, for instance, is no longer determined on
independent tiers of features and class nodes, but rather on a level
of articulatory gestures, or alternatively on phonological
segments. If a feature spreads, it spreads through all intervening
material: No Gapping is one of the inviolable principles of sound
systems. Similarly, some have argued that the role of the syllable in
phonotactics can be marginalized, and most phonotactic constraints can
be formalized in terms of consonant adjacency. Feature geometries
have been abandoned by many - there does not seem to be a lot of
consensus any more about segregation of vocalic and consonantal tiers.

Quite a different trend is to attribute long distance adjacency
effects to prosodic structure. Thus, two vowels are adjacent for vowel
harmony because they are the heads of adjacent syllables.

The question thus is what now is the best representation for
phonological adjacency. Can all the evidence that has been put
forward in favor of autosegmentalism, feature geometry and syllable
structure now really be explained in another way? Is No Gapping really
an inviolable principle? Or do we still have enough evidence for more
sophisticated phonological representations? Can we really dispense
with the (long distance) effects of the Obligatory Contour Principle?
And what about higher-level orders of structure? Are 'traditional'
ways of dealing with metrical adjacency phenomena (stress clash,
stress shift and the like) still satisfactory? And what about the
specific class of opacity facts caused by phonological processes
induced by a segment which gets deleted afterwards, so that it no
longer is present in the actual output?

A final class of questions concerns the interfaces between phonology
on the one hand and morphology and syntax on the other: what is the
most successful way to describe the effect morphological and syntactic
boundaries have on the mutual visibility of string-adjacent segments?


The colloquium will consist of approximately 27 (20 in the main
session plus 7 in the phonology parallel session) talks of 45 minutes
plus discussion. Abstracts should be sent anonymously in tenfold,
accompanied by a camera-ready original with the author's name,
address, e-mail, and affiliation to:

GLOW selection committee
c/o Joao Costa
Dept. de Linguistica
Faculdade de Ciencias Sociais e Humanas
Universidade Nova de Lisboa
Av. de Berna, 26-C
1069-061 Lisbon
Portugal

Abstracts may not exceed 2 pages with at least a 1 inch margin on all
four sides and should employ a font not smaller than 12pt. Please
indicate if the abstract is submitted to the main session or to the
parallel phonology session.

Submission of abstracts by e-mail or fax will not be accepted.

Speakers will receive partial reimbursement for their expenses.

DEADLINE FOR SUBMISSION OF ABSTRACTS: December 1, 2000

For more information, contact: glow2001netvisao.pt




The GLOW Workshops
April 11, 2001
University of Oporto

The GLOW Colloquium will be followed by two workshops:

I. Language change and variation.

The interest of Generative Grammar in linguistic change comes from the
early 1980s with the pioneering work of Lightfoot 1979. In this work
the central mechanism of change was reanalysis, which included not
only radical reanalysis, where new associations of form and content
are created, but also non radical changes, where there are just
extensions of old associations.

The notion of radical reanalysis underwent a new development with
Principles and Parameters Theory. Relating acquisition and change,
Lightfoot saw these two processes as the loci of new parameter
settings by the language learner on the basis of robust evidence. An
important idea of his work of 1991 was that it is necessary to find
acquisition triggers for the resetting of parameters, which come in
principle from simple sentences.

Within this framework, morphosyntax played an important role: the
parametric differences between languages and stages of one language
were mainly restricted to those properties of syntactic structure
which were related to inflection. Work on the functional domain of
the clause influenced the idea that parametric differences in the
history of languages may be explained by differences in functional
categories (Roberts 1993, Battye & Roberts 1995).

With the Minimalist Program (Chomsky 1993) new ideas come to light. In
the last texts by Chomsky (1995, chap. 4, 1998 and 1999), the only
admitted functional categories are those with interpretable features,
the role of interfaces is reinforced. To what extent will the
weakening of functional categories modify some previous assumptions
about linguistic change?

At the same time, the difficulty of the Minimalist Program to explain
"optional" and "stylistic" movements and the restrictive
status of the theory opened a large field of research and new theories
appeared.

For Optimality Theory, particular grammars are ranking of constraints;
different rankings of the same constraints give rise to different
grammars; the optimal sentences are those best satisfy the ranked
constraints; but the fact that some constructions do not satisfy a
constraint does not mean that they are ungrammatical. Can linguistic
change be explained by the effects of different rankings of
constraints?

In the theory of grammaticalization (Hopper & Traugott 1993), change
is seen as a gradual development where many grammatical categories
come from lexical items.

Kroch 1991, influenced by the work of Lightfoot but also by the work
of Labov, tried to conciliate radical change and graduality of change:
during a period of change speakers may use two grammars: a grammar
with the old value of the parameter and a second one characterized by
the new value, with a period of competition between them, followed by
a gradual shift in the relative frequencies of certain constructions
and the emergence of a new grammar.

If parametric approaches were adequate to explain language change can
they be useful to explain dialect variation?

Within phonology, our view of language change and language variation
should have changed quite dramatically since the almost general
replacement of rule-based theory by constraint-based alternatives,
since the 'standard' analysis of these phenomena involved devices such
as rule addition, rule reordering and rule deletion. Somewhat
unfortunately, however, neither language change nor (micro-)variation
has been in the focus of attention for many phonologists in recent
years. It is therefore still largely unclear how to capture the
insights of previous theories into these phenomena in the new
notations.

One of the central ideas of Lexical Phonology, for instance, was that new
rules were typically added at the end of the postlexical grammar, from which
they would sometimes start to move upwards, until they would end up as
lexical exception marking. The question still stands whether this was a
viable description of some types of language change, and if yes, in what way
(if at all) it could be described within Optimality Theory or other
constraint-based frameworks.

Similarly, there does not seem to be any consensus yet about the best
way to describe the minimal variations between dialects of the same
language. In the days of SPE-phonology, it was sometimes supposed that
all dialects of a language have the same underlying representations
for a given word, they only differ in the way rules map those inputs
to outputs. Given Richness of the Base, this assumption seems no
longer tenable.

Another topic worth exploring is the question whether there are any
restrictions as to the way languages can vary and the way they can
change? How are we to evaluate for instance the device of universal
constraint rankings? Similarly, is language change always to be
described as a relative reranking of IO faithfulness, or is there more
to it than just that?

It seems time, then, that experts on phonological variation and
phonological change gather in order to discuss these topics.




II. Syntax-Semantics Interface: nominal and temporal
anaphora

What speakers know about their language is part of the description of
that language and this is one of the objectives of linguistic
theories. Some of the properties of natural languages that interest
semantics are also studied by syntax. The problem is how to articulate
both domains. That is, which kind of semantics, lexical and/or
compositional, should be considered? At which level of representation
may that articulation occur and in which way, if we take into account
the minimalist program (Chomsky 1995), and the semantic (referential)
theories, namely the dynamic theories of meaning or the static ones
(Kamp 1981, Heim 1982, Chierchia 1995, among others)?

Some of the topics that are more extensively studied in semantics are
anaphora, quantification and scope. From a semantic point of view the
anaphoric relations are not so constrained as they are in syntax. So
we can say that they cover not only anaphora and coreferenciality, but
also binding. But we may also consider the cases of anaphora that
occur outside the scope of their binder, this is why there are the
e-type approach for anaphoric pronouns, unselective binding and
dynamic existential quantifiers for indefinite NPs.

In the past years extensive work has been done on indefinites, and,
depending on the approach, they are considered as free variables
subject to existential closure in their immediate environment, or as
existential quantifiers subject to existential disclose in some
environment, particularly adverbs of quantification.

Since Partee (1973, 1984) parallels between nominal and temporal
anaphora have been drawn, namely, non-linguistic 'antecedents',
definite anaphors with definite antecedents, indefinite antecedents
and also bound variables and 'donkey sentences'.

'Temporal anaphora' (Hinrichs 1986) is more subtle than nominal
anaphora because of the variety of categories involved (tenses,
adverbials, adverbial clauses, main clauses, aspect). Nevertheless, it
is well known that in complex sentences and discourse the
interpretation of a particular tense may be dependent on another one
(whether we take interval or event semantics). This is good reason to
use the reichenbachian reference point (or temporal perspective
point). Apart from that, we know that some tenses, in order to be
interpreted, need some kind of reference point, as it is the case of
the past perfect in English (and other languages) and the
'Imparfait' in Romance. Besides this, problems of sequence of
tenses and in particular double accessibility readings (Abush
1991,1997, Ogihara 1997, Stowell 1993) can be viewed as depending on
semantic factors as well as on special properties of verbal forms or
on properties of clause structure ( Gueron and Hoekstra 1988, Giorgi
and Pianesi 1997).

The aim of this workshop is to get together semantic and syntactic
approaches in order to enrich our knowledge of these topics.

All workshops will have 45 min. presentations plus 15 min. discussion.
Abstracts should consist of five two-page anonymous copies accompanied
by a camera-ready original with author's name, address, e-mail, and
affiliation and should be sent to:

GLOW Workshop {Language Change/Semantics}
c/o Joao Costa
Dept. de Linguistica
Faculdade de Ciencias Sociais e Humanas
Universidade Nova de Lisboa
Av. de Berna, 26-C
1069-061 Lisbon
Portugal

Speakers will not be reimbursed for their expenses.

DEADLINE FOR SUBMISSION OF ABSTRACTS: December 1, 2000

For more information, contact: glow2001netvisao.pt


The Colloquium venue

The Colloquium will be held at the University of Minho, in Braga. In
the next newsletter, there will be information concerning how to get
from the International Airport of Oporto to Braga. The workshops will
be held at the University of Oporto

Accomodation in Braga

We urge all prospective participants to reserve rooms in the hotels as
soon as possible, since April is a peak tourist month. Please consult
GLOW newsletter for a list of hotels.

CALL FOR PAPERS (the first announcement)

The Third International Conference on Cognitive Science (ICCS2001)
August 27-31, 2001, Beijing, China

The Third International Conference on Cognitive Science (ICCS),
following the two very successful ones held in Korea and Japan in 1997
and 1999 respectively, will be held in Beijing, China on August 27-30,
2001. The purpose of the Conference is to bring together researchers
in various active areas in cognitive science for exchanging recent
progresses made in various cognitive science research groups from
Asia, North America, Europe, Australia, and other places.

CALL FOR SYMPOSIUM TOPICS AND ORGANIZERS

In order to balance general and special interests of participants, the
conference will arrange several plenary talks by well-known cognitive
scientists as well as organize several symposiums for cutting-edge
research topics. Suggestions about topics and organizers of the
symposiums are particularly welcome, and should be sent to Huisheng
Chi <chipku.edu.cn> before February 28, 2001.

GUIDELINES FOR ABSTRACT/PAPER SUBMISSIONS 

The conference invites oral or poster presentations in broadly defined
studies of cognition. Pieces of work are welcome from, but not limited
to, cognitive psychology, artificial intelligence, cognitive
neuroscience, linguistics, philosophy, cognitive anthropology, social
cognition, and comparative cognition. Oral and poster presentations
all will be given in English, with English full papers published in
the proceedings. For oral presentations, please submit a full
paper. The paper must be 5 pages maximum, with fonts at least 12
pt. For posters and work in progress, please submit a one-page
abstract. Electronic submission is highly encouraged to the address
iccs2001orgetang.com in MS Word files, and pdf files. You may also
submit your work in the paper format (4 copies) to:

 Beijing Laboratory of Cognitive Science,
 University of Science and Technology of China
 Graduate School, Academia Sinica
 P. O. Box 3908, Beijing,
 100039 Beijing, P. R. China.
 
DEADLINES 
Full papers submission: March 30, 2001
One-page abstracts submission: April 30, 2001: 
Notice of acceptance: May 30, 2001

Sponsors

Ministry of Science and Technology of China;
Chinese Academy of Sciences;
Ministry of Education of China;
National Nature Science Foundation of China.
 and
Beijing Laboratory of Cognitive Science, Univ. of Science & Technology 
 of China;
Center for Brain and Mind Research, Chinese Academy of Sciences;
Institute of Artificial Intelligence and Robotics,
 Xian Jiaotong University;
Microsoft Research China;
State Key Laboratory of Pattern Recognition, 
 Institution of Automation, Chinese Academy of Sciences;
State Key Laboratory on Machine Perception, Peking University; 
State Key Laboratory of Intelligent Technology and Systems, 
 Tsinghua University;

Conference chair:
 Lin Chen (Beijing Laboratory of Cognitive Science,
 Univ. of Science & Technology of China)

Program committee:
Chair: Huisheng Chi (State Key Laboratory on Machine Perception,
 Peking University, China).
Co-Chairs: Chungmin Lee (Seoul Nat'l University, Korea)
 Yasuhiro Katagiri (ATR, Japan)

Organizing committee:
Chair: Nanning Zheng (Institute of Artificial Intelligence and Robotics,
 Xian Jiaotong University, China).
Co-Chairs: Koiti Hasida (Electrotechnical Laboratory, Japan)
 Charles Ling (University of Waterloo, Canada)

The web site http://www.ICCS2001.com has been set up and will be
updated constantly for new information on plenary talks, symposiums,
paper submission, registration, local arrangement, and so
on. Inquiries regarding the further information about the conference
should be addressed to The organizing committee at
iccs2001orgetang.com We invite you to visit our web site at
http://www.ICCS2001.com for up-date information on the conference.