The argument structure of the Dâw language

Author(s)

Jéssica Clementino da Costa

• Document type: Master's Dissertation
• Press: São Paulo.
• Institution: Universidade de São Paulo (USP). Faculdade de Filosofia, Letras e Ciências Humanas (FFLCH/SBD)
• Defense date: 2014-06-09

Examining board members

Luciana Raccanello Storto; Esmeralda Vailati Negrão; Kristine Sue Stenzel

• Advisor: Luciana Raccanello Storto

Abstract

This thesis describes and analyzes the argument structure and verbal classes of the Dâw language (Nadahup family, Amazon). We studied the verbs of that language from the semantic and syntactic perspective, identifying classes and subclasses according to the morphosyntactic behavior of verbal roots. Furthermore, we evaluated the descriptive and explanatory hypotheses of verb classes identified by Martins (2004), the first researcher to address Dâw morphosyntax. Our theoretical framework is the theory of argument structure developed by Hale & Keyser (2002), which proposes an analysis of the syntax and semantics of lexical items by means of the argument structure the pattern of structural relations between the head and its arguments within syntactical structures projected by the head itself. Through various language tests, including verbal valency alternation and judgment of (a)grammaticality, we reclassified the nine verb classes identified by Martins (2004) into three classes according to the verbal valency: the classes of intransitive verbs, transitive verbs and bitransitive verbs. Martins (2004) states that, in the sentence, the verbs may change in tone due to the presence of a transitivizing or intransitivizing tonal morpheme. However, we show in this paper that the tonal system of the language is predictable at the sentence level. Thus, regardless of the valency-increasing process involved, we realized that the tonal change of verbs arises due to the phonological phrasing of sentences. Regarding the transitivization process, subclasses of intransitive verbs were identified: alternating and non-alternating verbs. The restrictions on alternation are due to the argument structure of each verb type. In the case of unaccusative or alternating intransitive verbs, we observed that they are formed from a composite dyadic structure, projecting an internal specifier and a complement, which allows them to switch between intransitive and transitive forms. In the case of non-alternating verbs we found three patterns: denominal and unergative verbs, based on a monadic argument structure (that does not project internal specifier) that prevents alternation; non-alternating unaccusative verbs based on a monadic structure that takes a basic dyadic structure as a complement verbs of this type do not alternate because they are not formed by a dyadic structure, but contain such a structure n e jectiv l ve bs, f me f m c p l ve b th t t kes n jective s complement. Since the root and verbal head have full (non-empty) phonological content, no conflation is possible between head and root, which prevents the formation of the verbal predicate. This structure explains the agrammaticality of these verbs with regard to the automatic transitivization process. We also tested the syntax and semantics of intransitivization (inchoative constructions, passive, reflexive and middle voices). In general, we found that there is no specific morphology for constructing middle, inchoative or anticausative sentences. There are no passives in Dâw; in place of this voice, the speakers form sentences that are inchoative or have a subspecified subject. Reflexive sentences are created using reflexive pronouns in the position of the object of the sentence. Finally, we found that direct objects of transitive sentences are marked by the {-uuy\'} morpheme analyzed by us as DOM. Its use is subject to semantic constraints of definiteness and animacy (AU)

• FAPESP's process: 11/16168-2 - Argument structure in Dâw
• Grantee: Jéssica Clementino da Costa
• Support Opportunities: Scholarships in Brazil - Master