Pre-boundary lengthening in a cross-linguistic perspective

点击翻译按钮获取中文摘要

The project investigates the phenomenon of segmental lengthening preceding a prosodic boundary, commonly referred to as ‘pre-boundary lengthening’ (henceforth, PBL). Unlike in prior studies, this phenomenon is approached from a cross-linguistic perspective, comparing the lengthening domains of languages with different prosodic systems. This contributes to the question in how far aspects of PBL are universal, language-specific and/or tied to the word-prosodic system of a language. The project also investigates the temporal dynamics of lengthening among adjacent segments in the PBL domain, exploring the underlying mechanisms in the planning of duration in speech.PBL has been attested in various languages, which led to the suggestion that it is a universal tendency in speech production. However, it has also been found that PBL is implemented in language-specific ways, which suggests that it is a mechanism that is learnt by speakers, and raises the question whether and how it is linked to the word-prosodic system of a language. Furthermore, it has been attested that listeners employ PBL as a cue for the detection of a prosodic boundary, which may lead to structural disambiguation.Despite the existence of PBL in various languages and its attested relevance for perception, questions concerning the delineation of its domain of application, its temporal dynamics, and its cross-linguistic differences remain unanswered. As for the PBL domain, two major views on its specification are distinguished in the literature: structure-based and content-based views. Structure-based views assume that a specific linguistic constituent, such as the word rime, serves as the domain. The word rime is the stretch starting with the nuclear vowel of the last main stress syllable going up to the end of the prosodic phrase. Content-based views, on the other hand, assume structural variability of the PBL domain: the extent of the domain depends on properties of the final segment or syllable. As for the temporal dynamics within the PBL domain, prior studies found an increase of segmental lengthening towards the end of the prosodic phrase.The overall objective of the project is to test the aforementioned hypotheses across languages, thereby identifying the PBL domain of the respective language as well as the dynamics within the domain. The investigation focuses on three languages: German, French, and Tswana. These languages are chosen because they comprise typologically different prosodic systems that potentially influence the implementation of PBL. Detailed knowledge about the domain and dynamics in the production of PBL within and across languages is of paramount importance for investigating the processing of PBL in monolingual and bilingual speakers and for an understanding of the mechanisms for the computation of duration in speech.