Doctoral Dissertation Research: Combining quantitative analysis and corpus-based methodologies in the documentation of an endangered language

大多数语音语言学理论，以及为之提供信息的跨语言研究，都是基于对少数几种研究得很好的语言的研究。人们对使用者较少的语言的语音系统（音系）知之甚少。此外，对未充分研究的语言，特别是语音系统的描述，很容易受到语言学家母语的影响。非本族语听者的偏见往往会导致相互矛盾的描述，因此不适合进行跨语言比较。这个项目研究了一种未被充分研究和濒危语言的语音系统。先前对其语音系统的描述很少，但现有的文献表明，这种语言包括一些具有特殊理论或类型学意义的特征。例如，该语言使用小舌音，声门化响音和无舌侧摩擦音。这些音（或音组）中的每一个在跨语言中都是不常见的。它们在这种特殊语言中的出现也是出乎意料的;与其他语言的语音系统相比，这种系统中共同出现的语音和特征是不寻常的。此外，这项研究具有区域性意义，它将扩大语言学家对接触驱动的语言变化的理解，并揭示该地区各群体之间长期接触的历史。本研究结合了传统的方法学，包括记录语音的音频视频语料库，书面转录，并控制启发定量实验方法。该项目的实验部分针对理论和类型学方面的问题。例如，说话者使用什么声学线索来产生和感知压力？长元音和短元音是分开的吗？“吱吱作响的声音”是否应该作为单个片段、单词或短语的特征进行分析？控制实验的结果将与传统的，语料库驱动的结论相结合，以开发一个以前未描述的系统的全面的语音描述。精确的定量分析，结合全面的描述和公开的语言数据语料库，将使这些结果有助于更广泛的问题，涉及声音模式，语言类型学，语言接触和历时变化。该奖项反映了NSF的法定使命，并已被认为是值得通过使用基金会的智力价值和更广泛的影响审查标准进行评估的支持。

Most phonological linguistic theory, and the cross-linguistic studies which inform it, is based on studies of a few, well-studied languages. Much less is known about the sound systems (phonology) of languages with fewer speakers. Furthermore, descriptions of understudied languages, particularly of sound systems, are easily biased by linguists' native languages. Non-native listener biases often lead to contradictory descriptions which are therefore unsuitable for cross-linguistic comparisons. This project examines the phonological system of an understudied and endangered language. Previous description of its phonological system is sparse, but existing documentation indicates that this language includes some features which are of special theoretical or typological interest. For example, the language uses uvular sounds, glottalized sonorants, and a voiceless lateral fricative. Each of these sounds (or sets of sounds) is cross-linguistically uncommon. Their appearance in this particular language is also unexpected; in comparison to other languages' sound systems, the set of sounds and features which co-occur in this system is unusual. In addition, the study is of regional interest; it will expand linguists' understanding of contact-driven language change and shed light on long-term histories of contact among groups in the region. This study combines traditional methodologies—including recording an audio-video corpus of speech, written transcription, and controlled elicitation—with quantitative experimental methodologies. The experimental component of the project targets questions of theoretical and typological interest. For example, what acoustic cues to speakers use to produce and perceive stress? Are long vowels separate segments from shorter vowels? Should 'creaky voice' be analyzed as a feature of individual segments, words, or phrases? Results from controlled experiments will be integrated with traditional, corpus-driven conclusions to develop a comprehensive phonological description of a previously undescribed system. Precise, quantitative analysis, integrated with a holistic description and publicly-available corpus of language data, will allow these results to contribute to broader questions related to sound patterns, language typology, language contact, and diachronic change.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.