Retour vers liste

Détail de la contribution

Auteur: Anna SÖRÉS

Middle verb morphology in Hungarian and in a cross-linguistic perspective

Abstract/Résumé: The main goal of this work is to propose the mapping of form and meaning in the domain of middle voice by using a set of semantic and morphosyntactic features. This analysis investigates Hungarian data which are placed into a crosslinguistic perspective based on recent typological literature. Hungarian has several morphemes that encode reflexive and reciprocal situation and different middle situation types: kodik, kozik, όdik,-odik, ul, -ad (and several allomorphs). The paper addresses the following questions: 1. Is it possible to map these middle markers into the semantic subclasses proposed by Kemmer (1993)? 2. Does Hungarian confirm the typological generalization found in the literature (Kemmer 1993, Kazenin 2001) according to which verbs of change of body posture, translational and nontranslational motion patterns with grooming? 3. Can middle forms systematically be derived from reflexives (Maldonado 2009)? Methodology: The list of Hungarian middle verbs has been established according to Kemmer’s list. They have been grouped according to their suffix type. The middle domain will be presented by as continuum between active and passive, using the following set of criteria: •Properties of the Subject: ±human; ± animate; ± volitional; ± control. •Thematic role of the subject; •Mapping into the semantic subgroups of verbs. Results of the analysis: 1. There is no one-to one mapping between middle markers and semantic subgroups, with exception to anticausatives expressing change of state. In contrast to the other groups, not all middle verbs are derived: some of them form a functional pair with their transitive-causative counterpart, e. g. fordul ‘to turn intr.’ ~fordít ‘to turn trans.’. 2. In Hungarian middle forms may have different morphological realizations. E. g. grooming verbs are marked by one of the middle suffixes. Motion verbs show unmarked intransitive morphology. The typological generalization proposed by Kemmer (1993) is not confirmed. 3. Since reflexives are only anaphoric, middle verbs cannot be considered to be derived from these forms. Consequently, the Hungarian data fall within the scope of analysis proposed by Maldonado (2009). In Hungarian, reflexive pronouns are inflected for case, thus, the two participants are morphosyntactivally distinguished, even in case of coreference. Conclusion: This research shows that some parameters may be extremely relevant to the analysis of the middle domain, namely the semantic properties of the subject and the alternation between middle marking and unmarked intransitive morphology.