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  Romance Languages and Linguistic Theory 2010

Reviewer: Anna Alexandrova
Book Title: Romance Languages and Linguistic Theory 2010
Book Author: Irene Franco Sara Lusini Andrés Saab
Publisher: John Benjamins
Linguistic Field(s): General Linguistics
Language Family(ies): Romance
Issue Number: 24.2851

Discuss this Review
Help on Posting

The present volume contains a selection of ten papers presented at the twenty-fourth edition of the annual conference series ‘Going Romance’, which was held at Leiden University. It represents a variety of fields, including syntax, semantics, morphology, phonetics, and phonology. As the conference was preceded by a workshop on morphosyntax-phonology interface theories, research into interfaces appears to be one of the central issues of the volume (see the papers by D. Embrick, F. Torres-Tamarit & C. Pons-Moll, V. Déprez, K. Syrett & S. Kawahara). The papers range in topics that cover, among others, argument structure, differential object marking, stem alternations, nominalizations, event structure, word stress, the interaction of information structure with syntax and prosody, and ellipsis. The diversity of Romance languages is represented in the collection mainly by Catalan, French, Italian, Portuguese, Romanian, and Spanish; it is also worth noticing that two of the articles are dedicated to Capeverdean, a Portuguese-based Creole language. Apart from the papers, the volume also consists of a short foreword by the editors (p. VII) and a subject index (pp. 221–223).

Each of the papers is summarized briefly here, in the order that it appears in the volume.

“From Romance clitics to case: Split accusativity and the Person Case Constraint”, by Maria Rita Manzini, provides a new account of interactions between dative and 1st/2nd person in Romance within the minimalist framework. The author proposes a syntactic account which links both split accusativity phenomena and person-case interactions in the Person-Case constraint, which becomes possible as the notion of a dative is reconstructed in terms of a Q(⊆) category, denoting an ‘inclusion’ relation. The analysis focuses on Italian, though Romanian, Arbëresh (an Albanian variety spoken in Italy) and Greek data are also considered. As is well-known, many Romance varieties present the same clitic form for the accusative and the oblique in the 1st/2nd person, while in the 3rd person, accusative and dative are distinguished. A widely accepted approach to this type of asymmetry is to postulate a single underlying case system for all the persons, while the surface syncretism is considered to be of morphological nature. Manzini’s objection is that within this type of account, morphology’s only function is “opacizing the syntax” (p. 6). According to Manzini, Italian 1st/2nd object clitics should be analyzed in terms of split accusativity because they show overt dative morphology, as they have the same -i inflection as the 3rd person dative ‘gli’, and thus, are embedded as datives. Moreover, the paper suggests a reworked version of the Person Case Constraint as a constraint on the interpretation of Q(⊆), such that 1st/2nd person objects must be interpreted as the argument of Q(⊆), if present, so the dative clitic is left without an interpretation.

“Contextual conditions on stem alternations: Illustrations from the Spanish conjugation”, by David Embick, examines the phenomenon of stem alternation, or stem allomorphy. In Spanish, there are numerous verb-specific alternations in stem-vowels, e.g., the 1SG form of the verb ‘pensar’ (‘think’) is ‘pienso’, and the 1PL form is ‘pensamos’. Such alternations are morphological, as diphthongization applies only to certain root morphemes with /o/ and /e/ stem vowels, whereas other verbs do not undergo this process, and this distinction cannot be predicted synchronically. At the same time, the received view is that the alternation is conditioned by stress, so there is nothing morphological about it. The interpretation of such alternations is controversial because in principle, in such cases, one can postulate a single underlying form (Morphophonological Theory) or different irregular alternants that are listed in the memory as separate vocabulary items (Stem Storage Theory). The author’s claim is that this controversy cannot be solved without a general theory of the locality conditions under which stem alternations take place. On the basis of several case studies, he arrives at the conclusion that morphological or lexical conditioning does not imply storage of alternants; at least some morpheme-specific alternations must be treated as part of the morphonology. The alternations under analysis cannot be treated with stored stems because they don’t occur under the locality conditions that apply to contextual allomorphy, so they must be treated (morpho)phonologically.

In “State nouns are Kimian states”, by Antonio Fábregas and Rafael Marín, the recently established distinction between the so-called Kimian (e.g. be tired, know, resemble, etc.) and Davidsonian (e.g. sit, stand, sleep, etc.) states is extended to state denoting nouns (though they don’t consider this classification exhaustive). The paper is aimed at filling a gap in the existing literature on deverbal and verb-related nouns which is mainly focused on nominal constructions denoting events, while those denoting states have received little attention. The analysis is based on European Spanish data. The relevant features of Spanish state nouns are the rejection of the plural, the incompatibility with the predicate ‘tener lugar’ (‘take place’) and manner denoting adjectives, and the unavailability of a temporal reading with adjectives ambiguous between a time and a degree interpretation. The authors argue that all state nouns behave like Kimian states, even when they are derived from a verb which is a Davidsonian state. However, only the Davidsonian-state-denoting verbs that contain a gradable property equivalent to a Kimian state as a part of their internal structure (Flexible D-States) are associated with state nouns.

In “‘I know the answer’: A Perfect State in Capeverdean”, by Fernanda Pratas, a solution to the puzzle of the temporal reference of Capeverdean bare verb forms is suggested. It has been claimed that in Capeverdean, as well as in many other Creole languages, stative predicates unmarked for tense and/or aspect have a present reading, whereas bare eventive predicates have a past reading. However, the lexical stativity criterion fails to account for the fact that a series of stative verbs consistently patterns with eventives, i.e., their bare forms cannot have a present reading (e.g. kridita ‘believe’, lenbra ‘remember’, ama ‘love’, odia ‘hate’, etc.). In order to provide a solution to the problem, the author resorts to the concept of ‘Perfect State’. The structure of bare predicates can be better accounted for if a zero operator is introduced, which is a null Perfect marker. In this analysis, all bare verbs in simple sentences, both stative and eventive, are marked with a null Perfect morpheme. For predicates such as ‘sabe risposta’ (‘know the answer’), the Perfect State is a type of result state of past eventuality, whereas for other predicates, it is merely an abstract state of an event that has occurred.

“Stressed vowel duration and stress placement in Italian: What paroxytones and proparoxytones have in common”, by Stefano Canalis and Luigia Garrapa, examines some controversial points of the interplay between stressed vowel duration and word stress in Italian. According to the standard position, Italian stressed vowels in non-final open syllables are lengthened (no matter whether they are penultimate or antepenultimate), whereas all other vowels are considered to be short. However, in many works, stressed vowels in open syllables are reported to be shorter in proparoxytones than in paroxytones. The paper presents experimental data suggesting that penultimate and antepenultimate stressed vowels have roughly the ‘same’ duration in Italian: absolute duration is slightly longer in paroxytones, as reported in previous studies, while relative duration is constant. The ratio between stressed vowel duration and post-tonic duration is demonstrated to be unaffected by stress position. The experiment also shows that there is a final, secondary stress in proparoxytones. As for weight-sensitivity in Italian, stress placement in loanwords, acronyms and non-standard pronunciations provides evidence that it is no longer productive. As the result of a conflict between a preference for penultimate stress and the need to parse all syllables, even-syllabled words tend to receive penultimate stress, whereas odd-syllabled ones tend to be antepenultimate.

“Serial prosodification and voiced stop geminates in Catalan”, by Francesc Torres-Tamarit and Claudia Pons-Moll, provides an account of the process of voiced stop gemination in Central Catalan within the framework of Harmonic Serialism. In Catalan, root-final clusters consisting of a labial or velar stop followed by an alveolar lateral undergo gemination. If these clusters appear before a vowel belonging to the root, voiced stops spirantize and the cluster is parsed as a complex onset. Gemination can be triggered only when the voiced stop is syllabified in coda position in order to fix ill-formed rising sonority arising through intersyllabic contact. An epenthetic schwa or a vowel-initial suffix do not block gemination, although due to their presence, a phonological context capable of bleeding the application of gemination is created, i.e., the voiced stop syllabified as part of a complex onset. An account of these data requires the ordering of different phonological operations. Thus, Harmonic Serialism, as it posits serial derivations with intermediate steps, provides a straightforward explanation of the data in question.

“Interfacing information and prosody: French wh-in-situ questions”, by Viviane Déprez, Kristen Syrett and Shigeto Kawahara, discusses the interaction of information structure, pragmatics, prosody and syntax in the licensing of wh-in-situ questions in French. The authors support the viewpoint that in French information structure, syntax, and prosody are tightly intertwined in the formation of questions. The paper presents an experimental study based on Cheng and Rooryck’s (2000) proposal, which attributed to the above-mentioned questions some particular properties. One of them is licensing by an intonation morpheme that induces an obligatory sentence-final rising intonation contour, also present in purely intonational yes/no questions. The experimental data indicates that the intonation contours of wh-in-situ and yes/no questions are both rising, but distinct from each other: most, but not all participants produced a rising contour in wh-in-situ questions, while the slope was not as steep as in yes/no questions. The authors also explore discourse conditions that affect the felicity of wh-in-situ questions and develop an analysis appealing to movement through givenness-marking that appears to provide an explanation for the observed pitch compression.

“VP Ellipsis: New evidence from Capeverdean”, by João Costa, Ana Maria Martins and Fernanda Pratas, examines a non-trivial pattern of verb-phrase ellipsis (VPE) in Capeverdean. In this language, VPE can occur in answers to yes/no questions, whereas it is unacceptable in coordination structures of the type “You bought a new book and Maria did too”. At the same time, both polar question/answer pairs and coordination structures are considered to be typical licensing contexts for VPE in the languages exhibiting this phenomenon, for instance, English or Portuguese. The paper aims to identify the structural locus of cross-linguistic parametric variation in the domain of VPE. It is claimed that VPE is licensed by the polarity-encoding head Σ. In contrast to coordination structures, in answers to yes/no questions, Σ is projected above VP in Capeverdean. Only in this configuration, the verb may move out of the VP such that VPE is licensed. The availability of VPE in a language is claimed to be dependent on the interaction between clause structure, the ±V-relatedness of Σ, and verb movement.

“Anti-repair effects under ellipsis: Diagnosing (post-)syntactic clitics in Spanish”, by Andrés Saab and Pablo Zdrojewski, examines clitic doubling and its relation with extraction and ellipsis in ‘Rioplatense’ Spanish (Argentina). The paper aims to show that accusative doubling ameliorates the same types of island effects as clitic left dislocation (CLLD), although both doubling configurations exhibit different behavior in other relevant respects. Island repair effects disappear under ellipsis only with clitic doubling. This fact can be explained if one assumes that resumption can take place at the syntactic level or at the level of phonetic form (PF) and that certain island effects are only calculated at PF. A further proposal is to consider CLLD as an instantiation of syntactic resumption and CD as a case of PF resumption. PF resumption is bled under ellipsis and, consequently, there is no island repair effect in contexts of CD and ellipsis. Thus, anti-repair effects under ellipsis can be diagnostic when (post-)syntactic processes are at stake. It can be concluded that Kayne’s Generalization, implying that clitic doubling is dependent on differential object marking, must be treated as a purely PF-phenomenon.

“On the argument structure of the causative construction: Evidence from scope interactions”, by Francesco Costantini, is dedicated to the argument structure of the causative construction (namely, the so-called ‘faire-infinitive’) in Romance languages and the respective empirical evidence in the domain of scope interactions. The data discussed in the paper show that the Romance causative construction does not share its argument structure properties with the double object construction. The cause and the object appear to be arguments of the causativized predicate, while the causer is introduced outside the causativized vP, arguably by the causative head, as was previously proposed in the literature. Moreover, the causee moves from its merger position to a higher functional position and checks dative case.


On the one hand, this collection of papers covers a wide range of topics and theories, and for this reason it can be useful for many linguists, whether they are specializing in Romance languages or not. On the other hand, each of the papers is a highly specialized contribution to the discussion of some hot issue in a particular field, so the reader needs to have a good background in specific formal approaches in order to be able to appreciate it. It should be noted that in spite of the strong theoretical focus of the volume, empirical evidence is given much importance by all the authors, which is undoubtedly positive.

For instance, Pratas, in her paper, successfully applies the theory of Aktionsart to the Capeverdean verb system in order to account for apparent idiosyncrasies in the expression of tense, uniting semantic theory and solid typological evidence, which is valuable, as contemporary approaches to the classification of eventuality types can yield interesting results when applied cross-linguistically, both for the theory itself and for language description. Since the limits of space prevent discussion of every contribution, I would like to mention just one more paper I found particularly compelling, namely “State nouns are Kimian states”, by Fábregas and Marín, which not only provides new insights into the semantic structure of nominal states, poorly covered in the existing literature, but also gives rise to new questions concerning the (un)availability of certain types of nominalizations for different subclasses of stative verbs to be explored in further research.

As for the quality of editing, numerous misprints and various inconsistencies (for instance, the lack of an abstract and key words on p. 115) are somewhat deluding, especially for such a prestigious (and expensive) volume. However, on the whole the book leaves a very good impression due to the high quality of its contents.


Cheng, Lisa Lai-Shen & Rooryck, Johan. 2000. “Licensing wh-in-situ.” Syntax 3: 1–19.
Anna Alexandrova holds a degree in Russian and English philology. Now she is a second year PhD student of linguistics at Scuola Normale Superiore di Pisa (Italy). Her research interests include linguistic typology, Aktionsart, aspectual systems and verbal morphology both in synchrony and diachrony.