Retour vers liste

Détail de la contribution

Auteur: Carola TRIPS

An analysis of phrasal compounds in the model of Parallel Architecture

Abstract/Résumé: Phrasal compounds (PC) are very interesting from a theoretical point of view since they challenge traditional (generative) frameworks based on syntactocentricity and a strict division between the lexicon and grammar. The few analyses put forward so far (cf. Lieber, 1988, 1992, for English, Gallmann, 1990, Wiese, 1996, Meibauer, 2003, 2007 for German) have failed to fully account for them, because PCs violate the No Phrase constraint (Botha, 1981) and the Lexical Integrity Hypothesis. This paper seeks to bridge this gap by providing an account of phrasal compounds in English, contrasted with German, in Jackendoff’s model of Parallel Architecture (Jackendoff, 1983, 1990, 2002, 2008, 2010b) which is not based on these tenets and therefore handles PCs in a much more satisfying way. More precisely, it tries to answer the following two questions Jackendoff has posed in his 2009 (2010) paper on compounding: 1. Concerning the basic functions of compounds, which of these are generally available for pragmatics and non-linguistic conceptualisation? 2. Concerning the structure of compounds, which other morphosyntactic patterns must be added to the simple N+N compounds in order to account for quasi-syntactic elaborations as e.g. health and welfare fund, and what is their status vis-à-vis morphology and syntax? A distinction between two types of PCs will be made on semantic grounds: the first type contains a predicate in the phrasal non-head and is based on the IS-A relation (this “Steffi is Great” attitude), the second type lacks a predicate and is based on general functions assumed for N+N compounds (a chicken-and-egg situation; cf. e.g. Fanselow, 1981, Meyer, 1993, Jackendoff, 2009, 2010a). Thus, the syntactic structure of PCs belies their true nature, implying that what is critical for an adequate interpretation is not their phrasal structure but their conceptual-semantic properties. Further, it will be shown that the “expressive flavour” attributed to PCs by Meibauer (2003, 2007) can be explained by the interplay between the IS-A relation for the first type of PC, and basic functions for the second type of PCs, the morphological redundancy rule for N+N compounds, type (mis)matching and, as a result, instances of metonymic coercion. This contribution shows that the model of PA, where semantic structures are built in an independent generative component of semantics linked via components of interface rules to generative components of syntax and phonology, can successfully be applied to compounding, and is cross-linguistically valid. It thus provides us with new insights into the nature of compounding and, in a more general sense, the workings of morphology.