Evaluation of Vowel Changes in Global Englishes
(Purnell with Raimy and Balogun)
[Sponsor: University of Wisconsin-Madison Graduate School]
This project examines the extent of language change in English as evidenced by a comparison of the phonological features yeilding American English and Global English vowel variation. Modern linguists explicitly and implicitly refer to language change in vowels as if change has a known start and end point, repeating a 19th-century error. English varieties are particularly interesting as there have been widespread reports for the last 50+ years of vowels changing their articulatory position. This project is cataloging English vowels across the world. While significant change may occur among vowels (consider the spelling and pronunciation mismatches of bate and beet), contemporary English vowels strengthen the idea that modern English vowels largely vary in “flavor” but not distinctiveness. Contrary to the claim that rampant change occurs in English vowels, the argument is that a robust abstract English grammar constrains articulation, resulting in fewer substantive changes than expected. The import of observed superficial vowel changes is to reveal speakers’ group affiliations, not grammatical change.
Wisconsin Englishes Project
(Raimy, Salmons and Purnell with Annear, Stickle and Williams)
Past students: Dilara Tepeli, Jennifer Mercer; Emily Claire, Mary Simonsen, Alicia Groh; Taylor Patton, Kassie Weber, Danica Rockney, Casey Olsen, Zach Degregorio, Chris Spoke, Allen More
[Sponsor: University of Wisconsin-Madison Graduate School; past sponsor: Wisconsin Humanities Council]
The Wisconsin Englishes Project collects extensive data about how English is spoken in Wisconsin to address questions about geographic dialectal variation, language change and generational & gender differences in Upper Midwestern speech. This research is driven by the collection of primary linguistic data through audio recordings and our immediate pressing need is to create a coherent corpus structure and archive of our present data in order to support analysis and future data collection. Due to increased interest in participation in this project from both students and local communities, we must increase our ability to process and store data.
Intoxicated Speech and Speech Variability
(Purnell with Williams)
Past students: Ryan Hanke
Intoxication provides a temporary ataxic setting for examining dialect variation. The data collected used a paradigm where each speaker’s sober speech can be compared to their speech on the ascending arm of intoxication, the peak intoxicated state (BAL = 0.08), and on the descending arm of intoxication. This data set has been described in Hanke & Purnell 2006, Purnell 2010, and Williams & Purnell 2010, 2011.
Neuromagnetic Correlates of American Dialect Perception
(Purnell with Abrams, in collaboration with Bill Idsardi, Univ MD)
[Sponsor: National Science Foundation Award #1125295]
Humans are remarkably adept at recognizing other human voices, not only for the messages they communicate, but also for social and personal information about the person they are talking with. For example, when we answer the telephone we can easily recognize whether it is someone we know (Mom) or someone we don’t (a telemarketer). Listeners also make rapid judgments about a speaker’s age, gender and where they are from. This project investigates human behavioral and early brain responses to the recorded spontaneous speech of speakers with the same or different dialects and genders using a neuromagnetic brain imaging technology, magnetoencephalography (MEG). Investigators will examine the rating, identification and discrimination of American English dialect stimuli by listeners while their brain activity is passively recorded. These brain activity data will be correlated with dialect differences and behavioral measures. While there has been considerable work on early neural correlates of prosody, very little work has examined the neural basis of within-and across-category perception of speech characteristics that have real-life importance in areas of prejudice and discrimination. Researchers investigating social aspects of language have had limited access to brain imaging technologies, and consequently this collaborative research will contribute to speech perception, neuromagnetic brain research and sociolinguistics, and will foster new connections between these fields. Moreover, this research has other important social and legal implications because its results will bear directly on how courts can and will interpret automatic neural processes and higher order cognitive processing resulting in discrimination and prejudice.
Geographically-Informed Linguistic Analysis: Change as local knowledge spillover effect
(Livengood, Purnell, Raimy and Salmons)
We know that geography matters to language variation and change, but sociolinguists, linguistic geographers and historical linguists lag in their knowledge—despite calls and promises from the mid-1990s (e.g., Lee & Kretzschmar 1993)—of how to quantify this relationship utilizing the available, rich geographic information system (GIS) technology. The basic question of how linguistic features diffuse through space has been central to linguistics since the field’s beginning, but research has not yet fully utilized tools currently available to geographers and cartographers. Linguists have generally not explored how digital geospatial technologies, such as GIS, can help them to challenge assumptions, test theories, and generate hypotheses about space and language change. (This is changing, as shown by the NSF-funded workshop “Language in Space” and Auer & Schmidt 2010.)
The proposed research refines and synergizes quantitative and visualization methods for analyzing historical and contemporary linguistic data in space. Drawing data from the Upper Midwest, the principal investigators will:
- Identify georeferenced factors contributing to language variation,
- Advance ways to quantify the complex relationship between regional geography, society and language through a local knowledge spillover model, and
- Create novel visualizations of this complex relationship.
Overall, this project will advance our understanding of proximate pressures on language variation and change, using a regional economic model to investigate quantitative geolinguistics modeling. In particular, we adapt the local knowledge spillover model from economics in explaining how linguistic features spread geographically. We present our model below and address conceptual issues in visualizing linguistic variables on maps. After a word on the state of linguistic mapping, we sketch the pilot work leading to this proposal. The body of the proposal lays out how the program works, illustrated from work to date. We conclude with a work plan, detailing the analyses to be undertaken.
Final Devoicing and Acoustic Energy
(Purnell and Salmons with Rodgers)
[Sponsor: National Science Foundation Award #1052819]
This dissertation research seeks to better understand how listening is informed by the act of moving our vocal apparatus as well as the acoustic and perceptual implications of crosslinguistic articulatory differences. A number of factors are known to contribute to our ability to make use of the acoustic signal such as articulatory strength, trading relations, and acoustic energy in contrasts. The fundamental theoretical question addressed here is whether articulation per se has direct implications for abstract gestures as distinctive features. The project builds on earlier work investigating the role of energy characteristics. Acoustic data will be collected for a range of native speakers of such typologically distinct languages as English, Dutch, German and French.
Tongue position and air pressure data for plosive sounds at the beginning and end of a syllable will be synchronized with the acoustic recordings. A comprehensive data set for these languages will be used to, first, establish percepts and the boundary conditions within which a perception study will be conducted, and, second, increase our understanding of the articulation-acoustic relation across the speech chain. The relative spectral energy levels in lower harmonics will be measured and evaluated, along with their changing characteristics as a speaker makes a vowel after or before one of these consonants. These spectral energy change characteristics have been found to be more correlated with voicing contrasts than the traditional measures of voice onset time, vowel length, frequency changes, etc. across contexts and across language types. Information gained from the acoustic and articulatory data collection will be used to generate tokens to be tested in a subsequent set of perception experiments. The work has implications for automatic speech recognition technology.
Articulation Variation
(Purnell with Rodgers, Williams and Schrank)
Past students: Ryan Hanke; Lindsay Ginsberg
[Past sponsor: University of Wisconsin-Madison Graduate School]
Analyze variation in articulation with x-ray microbeam and ultrasound data. X-ray data has been collected for 27 female speakers in three areas of the upper midwest: the greater Milwaukee area; the greater Minneapolis area; and the region between, especially around Eau Clare. Analysis of the data is underway and has been partially reported in Purnell 2008a,b and Rodgers, Purnell & Salmons 2008. Currently, we are conducting ultrasound experiments to examine regional changes in three areas: a) different tongue shapes in diphthongization in /o/ for word classes of BOAT, BODE and BEAU; b) tongue height in word classes of BITE, BIDE and BUY; and c) changes in tongue shape over different speech styles.