The phonetics and phonology of vowel nasalization have been studied for some time, but the oral articulation of nasal vowels has been largely overlooked in both the phonetics and phonology literature. The acoustic consequences of nasalization obscure the oral configuration of a nasal vowel. Nevertheless, it is sometimes assumed that nasal vowels are produced with the same lingual and labial configurations as their oral equivalents. To better understand the synchronic and diachronic behavior of nasal vowels, the graduate student will examine the oral articulation of nasal vowels directly. The current project will combine articulatory, aerodynamic, and acoustic signals in order to observe lingual and labial articulation during the production of oral and nasal vowel congeners in two dialects of French: Quebec French (QF) and Northern Metropolitan French (NMF). The findings from this research are likely to have a significant impact on historical linguistics, specifically on sound change in Romance languages, as well as on the understanding of vowel nasalization more generally.
Pilot results from five speakers suggest that the speakers of NMF and QF produce nasal vowels shared by both systems in different ways which are suggestive of important characteristics of vowel nasality. When the oral articulation of a nasal vowel enhances the acoustic effect of nasalization, it is possible that the degree of naso-pharyngeal coupling can be reduced while still conveying nasality of the vowel. On the other hand, when oral articulation minimizes the acoustic effect of nasalization, an emergent nasal consonant may convey vowel nasality instead. The results suggest that the inter-dialectal differences in the nasal vowel systems of QF and NMF can be explained, in part, by the inherent characteristics of vowel nasality rather than by circumstances that are specific to the evolutionary development of French. This research will also support the scientific training of a graduate student.
Nasal vowels are characterized by some degree of coupling of the nasal cavity to the oral cavity, referred to as "nasalization". The articulations of the tongue and lips during the production of nasal vowels has remained of considerable interest to articulatory and acoustic phoneticians. Knowledge of these articulations has implications for both diachronic and synchronic studies of language. Acoustic analysis is often not sufficient to characterize these articulations; thus, observing the articulations directly is desirable. The current study has focused on the oral articulations of three nasal vowels in French, comparing these articulations to those of their oral vowel "counterparts". From 2011-2012 Carignan lived in Dijon, France, where he worked as an English instructor at the Université de Bourgogne. He was also an affiliated researcher at GIPSA-lab, Université Stendhal-Grenoble 3. Carignan travelled between Dijon and Grenoble, where funds from this NSF grant were used to record a total of 10 native French speakers for this project. Carignan collected data from simultaneously recorded acoustic, articulatory, and aerodynamic signals. During this time, Carignan worked closely with an undergraduate research assistant via video chat. Together with PI Ryan Shosted, Carignan created a Matlab graphical user interface (GUI) for segmentation of the data collected at GIPSA-lab. Carignan trained the assistant on how to use this tool and, in the process, she learned valuable information about how to work with acoustic signals as well as how both lingual and labial articulations can affect the acoustic signal. From 2012-2013, Carignan analyzed the data for this project, as well as completed and successfully defended his thesis (August 5, 2013). Carignan and Shosted also recorded three additional native French speakers using real-time MRI at the Beckman Institute at the University of Illinois, work which has resulted in transformative interdisciplinary collaboration. The articulatory results from this research suggest that lingual, labial, pharyngeal, and velic articulations of the French nasal vowels studied here can be interpreted as mechanisms of enhancement of the acoustic effects of nasalization. These results give strong support to the view that the goal of speech acts is acoustic, not articulatory, by showing that speakers use a variety of articulatory combinations in order to achieve a similar acoustic output. The "Ohalian" view of sound change (Ohala, 1975, 1981, 1996) considers change to be initiated by the listener via misperception of the acoustic effects of the articulatory source. The results from the current study support this view: the formant-frequency-related acoustic effects of nasalization were likely to have been misperceived by listeners as changes in oral articulation and, in turn, were produced as changes in oral articulation (presumably to match perception). This suggests that the synchronic vowel qualities observed here for the French nasal vowels are not merely coincidental, but have arisen diachronically due, in part, to misperception by the listener. Results from previous research on the oral articulation of French nasal vowels are somewhat limited, and also partially contradictory; without any further information, we are left to wonder if the contradictory results are due to differences in methodology, choice of speakers, analysis of the data, etc. However, the results from the current project suggest that these contradictory findings are due, rather, to inter-speaker variability in how the nasal vowels in French can be produced. In other words, different French speakers can articulate these vowels in different ways, even though these articulations result in similar-sounding productions. The findings from this project have important implications for second language French pronunciation teaching. For many of the sounds in French, it is helpful to teach students how to adjust their articulatory configuration to reach the acoustic target. With a more fine-grained understanding of the articulatory configurations used by native speakers for the vowels included in the current study, French instructors will be able to teach French learners more about the articulation of these sounds, which will help the students achieve a more native-like accent. Moreover, a greater understanding of the inter-speaker variability that exists among native speakers will allow French instructors to know which vowels should be taught with reference to articulatory targets, and which vowels should be taught with reference to acoustic targets. Carignan presented the results from this research at three major phonetics conferences and gave invited talks at University of California, Berkeley, and at North Carolina State University. The results from the data collected thanks to this NSF grant have resulted in a total of eight journal articles and conference proceedings which have been published (3), submitted for publication (3), or are currently in preparation for submission (2). Moreover, the acoustic, articulographic, and aerodynamic data and measures have been made publicly available on Carignan's professional website at the following address: www.christophercarignan.com/dissertation/database The public dissemination of these data and measures will, ultimately, provide an important benefit to the linguistics community in the form of a freely-available source of research.
