This dissertation presents a qualitative and quantitative analysis of the interaction between grammatical structure and prosodic structure in two Australian languages, Dalabon and Kayardild. The typological profiles of these languages contrast dramatically: Dalabon is an extreme head-marking polysynthetic language, in which a single verb can carry all of the information typically associated with a clause in a language like English (e.g. M. Baker, 1996; Evans & Sasse, 2002). Kayardild is an extreme dependent-marking language, with syntactic relationships encoded on nominal dependents through sequences of case morphology (Dench & Evans, 1988; Evans, 1995a; Nordlinger, 1998). One of the main motivations for studying the interaction between grammatical structure and prosodic constituency in two grammatically distinct languages is that a comparison will potentially test the effects that grammatical structure may have on prosody. By examining two structurally different languages it then becomes possible to test certain hypotheses regarding the grammatical influence on prosodic structure. Whether an extreme dependent-marking language and an extreme head-marking language show similar or different prosodic structural patterns may either support or disprove statements regarding the universality of the interaction between prosodic and grammatical structure. This dissertation presents both quantitative and qualitative findings, based on approximately 75 minutes of recorded Dalabon and Kayardild narratives. The hypotheses examined in this dissertation concern whether factors such as grammatical complexity, the location of clause boundaries, discourse effects such as the introduction of new information, as well as prosodic length, affect prosodic constituency boundary location and strength, irrespective of the language type in question. Although the two languages differ dramatically in their grammatical structures, the overall interaction between prosody and the various grammatical factors may be expected to show similar patterns. In order to test these hypotheses, the relationships between grammatical complexity and prosodic constituency, between the clause and prosodic constituency, and between prosodic length and prosodic constituency are examined. The results presented here show that there is a large amount of consistency across both languages in terms of the prosodic phrasing of clauses overall, supporting the view that prosodic structure is independent of grammatical structure. However, the results also show there is interesting variation in the prosodic phrasing of constituents within the clause as well as overall pausing patterns, suggesting that typological structure does have a role to play in prosodic structure. The chapters are structured as follows: Chapter 1 provides the introduction and context for the research, with definitions of prosody and intonation, a review of the literature on the interaction between prosody and grammar, and the aims and hypotheses. Chapter 2 provides language overviews of Dalabon and Kayardild. Chapter 3 provides a description of the methods and materials, including an overview of the corpus, and the statistical methods and annotation conventions used. Chapter 4 provides descriptions of the intonation phrases found in Dalabon and Kayardild, including a description of the contour types found, the makeup and boundary tones of these phrases. Chapter 5 provides an examination of the relationship between the clause and prosodic phrasing including analyses of pause, IP boundaries, clause types, constituents and word order. In Chapter 6 some of the interesting findings to emerge from the preceding analyses are investigated. This includes the prosodic behaviour of nominals and examples of IPs which comprise multiple verbs. Chapter 7 concludes the dissertation, providing an overview of the main findings and their implications, as well as a discussion of directions for future research.
Stored only when both papers are in the corpus — 126 citations exist upstream.
Imported from OpenAlex (CC0) — nothing here is generated.