Catálogo de publicaciones - libros

The article presents a formal analysis in the framework of bi-directional optimality theory of the free choice, ignorance and indifference implicatures conveyed by the use of indefinite expressions or disjunctions. Ignorance is expressed by standard means of epistemic logic. To express indifference we use Groenendijk and Stokhof’s question meanings. To derive implicature, Grice’s conversational maxims, and an additional principle expressing preferences for minimal models, are formulated as violable constraints used to select optimal candidates out of a set of alternative sentence-context pairs. The implicatures of an utterance of are then defined as the sentences which are entailed by any optimal context for (but not by itself). Entailment is defined in a version of update semantics where contextual updates are derived by competition among contexts. Free choice and other modal implicatures of disjunctions and indefinites will follow, but also scalar implicatures and exhaustification.

Structuring of information proceeds through the foregrounding of certain parts of the information. In general, foregrounding can be realized on various linguistic levels and it is possible to distinguish: Conceptual,Functional, Discourse and Pragmatic devices, which can be represented by various formal means: Phonetic-Phonological, Morphological-Syntactic and Lexical-Pragmatic. All the devices can co-occur during the information packaging. Some of them are obligatory and are on the high level of the hierarchically organized processes of foregrounding (e.g. conceptual or functional foregrounding); some of them are optional and they are defined by the specific discourse and/or pragmatic values of a sentence (e.g. focus or topic); some forms of foregrounding are implicational (e.g. sometimes reordering implies emphasis of intonation) and so on. The relations between the different kinds of foregrounding are language specific, but it seems possible to speak about universal models of formalization of the information structures. In Georgian there is no morphological topic marker, but all other devices of foregrounding are possible. The paper examines the main models of such devices.

The main of this paper to investigate Hungarian focus interpretations. Hungarian has a special pre-verbal position for focussed constituents, which receive an exhaustive interpretation. Since the focus sensitive particle, ‘only’ goes together with this exhaustive focus, ‘only’ seems to be redundant or superfluous. For this reasons, we will investigate focus and ‘only’ in answers and multiple focus constructions and propose an analysis where exhaustivity-operator and only-operator are distinct.

Simplicity and subdirect irreducibility of complex algebra duals of Kripke frames can be readily characterized in terms of roots of the corresponding Kripke frames.

Here these characterizations are generalized to the case of distributive modal algebras , and their duals. Such an algebra consists of a distributive bounded lattice (, ∨ , ∧ , 0, 1) together with a join preserving operator , a meet preserving operator , a join reversing operator ⊳, and a meet reversing operator ⊲.

This paper argues that at least some cases of productive language use, specif ic ally cases of apparent variation in word sense, require a treatment at the conceptual level rather than a lexical semantic solution. It is argued that the lexical semantics should be left underspecified in these cases, and the observed variation in truth-conditions should be attributed to differences in conceptual represent ation that result from differences in the utterance context. This involves rather drastic changes in our conception of the semantics-pragmatics interface, which are discussed in this paper.

This paper uses standard syntactic scoping properties (as found with lambda calculus and predicate logic) to account for linking dependencies in Tukang Besi, an Austronesian language of Indonesia. Basic argument dependencies are established with a combination of verbal agreement, case marking and constituent order. If taken alone, no one of these is sufficient to determine the grammatical relations in the clause. Rather these different factors operate in combination to yield unambiguous clauses. What is shown is that scope taking options determine which combinations are possible and which are impossible.

The purpose of our research is to find a transparent and consistent basis for representing the relations between the Georgian verb and its actants. We use the metaphor of an ”action chain” as a means to elucidate the complex interdependencies representing the most general aspects of the verbal meaning, as well as the core component of the utterance as a whole, namely the standpoint from which the speaker considers the event described in the utterance.

In this paper we consider some basic questions regarding the extensions of modal logics with bisimulation quantifiers. In particular, we consider the relation between bisimualtion quantifiers and uniform interpolation for modal logic and the -calculus. We first consider these questions over the whole class of frames, and then we restrict to specific classes, where we see that the results obtained before can be easily falsified. Finally, we introduce classes of frames where we found the same good behaviour than in the whole class of frames. The results presented in this paper have been obtained in collaboration with other authors during the last years; in alphabetical order: Tim French, Marco Hollenberg, and Giacomo Lenzi.

This paper is concerned with a possible mechanism for learning the meanings of quantifiers in natural language. The meaning of a natural language construction is identified with a procedure for recognizing its extension. Therefore, acquisition of natural language quantifiers is supposed to consist in collecting procedures for computing their denotations. A method for encoding classes of finite models corresponding to given quantifiers is shown. The class of finite models is represented by appropriate languages. Some facts describing dependencies between classes of quantifiers and classes of devices are presented. In the second part of the paper examples of syntax-learning models are shown. According to these models new results in quantifier learning are presented. Finally, the question of the adequacy of syntax-learning tools for describing the process of semantic learning is stated.

The present paper presents the structure of a cross-linguistic database of production data. The database contains annotated texts collected from a sample of fifteen different languages by means of identical data gathering methods, which are designed to enable studies on typology and universals of information structure. The special property of this database is that it combines the features of a natural language corpus and the features of a typological database. The challenge for the exploration interface is to provide user-friendly support for exploiting this particular type of resource, thus facilitating empirical generalizations about the collected data in the individual languages and comparison among them.