Search

It has long been proposed that noun and verb are better conceived not as discrete, but as continuous categories (Ross 1973, Simone & Pompei 2007). The verb-noun continuum includes different kinds of nouns, some of which are characterized by verbal semantic features, such as aspectual and actional ones; for example, nouns referring to events (i.e. event nouns, Gross & Kiefer 1995) can show the verbal characteristics of duration, dynamicity and telicity. Verbal nouns are realized by language-specific morphosyntactic strategies. For example, event nouns can be rendered by forms belonging to verbal paradigms: alongside the gerund, the maṣdar and the participle, many languages use the infinitive (Simone 2003). This is the case with Italian and other Romance languages. The nominal uses of the Italian infinitive have been widely analyzed (Tekavčic 1972, Vanvolsem 1983, Skytte 1983, Skytte & Salvi 1991, Jansen et al. 2002, Thornton 2004, Simone 2004). Nevertheless, one feature has been underestimated - the syntactic and semantic differences between bare infinitives (1) and infinitives preceded by a determiner (2) -: (1) Girovagare per Roma mi rilassa. wander.INF for Rome PRON.1SG-ACC relax. PRES.3SG 'Wandering around Rome relaxes me'. (2) Dopo un’ ora di questo improduttivo girovagare tornammo indietro After a hour of this pointless wander.INF go.PAST.1PL back ‘After one hour of this pointless wandering we went back’. This study firstly aims at identifying the main syntactic differences between bare infinitive and infinitive preceded by a determiner. Secondly, our proposal is to distinguish different patterns of determiner and infinitive: definite article plus infinitive (e.g. il girovagare ‘the wandering’), indefinite article plus infinitive (e.g. un girovagare ‘a wandering’), proximal demonstrative plus infinitive (e.g. questo girovagare ‘this wandering’) and distal demonstrative plus infinitive (e.g. quel girovagare ‘that wandering’). We intend to show that features measurably more verbal or more nominal characterize the different forms of determined infinitives. Accordingly, they can be located on different positions along the verb-noun continuum. Our analysis includes three main points. First, it shows the relation between the distributional properties and the semantic variation of these different kinds of nominal infinitives. Secondly, it intends to identify different degrees of 'verbiness' and 'nouniness' related to each kind of nominal infinitive. Thirdly, in a broader perspective, it aims at showing differences and similarities within other Romance languages, in order to identify the Romance infinitive patterns. We drew data from three different corpora (La Repubblica written corpus for Italian, Corpus del Español Actual for Spanish and Chambers-Rostand du français journalistique.for French), in order to clearly show the syntactic differences between the patterns identified. Such kinds have been evaluated against their compatibility with two different sets of features. First, we have examined typical verb properties, such as the agent expression, the expression of duration with verb-specific constructions, the possibility to be modified by adverbs, to be negated, and to govern a direct object. Secondly, we have considered features typical to nouns, such as the possibility to be modified by adjectives, the expression of the subject and the object as a genitive, and the possibility to express the duration by specifically nominal syntactic structures. Les études précédentes sur les classes de mots ont démontré que nom et verbe ne sont pas des catégories discrètes, mais continues (Ross 1973, Simone & Pompei 2007). Le continuum entre le verbe et le nom comprend différents types de noms, dont certains sont caractérisés par des traits sémantiques verbales, tels que ceux aspectuels et actionnels; par exemple, les noms se référant aux événements (c.-à-d., noms d'événement, Gross & Kiefer 1995) peuvent montrer les caractéristiques verbales de la durée, du dynamisme et de la télicité. Les noms verbaux sont construits par des stratégies morphosyntaxique spécifiques aux langues. Par exemple, les noms d'événement peuvent être générés par les formes appartenant à des paradigmes verbaux : aux côtés du gérondif, le maṣdar et le participe, plusieurs langues utilisent l'infinitif (Simone 2003). C'est le cas de l'italien et des autres langues romanes. Les fonctions nominales de l'infinitif italien ont été largement analysées (Tekavčic 1972, Vanvolsem 1983, Skytte 1983, Skytte & Salvi 1991, Jansen et al. 2002, Thornton 2004, Simone 2004). Néanmoins, un point très important a été sous-estimé : les différences syntaxiques et sémantiques entre infinitifs sans déterminant (1) et ceux précédés d'un déterminant (2): (1) Girovagare per Roma mi rilassa. promener.INF pour Rome PRON.1SG-ACC détendre. PRES.3SG 'Promener autour de Rome me détend'. (2) Dopo un’ ora di questo improduttivo girovagare tornammo indietro Après une heure de cet improductif promener.INF aller.PAST.1PL arrière ‘Après une heure de ce promener improductif, nous revînmes en arrière’. Tout d'abord, ce travail a l'objectif d'identifier les différences syntaxiques plus importantes entre l'infinitif sans déterminant et celui avec le déterminant. En deuxième lieu, dans cet article on propose de distinguer entre quatre types syntaxiques différents d'infinitif déterminé : infinitif plus article défini (p. ex. il girovagare « le promener »), article indéfini plus infinitif (p. ex. un girovagare « un promener »), démonstratif proximal plus infinitif (p. ex. questo girovagare « ce promener(-ci) »), démonstratif distal plus infinitif (p. ex. quel girovagare « ce promener(-là) »). L'objectif de ce travail est celui de montrer que les différentes formes d'infinitifs déterminés sont caractérisées par éléments variablement plus verbales ou plus nominales. En conséquence, elles peuvent être situées sur différentes positions du continuum verbe-nom. Notre analyse comprend trois points principaux. Tout d'abord, il montre la relation entre les propriétés distributionnelles et la variation sémantique de ces différents types d'infinitifs nominales. En deuxième lieu, on a l'intention d'identifier les différents degrés de « verbalité » et « nominalité » associés à chaque type d'infinitif nominal. En troisième lieu, dans une perspective plus large, on montre les différences et les similitudes dans les autres langues romanes, afin d'identifier les configurations infinitives Romance. En ce qui concerne les deux premiers points, on a extrait les données de trois différents corpora (le corpus La Repubblica pour l'italien, le Corpus del Español Actual pour l'espagnol et le corpus Chambers-Rostand du français journalistique pour le français), afin de montrer clairement les différences syntaxiques entre les profils identifiés. Ces types ont été évalués contre leur compatibilité avec deux ensembles différents de propriétés. Tout d'abord, nous avons examiné les propriétés typiques du verbe, comme l'expression de l'agent, l'expression de la durée avec des constructions verbales spécifiques, la possibilité d'être modifié par des adverbes, la négation et la présence d'un objet direct. Deuxièmement, nous avons considéré les caractéristiques typiques du nom, tels que la possibilité d'être modifiés par les adjectifs, l'expression du sujet et de l'objet comme un génitif et la possibilité d'exprimer la durée par des structures syntaxiques spécifiquement nominales.

Questo contributo propone una descrizione corpus-driven di un tipo di strutture combinatorie, accomunate da specifiche proprietà morfosintattiche e semantiche: le combinazioni parzialmente riempite (CPR). Le CPR sono caratterizzate da un basso grado di specificità lessicale, a cui corrisponde un’alta predicibilità semantica; si configurano attraverso un formato superficiale caratterizzato da posizioni vuote, con una caratterizzazione semantica stabile e un livello di produttività variabile. Al pari delle costruzioni, le CPR rappresentano un’associazione di forma e significato (Goldberg), e sono in grado di veicolare specifici significati attraverso la configurazione sintattico/semantica. Le CPR possono essere classificate sulla base di diversi parametri, tra cui le proprietà funzionali, il grado di idiomaticità, il livello di specificità lessicale e la produttività della struttura. Attraverso l’analisi di dati estratti da corpora, il contributo si propone di identificare e analizzare i formati parzialmente riempiti in alcune lingue di ambito romanzo (italiano, spagnolo e francese), e di mostrarne una loro possibile rappresentazione formale e una classificazione in tipi, tenendo conto delle eventuali differenze interlinguistiche. Si noterà come le stesse CPR nelle diverse lingue possono essere caratterizzate da tratti di schematicità e livelli di produttività variabili. Infine, si dimostrerà che le CPR rappresentano un fenomeno di regolarità del lessico, nonostante siano caratterizzate da un basso grado di specificità lessicale. Nuovi elementi lessicali possono originarsi a partire da una specifica costruzione, e convenzionalizzarsi attraverso la ripetizione. This investigation proposes a corpus-driven description of partially filled constructions (PFCs), a subtype of phraseological units characterized by a low degree of lexical specification and a high level of semantic predictability. PFCs contain both fixed and empty slots, showing variable levels of productivity. As constructions, they are pairings of form and function (Goldberg), since their semantic/syntactic configuration can convey particular sets of meanings. Through the analysis of authentic data extracted from corpora of three different Romance languages (Italian, French, and Spanish), this contribution aims at i) identifying and analysing a set of partially filled units, ii) showing a formal representation of such syntactic-semantic patterns, iii) classifying them into different types on the basis of functional, syntactic, and semantic criteria, as well as on the degree of idiomaticity and lexical variation, and iv) driving a cross-linguistic comparison among the Romance languages. The same PFC may be characterized by different levels of schematicity and productivity in the various languages taken into account. It will be shown that PFCs may be associated with different syntactic structures, either at the phrase-level or at the sentence-level. Even though they are characterized by empty slots and by a lower degree of lexical specification, they still represent a regular area of the lexicon. New and independent lexemes can develop from specific constructions and become conventionalized through repetition.

This paper focuses on a specific type of Multiword Expressions, particularly widespread in Italian as well as in other Romance languages: Multiword Modifiers, i.e. prepositional phrases functioning as modifiers of a noun (Multiword Adjectives) and of a verb (Multiword Adverbs). Exploiting both syntactic and semantic analysis, this paper explores the hypothesis that Multiword Modifiers are formed on the basis of regular syntactic templates, which can structure and organize the semantic information associated with words. In this perspective, after a brief presentation of Multiword Lexical Units and the class of Multiword Modifiers, the methodology and the general theoretical framework of this study will be explained. The last section is devoted to the analysis of some semantic relations frequently fulfilled by Multiword Modifiers of Italian, French and Spanish. This investigation aims at demonstrating that all Romance languages considered make a regular use of this kind of analytical resource in adjectival or adverbial function, showing similar patterns and syntactic templates.