Motapanyane, Virginia, ed. (2000) Comparative Studies in Romanian
Syntax. North-Holland (imprint of Elsevier Science), hardback
ISBN 0-08-043871-7, vii+341 pages, $97.00, North-Holland
Linguistic Series 58.
Manideepa Patnaik, Harvard University
This book provides an up-to-date overview of the studies in
Romanian Syntax by bringing together linguists working on
Romanian within generative grammar. The volume's comparative
approach demonstrates the relevance of Romanian data to
grammatical theory. The editor's introductory chapter provides a
valuable summary of developments in Romanian syntax and is the
ideal preparation for the studies contained in this volume, both
for Romance specialists and for those less familiar with the
topic. The target audience for this book may be graduate students
in general linguistics and Romance linguists in particular, and
scholars of all theoretical frameworks who are interested in
Romance or Balkan languages. Syntactically, Romanian is a hybrid
between Romance and Balkan languages, and many peculiarities can
be understood only with reference to equivalent paradigms in
Romance and Balkan. However, majority of the works assume
Romanian as a Romance language. The present volume contributes to
redressing this comparative balance by proposing cross-linguistic
accounts that include Balkan languages (G. Legendre), English (V.
Motapanyane), German (J. Bayer & A. Grosu), French (Y. D'Hulst et
al.), and Hebrew (C. Dobrovie-Sorin).
The introductory chapter by Alboiu & Motapanyane "The generative
approach to Romanian grammar: an overview" provides an outline of
Romanian syntax from the perspective of generative grammar.
Interesting topics including the morphology, internal structure,
case and word order of nominal elements; pronouns and anaphora;
clitic doubling; object raising; Wh and quantification;
auxiliaries and modals; declarative and indicative verbs;
interrogatives; imperatives; sentential complements such as
embedded indicative CP, subjunctive clauses, infinitives; supines
and null operators. These topics have been introduced through
examples and inflectional paradigms, followed by presentations of
controversial analyses they prompted in generative studies.
Bayer & Grosu "Feature Checking meets the Criterion Approach:
Three ways of saying only in Romance and Germanic" approach
operator constructions in general by postulating an operator
feature whose "spreading" across the boundaries of the left-
branch islands is the licensing factor for pied-piping and VP-
scope effects. Data from three constructions i.e. adnominal
adjectives, only-clauses and constructions with left DP-external
adjectives in Romance, especially Romanian and German, provide
support for the existence of the operator feature spread and
indicate the structural environment in which it may or may not
apply. The three syntactic constructions they have addressed have
comparable semantic import, but they have distinct distributional
properties and distinct cross-linguistic privileges of
occurrence. The constructs with synonymous Italian, English and
Romanian examples are given below:
(1) La sola maria si e presentata. (Italian)
the only Maria REFL is presented
(2) Only Mary showed up. (English)
(3) Singura maria/Maria singura s-a prezentat. (Romanian)
only Mary /Mary only REFL-has presented
The first two constructions were discussed in Bayer (1996) at
some length and have been partly reanalyzed here; construction
(3) seems to have neither described nor analyzed in the earlier
generative literature. The authors have proposed to analyze the
intra-linguistic and cross-linguistic distributional properties
in terms of differences in the feature make-up of lexical cross-
linguistic distributional properties in terms of differences in
the featural make-up of lexical items with operator import that
individual languages may or may not have. In particular, they
proposed that in constructions type (1) the Op's features might
percolate up the phrase markers as a consequence of purely formal
and non-semantic checking operations. Constructions of type (2)
have only Op features, and that these are forced to undergo
movement to a scope position unless such movement is blocked by
known constraints on Move-alpha. Type (3) constructions reveal an
asymmetry well-formed structure can only be achieved in those
cases where the relevant phrase semantically well-formed
structure can only be achieved in those cases where the relevant
phrase semantically well-formed structure can only be achieved in
those cases where the relevant phrase is really in potential
scope position for reasons that have nothing to do with semantics
proper. In analyzing these various constructions, they pointed to
the need recognize to two types of covert movement operations,
one purely formal and one semantics-oriented. The present results
support conclusions about two types of covert movement that were
reached in Bayer (1996, 1998, 1999), but they clearly go beyond
that in presenting a more fine-grained picture of the interaction
between the syntax of functional morphology and semantic
interpretation. In doing so they have shown that important
insights from both the Criterion approach by Rizzi (1991, 1997)
and the feature-checking approach of Chomsky (1995) can be
successfully integrated into a more unified account.
Cornilescu "The double subject construction in Romanian"
discusses the pre-verbal subject position and concludes that it
is not L-related. On the basis of morphological and syntactic
facts, she argues that Romanian clauses are Mps. The functional
structure of the MP includes at least the following ordered
categories:
MP>CIP>AgrSP>TP>AspP....VP
There are two argumental subject positions in the Romanian
clause, both of them post-verbal:
A) The thematic position which is Spec VP;
B) The case and pro-licensing position which is SpecAgrSP
Case is uniformly checked in SpecAgrSP, except for the DSC, where
SpecTP is also involved. Depending on the intrinsic semantic
properties of the DP, it will check Case overtly. The pre-verbal
subject is in the non-argument position of left dislocated Topic
or Focus, depending on the particular [+Topic] or [+Focus]
features it must check. The position that we have reached is
similar tot hat expressed in Alexiadou (1994) on Modern Greek.
She also claims that in SVO the subject cannot be in SpecAgrSP
and that to license a pre-verbal subject an additional TopicP and
Focus P is needed. The Agr verbal features do not include the
Topic/Focus feature, so such features are checked in special
operator projections. Thus, although they have used different
arguments, they have confirmed the analysis in Dobrovie-Sorin
(1994).
D'Hulst, Coene & Tasmowski "Last resort strategies in DP: article
reduplication in Romanian and French" resumes the discussion of
genitive marking from a Minimalist (Chomsky 1995) perspective.
The authors rely on the concept of overt [case] feature checking,
implemented in a local Spec-head configuration. The two instances
of article reduplication they have considered in this article
both appear to exploit the same last resort strategy of template
DPs, albeit for different reasons. In this case they discuss more
extensively (Romanian possessor phrases introduced by al ), the
template DP offers the Poss both an escape hatch to check its
strong nominal feature and a way for the possessor to check its
genitive Case in overt syntax. French exploits this very same
strategy for interpretative reasons: the possibility of a
restrictive clause of superlative quantification asks for an
explanation; more specifically, the comparative particle needs to
have the noun in its scope so that restrictive clause can be
construed at the conceptual-interpretative interface. Although
the use of template DPs appears to be determined by language
specific rules, the Template DP hypothesis is based on minimal
assumptions. The assumption that empty Ns can be licensed, is
independently needed for other cases. Since Romanian and French
template DPs occur within other Dps or, as in the case of the
predicative use of genitives and superlatives, are locally
related to her DPs, the licensing of an empty N in the template
DPs should receive an account similar to the one in elliptical
constructions (the red and blue car), which, by the way,
constitute the most widespread case of empty N licensing. The
other basic assumption of the Template DP hypothesis is that the
definite determiner heading these DPs is expletive. This
assumption places an upper bound to the cross-linguistic
distribution of template DPs: only languages that independently
allow the definite determiner to function as an expletive can
exploit template DPs as a last resort. Both Romanian and French
appear to be such languages. This analysis offers a uniform
treatment for genitive case checking in adjacent and non-adjacent
Possessive-Possessor strings. Authors opine that further research
is needed to check whether the above assumptions are all that is
needed to license template DPs and whether these assumptions
account for the relatively restricted cross-linguistic
distribution of template DPs. According to authors, an
interesting test case for this is continental Scandinavian
article reduplication in adjective plus noun constructions.
Dobrovie-Sorin "(In)definiteness spread: From Romanian genitives
to Hebrew construct state nominals" argues that Romanian
morphologically marked genitives must be analyzed as arguments of
the function denoted by the main N. This paper basically aims to
establish the existence of two universal constraints imposed by
the SpecDP position:
(a) SpecDP can be lexically filled only if D is empty or
filled with a definite article.
(b) SpecDP cannot be filled with bare NPs.
These constraints are derived as consequences of the semantic
composition that underlies DPs that contain a SpecDP constituent:
(c) The head N denotes a function of type (e,e), which
applies to the individual denoted by the DP in SpecDP and yields
the individual denoted by the overall possessive DP.
The paper goes on to show that "indefiniteness spread" phenomenon
found in Saxon genitives and Hebrew Construct State nominals
(CSNs) is due to the universal semantic rule in (c), rather that
to some marked, construct-specific mechanism of Spec-head
agreement in +-def features between the genitives DP and the D of
the head N. Relying on the minimalism assumptions of Bare Phrase
Structure, she argues against the X bar theoretical formal
universal according to which all nominal projections are
projections of D. The functional analysis of the main N is
incompatible with fully semantic determiners on the main N: only
the definite article that can be an expletive is allowed. Thus,
the necessary co occurrence of morphological genitives and
definite articles in Romanian, which has so far been attributed
to a language-specific mechanism of Case assignment, is now a
result of the functional interpretation of the main noun, a
phenomenon that is expected to appear in other languages too.
Legendre "Optimal Romanian clitics: a cross-linguistic
perspective" reassesses the Romanian clitic system in the
framework of Optimality theory. Using a wide comparative, the
author argues for a morphological analysis of clitics in general.
The main departure from the earlier work lies in the claim that
alignment constraints are violable, a claim independently made in
Anderson (1996). Different alignment constraints regulating the
realization of distinct features compete for the left edge of the
clause; hence they must be violable. Which one prevails depends
on two factors: the input to optimization that contains the
features themselves and thus determines which constraints are
applicable in particular context, as well as the ranking of the
alignment constraints themselves. The latter constitutes a
partial grammar of a given language. Partial rankings for
Romanian, Macedonian, Bulgarian, and Italian are proposed.
However, the major claim is that clitics, as lexical affixes,
instantiate functional features and become subject to alignment
constraints. This analysis covers a larger domain of empirical
data, accounting not only for the cross-linguistically variable
cluster-internal order but also for variable position in the
clause.
Motapanyane "Parameters for focus in English and Romanian"
reinterprets the analyses of earlier studies in a minimalist
framework. This paper argues that [Focus] features surface in the
grammar in conjunction with semantically related formal features,
in particular, [wh] and [tense]. Cross-linguistic variation
arises from the parametric choice of the fundamental host for
[focus], that is, C with [wh], or T with [tense]. This hypothesis
is tested on English and Romanian, with systematic contrasts in
pre-verbal Focus constructions that can be reduced to parameter
settings, opposing the [Focus/wh] option in English to the
[Focus/tense] option in Romanian. Subsequent application of the
parameterized pre-verbal Focus analysis leads to a uniform
account for cross-linguistic variation in a variety of structures
involving wh-movement. Pre-verbal Focus constructions in English
and Romanian rely on similar syntactic structures with multiple
Specs, as in a structure 'FOC be phrase who focus/who' and '
subject focus V+T', needed for checking on heads with two sets of
features. Variation occurs in the distribution of the two-feature
set, specified on C in English and on T in Romanian. This,
according to the author, follows from language internal
properties, in particular properties of scope positions and of
A'-chains in the two languages. The author points out that
[Focus}, as a non-categorical feature, cannot be recognized by
the computational system unless it combines with semantically
related formal features, in particular, [wh] and [Tense]. This
hypothesis leads to a parameterization of Focus: (i) [Focus/wh]
as in English; (ii) [Focus/Tense] as in Romanian. Since the
syntactic manifestation of [Focus] depends on [Tense] in
Romanian, [Focus/Tense} merges in T and leads to the projection
of two SpecTP positions: one to check the [D] / [EPP] features of
T, and one to check the [Focus] features of T. This approach is
in line with the theoretical assumptions of the Minimalist
Program that restrict syntactic relevance to formal features.
Pirvulescu & Roberge "The syntax and morphology of Romanian
imperatives" offer a detailed analysis of the verbal morphology
in Romanian imperatives and consider its relevance for the syntax
of imperative constructions. The inflectional paradigms of
imperative verbs are construed as resulting from the assignment
of default forms where the default forms are determined through
feature geometry; among the competing affixes, only the ones that
are not specified for Tense or those that are the least marked
enter into the formation of imperative verb forms. The authors
notice that the alternation -e/-i in the ending of a certain
class of imperatives correlates with the transitivity value of
the verb. The contrast between the transitive and the
intransitive values of the same verb is related to the
composition of the functional domain in imperative clauses: in
the absence of Tense, imperative constructions force the
projection of a position for the internal argument, which is
reflected in the morphological alternation.
Although the introduction has a list of a list of references,
more about each topic can be found by consulting the Appendix to
the volume, which contains a general bibliography of the
generative studies on Romanian syntax from 1980 to 2000.
References:
Alexiadou, A. (1994) Issues in the syntax of adverbs. Doctoral
dissertation, Postdam University
Anderson, S. (1996) How to put your clitics in their place or why
the best account of second-position phenomena may be something
like the optimal one. The Linguistic Review 13,165-191
Bayer, J. (1996) Directionality and Logical Form. On the scope
of Focusing particles and Wh-in-situ. Kluwer, Dordrecht
Bayer, J. (1998) Tow types of covert movement. Paper presented
at the Workshop on acquisition and variation in syntax and
semantics, Trieste, SISSA
Bayer, J. (1999) Bound Focus or: How can association with Focus
be achieved without going semantically astray. In: The Grammar of
Focus (G. Rebuschi and L. Tuller, eds.) John Benjamins, Amsterdam
Chomsky, N. (1995) The Minimalist Program. MIT Press, Cambridge,
Mass
Dobrovie-Sorin, C. (1994) The syntax of Romanian. Comparative
studies in Romance. Mouton de Gruyter, Berlin
Dobrovie-Sorin, C. (1994) On impersonal se/si in Romanian, French
and Italian. In: Paths Towards Universal Grammar: Studies in Honor
of Richard S. Kayne (G. Cinque, J. Koster, J-Y. Pollock, L. Rizzi
and R. Zanuttini, eds.), pp.137-153. Georgetown University Press,
Washington
Rizzi, L. (1991) Residual Verb Second and the Wh-Criterion.
Technical Reports in Formal and Computational Linguistics
2, University of Geneva.
Rizzi, L. (1997) The fine structure of the left periphery. In:
Elements of grammar (L. Haegemann, ed.), pp.281-337. Kluwer
Academic Publishers, Dordrecht
About the reviewer:
Dr. Manideepa Patnaik did her Ph.D. on Aspects of
Junag Syntax within the Minimalist framework. She is the
co-author of an Oriya grammar book. She is working on the
Complementizer System of South Asian languages at present.
|
