Publishing Partner: Cambridge University Press CUP Extra Publisher Login
amazon logo
More Info

New from Oxford University Press!


Style, Mediation, and Change

Edited by Janus Mortensen, Nikolas Coupland, and Jacob Thogersen

Style, Mediation, and Change "Offers a coherent view of style as a unifying concept for the sociolinguistics of talking media."

New from Cambridge University Press!


Intonation and Prosodic Structure

By Caroline Féry

Intonation and Prosodic Structure "provides a state-of-the-art survey of intonation and prosodic structure."

Review of  UG and External Systems

Reviewer: Luis Vicente
Book Title: UG and External Systems
Book Author: Anna Maria Di Sciullo
Publisher: John Benjamins
Linguistic Field(s): Computational Linguistics
Discourse Analysis
Linguistic Theories
Cognitive Science
Language Acquisition
Subject Language(s): American Sign Language
Language Family(ies): Romance
Issue Number: 16.3500

Discuss this Review
Help on Posting
Date: Fri, 28 Oct 2005 13:17:49 +0200
From: Luis Vicente
Subject: UG and External Systems

EDITOR: di Sciullo, Anna Maria
TITLE: UG and External Systems
SUBTITLE: Language, Brain, and Computation
PUBLISHER: John Benjamins
YEAR: 2005

Luis Vicente, LUCL, Leiden University

[We apologize to the reviewer, the editors of this volume and the
publisher for our delay in posting this review. -- Eds.]

The stated goal ''UG and external systems'' is to advance our
understanding of the nature of interfaces, i.e., how the requirements
imposed by conceptual-intentional and sensory-motor systems have
influenced the way UG is. To this end, di Sciullo has put together an
eclectic collection of articles, which deal with a wide range of topics
and subdisciplines of linguistics. There is, nonetheless, a common
thread through them, as just mentioned: the exploration of the
interfaces and their effect on the shape of human language. What is
not so parallel amongst the articles, though, are the depth of the
studies and the success of the hypotheses.


The book consists of three parts (''Brain'', ''Language'',
and ''Computation''), comprising a total of 18 articles. In what follows,
we'll go briefly through them.


In ''Depictives: syntactic and semantic asymmetries'', Daniela Isac
examines the differences between subject and object orientation of
depictive secondary predicates. She claims that their asymmetries
stem not only from their syntactic differences, but also from their
different semantics. More specifically, she argues that object oriented
depictives compose directly with the object they modify, whereas
subject oriented depictives compose with an entire vP. This analysis
recalls certain recent proposals about the different structures of
ditransitive predicates (e.g., Pylkkänen 2002), although Isac doesn't
mention this parallelism. Although a comparison between Isac's
proposal and Pylkkänen's would certainly have been interesting, the
article is still highly readable and puts forward a consistent and neat

The second chapter is Stanca Somesfalean's ''On two issues related
to clitic clusters in Romance languages''. Here, Somesfalean argues
against templatic analyses of clitic clusters, which had been proposed
in the past largely due to some intra- and crosslinguistic idiosyncrasies
that apparently couldn't be reduced to independent principles.
However, in this rather data-heavy article, Somesfalean argues that
the properties of clitic clusters are actually a reflection of the order of
arguments in the clause.

In the third chapter (''On the question of (non)agreement in Russian
imperatives''), Edit Jakab examines constructions in which an
imperative verb form doesn't have an imperative meaning, but rather
either a conditional or a contrastive reading. These two constructions
are further characterized by the lack of agreement between the verb
and the subject. His proposal is that the two anomalous imperatives
lack the higher part of the functional layer, namely, TP and AgrSP,
which accounts for the lack of agreement morphology.

Nicola Munaro's chapter (''Computational puzzles of conditional clause
preposing'') starts by establishing the correlation that, in Italian
dialects, whenever a conditional clause exhibits subject-verb
inversion, it tends to precede the main clause. The analysis consists
on positing that subject-verb inversion involves verb movement to the
CP area, which activates the TopicP layer and defines the conditional
clause as topical. As a consequence, the entire subordinate clause
needs to move to a topic position in the main clause. Apart from this
particular analysis, Munaro's paper is of interest in that it includes an
exploration of the left periphery of Italian dialects couched in the
cartographic approach.

In chapter five (''Clefts and tense asymmetries''), Manuela Ambar
argues against the hypothesis that cleft sentences have a relative-like
structure. On the basis mainly of Portuguese data, she recognises
various types of clefts and argues that one determining factor
distinguishing them is whether the tenses of the copula and the lexical
verb have to be identical or not. She takes this requirement to be a
point of parametric variation, which accounts for the crosslinguistic
distribution of infl-less clefts and that-clefts.

This first part closes with Evan Mellander's paper ''Generating
configurational asymmetries in prosodic phonology'', where he
proposes an OT analysis of various types of crosslinguistic variation in
the structure of feet and syllables. The crucial claim of the paper is
that these asymmetries are reducible to a small set of rhythmic well-
formedness constraints.


In their paper ''Language learnability and the forms of recursion'',
Willian Snyder and Tom Roeper argue that recursion is at the heart of
linguistic competence. They support this claim through a case study
on endocentric root compounding in various languages (French,
English, and Swedish), where they claim that various types of
recursion are possible, and the language learner's task is to
determine which ones are available in their language. A number of
crosslinguistic differences are derived from this hypothesis.

In ''The autonomous contribution of syntax and pragmatics to the
acquisition of the Hebrew definite article'', Sharon Armon-Loten and
Idit Avram argue that pragmatics and world knowledge are important
factors in the acquisition of the distribution of definite articles in
Hebrew and English. On the basis of these data, they claim the
pragmatic concept of shared knowledge doesn't arise until age four or
so, accounting for some earlier-age errors in the use of articles.

Chapter nine is Helen Goodluck's ''D-linking and question formation:
comprehension effects in children and aphasics''. She shows that
there is a subject/object asymmetry in the comprehension of d-linked
questions by children and aphasic, in that object questions are harder
to parse than subject question. However, this contrast is absent in
non-d-linked questions. She suggests that the asymmetry in d-linked
questions can be traced back, at least partially, to perceptual factors
in the set-up of the experiment.

In chapter ten (''Evidence from ASL and ÖGS for asymmetries in UG''),
Ronnie Wilbur shows that both American and Austrian Sign
Languages provide evidence for the existence of structural
asymmetries in UG. He considers the behaviour of the verb with
respect to various elements like negation or stage/individual level
predicates; the clause structure of ASL and ÖSG, in particular the
directionality of various projections; and the scope of wh- elements, as
indicated by non-manual gestures like brow furrowing. The take-home
message here is that languages, whether spoken or signed, are
inherently asymmetric.

Ning Pan and William Snyder (''Acquisition of phonological empty
categories: a case study of early child Dutch'') examine the order in
which Dutch children acquire different syllable types. On the basis of a
reexamination of available CHILDES data, they conclude that CV is
the first syllable type to be acquired, followed by a group formed by
CVC, V, and VC (contra Levelt, Schiller & Levelt 2000, who claimed
that the latter three were acquired at different stages). Pang and
Snyder account for this pattern in terms of Government Phonology, by
postulating two parameters [+/- empty onset] and [+/- empty nucleus].
This analysis thus lends support to the onset-nucleus theory of the
syllable, as opposed to the onset-nucleus-coda one.

The ''Brain'' section finishes with Matt Bauer's article ''Prosodic clues
during online processing on speech''. In this paper, Bauer tries to
determine whether hearers make use of prosody as an aid to parse
the syntactic structure of the incoming string. The data come from two
case studies on stress shift in American English. However, the results
are negative, and the conclusion is that prosody doesn't seem to
provide a cue as to what the syntactic structure of the clause in
question is, although, Bauer adds a caveat that this result might be
related to a flaw in the design of the experiment.


Chapter thirteen is Annamaria di Sciullo & Sandiway
Fong's ''Morphosyntactic parsing'', in which they develop a parser
based on di Sciullo's (1996) own theory of morphological selection.
The main idea is that lexical properties of derivational affixes are not
lexically encoded, but are rather defined in terms of asymmetric
structural relations, from which premise they derive various properties
of affixation.

Next is Sourabh Niyogi & Robert Berwick's paper ''A minimalist
implementation of Hale & Keyser incorporation theory''. On the basis
of Hale & Keyser's well-known work on incorporation, they develop a
parser that covers most of Levin's (1995) English verb classes. In the
same way as Hale & Keyser's theory, Niyogi & Berwick's analysis has
the added advantage of replacing theta roles (qua specifications in
the lexical entries of verbs) with specific structural configurations.

Following up with the morphology theme, Henk Harkema (''Minimalist
languages and the correct prefix property'') develops a top-down
parser with the ''correct prefix property'', namely, one in which, given
an ungrammatical sentence, parsing is stopped on the first word that
doesn't fit the structure. The entire system is based on the basic
operations Merge and Move.

In ''Computation with probes and goals'', Sandiway Fong presents a
parser that implements the operations Merge, Move, and Agree.
These operations are redefined so that they can be applied in a top-
down parser. Although the paper gives a thorough overview of how
these operations are to be implemented, it fails to address how they
compare to their counterparts in bottom-up structure building.

Rodolfo Delmonte (''Deep and shallow linguistically based parsing:
parametrising ambiguity in a hybrid parser'') develops a theory of
processing that reconciles symbolic and statistical approaches. His
idea is that parsing should be based on grammar, not on statistical
methods. However, statistical considerations can apply whenever
lexical information introduces an ambiguity whose resolution requires
non-grammatical knowledge.

The final article of the book is Philippe Blanche's ''Towards a
quantitative theory of variability''. He proposes an ''extended interface''
of sorts, in which information coming from various linguistic
components (prosody, syntax, semantics…) adds up to the
communicative act. This enables him to account for variability in a
number of constructions, the choice between alternatives depending
on whether a certain ''equilibrium threshold'' is reached or not.


As I mentioned at the beginning of this review, this book is a rather
eclectic collection of papers, both in contents –- as can easily be seen
from the above paragraphs -- and in quality. While some of the papers
(e.g., Pan & Snyder's comes to mind) are important contributions to
the understanding of the topic they deal with, others are not so much.

I must also say that I found the subtitle of the book (''Language, brain,
and computation'') a bit misleading. I was expecting a collection of
interdisciplinary articles, in which language, brain, and computation
issues were discussed in relation to each other. However, what I
found is that these three categories are kept rather apart. That is, one
finds articles about linguistic theory, about ''brain'', and about
computation, but the three topics do not intermingle in individual
articles. The ''brain'' label is also a bit obscure for me, since it makes it
sound as though you were in for a set of papers on
neuro/psycholinguistics. Instead, most papers in this section deal with
acquisition, and only a couple of them with comprehension.

Now, all this doesn't mean this is a bad book. I enjoyed reading most
of the articles, including the ones dealing with topics I am not familiar
with. In this sense, it is a valuable and interesting book. My only
complaint is that it didn't deliver what it advertised, not that the content
is not worth the effort of reading through it.

I am a fourth year PhD student, specialising in formal syntax. In the
past I have worked on relativisation, reconstruction, head movement,
remnant movement, the syntax-phonology interface, coordination, the
structure of VP, and argument licensing. Currently I am writing a
dissertation on A-movement and agreement