Lothian Lockdown: The Lothian Diary Project I direct a team of multi-disciplinary social scientists studying how the COVID-19 stay-at-home measures have affected life in Edinburgh and the Lothians. We collected 195 self-recorded video/audio diaries and survey answers from a wide range of local residents. We have submitted an Executive Summary and Parliamentary Briefing for Scottish Parliament and we have created an oral history that is now housed with Museums and Galleries, Edinburgh. We are currently presenting talks and papers describing the corpus, describing methods, and showing some preliminary sociolinguistic analysis. The longer term outcomes will be research projects on a range of topics in linguistics, anthropology, political science, and psychology. For more about the project, visit our website: https://lothianlockdown.org/
Tourism & Language Commodification From 2014 to 2019 I was studying the commodification of Scots and Scottish English accents in Edinburgh, with a focus on the tourism industry. A brief summary of the project appears as a Case Study in the online textbook, Tourism Geography, 3rd Ed. (Williams & Lew 2014). In the following video I discuss the findings of one of the earlier stages of the project. I've written more generally about language and tourism for the Wiley Blackwell Companion to Tourism, with my father, Alan A. Lew. In 2021 was interviewed about this work by Bilingualism Matters, here.
Social Class Mobility in Edinburgh My former undergraduate (honours) student, Victoria Dickson, worked with me on phonetic variation and social class in Edinburgh from the angle of social class mobility, looking in particular at older people who were born into working class families who experienced upward mobility throughout their lifetimes. Our co-authored work building on her Honours dissertation was published in the Journal of English Linguistics. Our work is part of a larger collaborative project at the University of Edinburgh that's about language in Edinburgh, called Edinburgh Speaks.
Phonetic Variation & Politics / Politicians
Bre/ks/it or Bre/gz/it? In the summer of 2019, two MSc Applied Linguistics students worked under the supervision of me and Prof. Graeme Trousdale on the production, perception, and metadiscourse around the variable pronunciation of the portmanteau, Brexit. A write up of the results and further analysis have now been published and are available here.
The 'Iraq' Vowel I collaborated with Drs. Rebecca L. Starr and Elizabeth Coppock on a project analyzing the members of US House of Representatives in 2007 with respect to how they pronounced the word Iraq. We found that Democrats were more likely than Republicans to use the vowel /ɑ/ (as in cot) for the 'a' in Iraq rather than the more commonly used vowel, /æ/ (as in cat), even when controlling for speaker age, ethnicity, geographical region, and regional accent. 15% of the Representatives vary between both vowels in the same speech, and when we examined that intraspeaker variation we found additional suggestive evidence that the vowel is an index of political identity. I've discussed this work on the BBC World Service podcast, Crowdscience: http://www.bbc.co.uk/programmes/p04xtpnh.
Prosody & Negation My work in sociolinguistics began with my collaboration with Dr. Malcah Yaeger-Dror, and other colleagues, looking at the prosody and contraction of 'not' negation. My contribution to the project was primarily with respect to the analysis of adversarial speech as realised in the US presidential debates from Kennedy/Nixon to Clinton/Perot/Bush. We found that genre and register predict the realisation of 'not' negation, both with respect to whether it is full form (not) versus contracted (n't) and with respect to its prosodic contour.
Ethnicity & Sound Change in San Francisco
For my PhD dissertation and subsequent related projects I analyzed aspects of the California Vowel Shift in San Francisco, California. I interviewed about 100 San Franciscans of various social backgrounds, most of them lifelong residents of the large western neighborhood known as the Sunset District. I have focused specifically on the transition in the second half of the 20th century from the neighborhood's identity as an "Irish Catholic parish" to a "New Chinatown", and the sociolinguistic consequences of this of change. Some of my work focuses specifically on the Chinese Americans in the sample, such as a paper with Dr. Rebecca L. Starr on ethnic identity beyond the 2nd generation, and a paper with Dr. Amy Wong comparing vowel production among Chinese Americans in New York City with those in San Francisco. This is an early overview article of ethnicity and variation in SF English, and this page links to the most recent publication from the project, in collaboration with Dr. Amanda Cardoso, Yova Kementchedjhieva, and Ruaridh Purse. Some of my findings were discussed in this episode of the Outside Lands podcast, and this episode of KQED's Bay Curious podcast:
Arizona Between my undergraduate and postgraduate studies I researched back vowel production in Flagstaff, Arizona. I took the results to suggest that Northern Arizona English might be characterized as having a foundation of accent variation from the Southern states, overlaid with Californian influences in more recent years. After a dozen years' break, I addressed this question again in a forthcoming paper with collaborators Mirjam Eiswirth, Mary-Caitlyn Valentinsson and William Cotter, which shows that 'Californian' features in the front vowel system may have been evidenced in Arizona much longer than previously thought, perhaps just as long in Arizona as in California. This project also takes advantage of the generous assistance of the StoryCorps oral history project.
Texoma My work in Arizona fostered a curiosity in the linguistic correlates of identifying as 'country'. This interest led to a collaboration with Dr. Nola Stephens, who hails from the border of Texas and Oklahoma, also known as 'Texoma'. Based on interviews with Texomans, we argued that Country Talk is a register comprising links between linguistic features and cultural meanings that contrast in discourse with notions like rurality, Texan identity, and Southernness.
Homeschoolers Dr. Nola Stephens, Prof. Vickie Shamp Ellis, and I recently published a paper looking at quotative use (specifically, the use of BE LIKE to introduce a quotation) among students at a small Southern Christian college. While the don't specifically focus on region as a factor, most of the speakers in the study identified as Southern. In this study we compared those who had been homeschooled in high school with those who had gone to private or public high schools. We also compared those who said they chose their college for academic reasons with those who said they chose their college for social reasons. We found no differences in quotative use according to schooling type, but we did find lower rates of BE LIKE use among students who had chosen their college for academic reasons. We also see some interesting differences between students who had been homeschooled, depending on what type of homeschool they had.
Quantifying Vowel Overlap "Vowel mergers are some of the most well-studied sound change phenomena. Yet the methods for assessing and characterizing an individual speaker's participation in an ongoing merger (or split) vary widely, especially among researchers analyzing naturalistic corpora. We consider four methodological approaches to representing and assessing vowel difference: Euclidean distances, mixed effects regression modeling (Nycz 2013), the Pillai-Bartlett trace (Hay, Warren, & Drager 2006), and the spectral overlap assessment metric (Wassink 2006). We discuss the strengths and weaknesses of each method and compare them by applying all of them to three different data sets, each of which contains low vowel data from speakers whose status with respect to a vowel contrast may not be clear-cut." (Nycz & Hall-Lew 2013)
L-vocalisation The vocalisation of syllable-final /l/ (where the word 'coal' sounds like 'coe') is a relatively common feature across varieties of English, but notoriously difficult to measure acoustically. Dr. Sonya Fix and I studied how consistent linguists are in making auditory judgments about the quality of coda-/l/ productions in naturalistic speech corpora. We found that linguists are incredibly consistent with tokens that are definitely consonantal and those that are definitely vocalized, but that there are a many occurrences that lie somewhere in the middle, and these are not so consistently rated (but are sometimes the most interesting ones, indicating a middle stage of a change in progress, for example). I got interested in vocalisation because of some of the Asian Americans I interviewed in San Francisco. Preliminary results show vocalisation to be strongest and more frequent in the English of speakers who, as children, were proficient in a language without coda-/l/. It also appears in common words that index aspects of local Asian American culture, like the name of a local high school ('Lowell' often sounds like 'low', even among students who are non-Asian Americans).
Speech Elicitation Tasks A co-authored paper with my former PhD student, Dr. Zac Boyd, recently appeared in Linguistic Vanguard. It presents a phonetic analysis of variation according to speech elicitation task, bridging methods typical of sociolinguistics, phonetics, and lab phonology. With other colleagues at Edinburgh we initially analyzed vowel variation within a single speaker with respect to three styles: speech from sociolinguistic interviews, speech from "lab tasks" (e.g., a Map Task or a Spot-the-Difference Task), and speech from "self-recordings" (i.e., speech recorded without a researcher present). Based on those results, Zac and I collected parallel data from three additional, demographically similar speakers, considering additional styles such as picture book description, silent film narration, semantic differential questions, reading passages, and word lists. We are also considering under-studied factors such as the "same" task in monologue (e.g., a solo Map task) versus dialogue, and a comparison between portions of interview speech that are metalinguistic versus those that are not. Another recent publication is this paper in the Penn Working Papers in Linguistics, about sibilants.