Untitled

Funder

Organizational Unit

Publications

Capeverdean reflexives: the importance of a silent Voice

Publication . Pratas, Fernanda

In Capeverdean, a Portuguese-based Creole language, many reflexive contexts do not show any overt reflexive expression. This is the case of transitive verbs like bisti ‘dress’ in simple clauses: Ana bisti ‘Ana has dressed herself’. This is a perplexing fact, given that there is an anaphor of the SELF-type available in the language: (si) kabesa — literally ‘his/her head’ —, meaning ‘himself/ herself’, which participates in reflexive clauses with other verbs. The current paper explores this puzzle, ending with a proposal supported empirically and also by recent studies for other languages. This novel analysis goes as follows: all Capeverdean finite sentences, except unaccusatives, have a Voice head, responsible for assigning external theta-roles. This also includes middles, passives and this type of reflexives. It is this Voice head that, in spite of being silent, attracts the internal argument to a preverbal position and provides the interpretation for an implicit external argument, which is syntactically active.

2014Journal article

Open access

Comitative Coordination in Capeverdean

Publication . Brito, José António; Matos, Gabriela; Pratas, Fernanda

Comitative coordination in Capevedean, a Portuguese-based Creole language, differs from comitative coordination in Portuguese and other European languages by the wider range of syntactic categories it conjoins and its insensitivity to the lexical-semantic selection of the main verb. In this respect Capeverdean behaves like with-languages. However, the comitative conjunction ku may not coordinate adjectival or tensed verbal predicates, as well as finite sentences. In this paper it is claimed that the widespread occurrence of ku in Capeverdean is due to its full grammaticalized status as a conjunction, and its impossibility to conjoin predicates and tensed domains is imputed to its comitative meaning: ku conjoins referential arguments or properties to derive a group entity or a compound property; it is excluded from contexts that denote related but independent properties or events and situations.

2015Journal article

Open access

The Perfective, the Progressive and the (dis)closure of situations: comment on the paper by María J. Arche

Publication . Pratas, Fernanda

In the present paper, inspired by María J. Arche’s work, “The construction of viewpoint aspect: the imperfective revisited” (2013, this issue), I add several pieces of evidence in favor of her proposal that viewpoint aspect does not alter the fundamental situation aspect properties of predicates. Namely, I discuss the temporal interpretations in Capeverdean, a Portuguese-based Creole language for which the salient opposition in the domain of viewpoint aspect is not between the imperfective and the perfective, but rather between the Progressive and the Perfect, here taken as semantically complex categories that involve certain temporal characteristics; crucially, imperfectivity is one of the features of the Progressive and a perfective viewpoint is part of the semantic complexity of the Perfect. I also discuss the role of for-time durational adverbials when combined with the perfective and propose that, in their presence, the relevant final boundary when telic predicates are at stake is not the culmination of the event, but rather the final point described by that time-argument. This proposal accounts nicely for the fact that, in these specific contexts, there is no contradiction in having this perfective clause conjoined with the assertion that the underlying telic situation is not completed.

2014Journal article

Open access

A alomorfia dos pronomes de objeto em caboverdiano

Publication . Salanova, Andrés Pablo; Pratas, Fernanda

In Capeverdean, object pronominals can be phonological enclitics or free-standing forms, either in accusative or dative contexts. Crucially, pronominal enclitic forms are ruled out on verbs carrying the past suffix -ba. This fact has been analysed (Baptista, 2002) as being related to the banning of clitic clusters. In this view, the affix -ba is considered a clitic, barring other enclitics. A problem of this view is that -ba has to be stipulated to be a clitic, where there is no evidence for this. In this paper, we propose a phonological solution to these Capeverdean facts, based on the following two elements, independently justified: (i) a simple stress rule for the language; (ii) a filter against shifting the stress out of the verb stem. If our phonologically based proposal is correct, it gives support to the distributed morphology framework (Halle e Marantz, 1993): morphemes are bundles of abstract (syntactic-semantic) features that are provided with phonological features at Vocabulary Insertion (VI), which is conditioned by phonological rules and constraints.

2014Journal article

Open access