Frank Van Eynde

NP-internal agreement and the structure of the noun phrase

For the analysis of the noun phrase, the treatment which currently prevails in generative grammar is the one in which the head of the noun phrase is identified with the determiner, rather than with the noun. This D(et)P treatment has the advantage of providing a uniform account of all syntactic categories, both the substantive and the functional ones, and it provides a natural way to capture the co-occurrence restrictions between nouns and determiners, but it also faces a number of empirical problems. To solve them I propose an analysis in which the head of the noun phrase is identified with the noun, but in which the advantages of the DP treatment are incorporated as much as possible. This is done in two steps. First, I argue that the requirement (or the desirability) of a uniform treatment of all syntactic categories does not by itself favour the DP treatment, since there is no empirical evidence for the postulation of a separate syntactic category for the determiners. The argumentation is mainly based on an analysis of NP-internal agreement data and leads to the conclusion that the class of determiners is syntactically heterogeneous: there are the adjectival determiners, which are subject to morpho-syntactic agreement, and (pro)nominal ones, which are exempt from this agreement. Second, I dissociate the roles of head and selector. All prenominals, both the specifying and the modifying ones, are treated as functors which select a nominal head, rather than as heads which select a nominal complement. This functor treatment accounts in a natural and straightforward way for both morpho-syntactic agreement and semantic types of agreement. The language which is used for exemplification is Dutch, but at various points comparisons are made with German and English.

Large Scale Syntactic Annotation of Written Dutch: Lassy

This chapter presents the Lassy Small and Lassy Large treebanks, as well as related tools and applications. Lassy Small is a corpus of written Dutch texts (1,000,000 words) which has been syntactically annotated with manual verification and correction. Lassy Large is a much larger corpus (over 500,000,000 words) which has been syntactically annotated fully automatically. In addition, various browse and search tools for syntactically annotated corpora have been developed and made available. Their potential for applications in corpus linguistics and information extraction has been illustrated and evaluated in a series of case studies.

The Eurotra linguistic specifications: An overview

In this article, we outline the contents of the linguistic specifications of the Eurotra machine translation system. We start in sections 1 and 2 from some of the requirements placed by multilingual MT on the overall design of the linguistic components. We then move on to a characterization of the Eurotra interface structure (section 3), the nature of transfer (section 4), and trends towards more interlingual representations within the project (section 5). Thereafter, we concentrate on the contents of the various levels beside the interface structure (section 6) before giving a brief survey of word structure (section 7) and outlining some areas for further research (section 8)The authors of this article are indebted to many other members of the project too numerous to be mentioned here. They wish to record a special intellectual debt to previous members of the Eurotra Linguistics Specification team and, in particular, Doug Arnold, Louis des Tombe and Lieven Jaspaert who did so much to establish sound theoretical bases for multilingual MT (see inter alia Arnold, Jaspaert and des Tombe 1985; Arnold 1986; Arnold and des Tombe 1987). For an extensive version of the overview presented here, see Allegranza et al. 1991. For another recent presentation of Eurotra, see Raw, Vandecapelle and Van Eynde 1988..

The analysis of tense and aspect in Eurotra

This paper presents a framework for the model-theoretic analysis of tense and aspect forms in discourse. It has been developed for Eurotra, the MT project of the European Community, and has been applied to the nine Eurotra languages: English, German, Dutch, Danish, Greek, Italian, French, Spanish and Portuguese.The paper consists of six parts. The first presents the problem of translating tense and aspect forms and indicates the type of solution I envisage. The second contains a formalism for the representation of time meanings. The third and the fourth present a theory of tense and aspect respectively. The fifth discusses the issue of compositionality and the sixth is about the use of the system in the Eurotra framework.

Iteration, habituality and verb form semantics

The verb forms are often claimed to convey two kinds of information:1. whether the event described in a sentence is present, past or future (= deictic information.2. whether the event described in a sentence is presented, as completed, going on, just starting or being finished (= aspectual information)It will be demonstrated in this paper that one has to add a third component to the analysis of verb form meanings, namely whether or not they express habituality.The framework of the analysis is model-theoretic semantics.

Morpho-Syntactic Agreement and Index Agreement in Dutch NPs

For the treatment of agreement in Dutch NPs I adopt a distinction, familiar from HPSG, between morphosyntactic agreement and index agreement. In order to determine their respective roles, I make another distinction, proposed in Van Eynde (2003), between marked and unmarked nominals. Employing these distinctions, I will demonstrate that the combination of prenominal adjectives and determiners with unmarked nominals is subject to morphosyntactic agreement, whereas the combination of these prenominals with marked nominals is subject to index agreement.

Natural Language Generation from Pictographs

We present a Pictograph-to-Text translation system for people with Intellectual or Developmental Disabilities (IDD). The system translates pictograph messages, consisting of one or more pictographs, into Dutch text using WordNet links and an ngram language model. We also provide several pictograph input methods assisting the users in selecting the appropriate pictographs.

The semantics of tense and aspect

The present chapter describes a range of formal semantic accounts of tense 10438 and aspect, constituting a modest portion of the vast literature on tense and 10439 aspect (e.g., Binnick, 2012; Mani et al., 2005). The focus is on the nature of 10440 the ingredients assumed, including the pairs �w, t� of possible worlds w 10441 and moments t of time in Montague (1973), expansions of the moments t to 10442 intervals (Bennett & Partee, 1972; Dowty, 1979) which generalize to formal 10443 occurrences (Galton, 1987), reductions of worlds w to situations (Barwise & 10444 Perry, 1983) events/eventualities (Kamp, 1979; Bach, 1981; van Lambalgen & 10445 Hamm, 2005), incomplete events (Parsons, 1990), branching (Landman, 1992), 10446 event nuclei (Moens & Steedman, 1988), and related complexes (Pustejovsky, 10447 1991; Kamp & Reyle, 1993; Pulman, 1997). The chapter formulates these 10448 notions in finite-state terms, building strings that approximate timelines, a 10449 logical starting point for which is Priorean tense logic (Prior, 1967). 10450 At the heart of Priorean tense logic, commonly called temporal logic (e.g., 10451 Emerson, 1992), is a satisfaction relation |=A defined relative to a model 10452 A. A simple example of |=A at work is the analysis (1b) below of (1a) as 10453 Past(adam-leave-the-garden), with a time parameter changing from t to t�. 10454

CONSTRAINT-BASED LEXICA

As the field of generative linguistics has developed, the lexicon has taken on an increasingly important role in the description of both idiosyncratic and regular properties of language. Always viewed as a natural home for exceptions, the lexicon was given relatively little work in the early years of transformational grammar. Then Chomsky proposed in 1970 (Chomsky, 1970) that similarities in the structure of deverbal noun phrases and sentences could be expressed in terms of a lexical relationship between the verb and its nominalization. Jackendoff (1975) characterized further lexical regularities in both morphology and semantics, and Bresnan (1976, 1982) pioneered the development of a syntactic framework (Lexical Functional Grammar) in which central grammatical phenomena such as passivization could be explained within the lexicon. A parallel line of work by Gazdar (1981) called Generalized Phrase Structure Grammar sought to provide a nontransformational syntactic framework, by employing metarules over a context-free grammar. Gazdar et al. (1985) constrained the power of those metarules by restricting them to lexically-headed phrase structure rules. Pollard and Sag (1987, 1994) built on the work in GPSG, outlining the more radically lexicalist framework of Head-driven Phrase Structure Grammar (HPSG), abandoning construction-specific phrase structure rules in favor of a small number of rule schemata interacting with a more richly articulated lexicon to capture relevant syntactic generalizations. As a well-known and widely used constraint-based grammar formalism1 HPSG will serve us well in this chapter by providing a precise linguistic framework within which we can organize the relevant data and examine the technical devices available for analysis of that data. For those who are not familiar with the notation, we first provide a brief introduction to this framework.

Translating text into pictographs

We describe and evaluate a text-to-pictograph translation system that is used in an online platform for Augmentative and Alternative Communication, which is intended for people who are not able to read and write, but who still want to communicate with the outside world. The system is set up to translate from Dutch into Sclera and Beta, two publicly available pictograph sets consisting of several thousands of pictographs each. We have linked large amounts of these pictographs to synsets or combinations of synsets of Cornetto, a lexical-semantic database for Dutch similar to WordNet. In the translation system, the Dutch input text undergoes shallow linguistic analysis and the synsets of the content words are looked up. The system looks for the nearest pictographs in the lexical-semantic database and displays the message into pictographs. We evaluated the system and results showed a large improvement over the baseline system which consisted of straightforward string-matching between the input text and the filenames of the pictographs. Our system provides a clear improvement in the communication possibilities of illiterate people. Nevertheless there is room for further improvement.