Researcher choice and overlooked variables: A "bottom-up" reanalysis of Villarreal (2018)

Dan Villarreal (Pitt Linguistics)

An immense body of sociolinguistic research has demonstrated how social processes (from macrosocial structures to momentary identity performance) are reflected and reproduced in speakers' production and perception of individual linguistic variables. The approach of isolating individual variables for investigation, despite being clearly fruitful, misses two critical facts about language variation in actual use. First is variable co-occurrence: variables do not exist in isolation. In between tokens of the individual variable under investigation are numerous socially meaningful variables (some in structurally related changes) that may mediate or change the social meanings of the studied variable. Second is researcher choice: the process by which we choose what to investigate may lead us to miss meaningful variation. The present study attempts to address these shortcomings using "bottom-up" methods to investigate Californian listeners' attitudes toward a multiplicity of co-occurring vowel variables, comparing these variables' influence on social meanings to previous research on California vowels.

To investigate this question, my co-author, James Grama, and I re-analyzed the results of my earlier matched-guise research on California English perceptions (Villarreal 2018). In that study, 97 Californian listeners rated excerpts from a cartoon-retell task (produced by 12 Californian speakers) on 12 attribute scales. Because stimuli were spontaneously produced (albeit all on the same topic), they all contained slightly different content and thus different vowel variables. Aside from the two vowels acoustically manipulated into guises (TRAP and GOOSE), all other vowel phonemes were left to vary naturally. The original analysis found that, despite substantial variance in attribute ratings overall, guise significantly affected three scales—suggesting stimuli contained additional socially meaningful variation that guise failed to capture.

To model this variation, we treated each stimulus as a "bag of features", mirroring "bag-ofwords" approaches to text corpora (Jurafsky & Martin 2022). These comprised vowel changes that are well-attested in California English (TRAP, DRESS, KIT, GOOSE, GOAT, LOT/THOUGHT), marginally attested (FOOT, STRUT), and largely unattested (FLEECE, FACE, PRICE). Vowels' F1 and F2 measurements were normalized and translated to discrete features using Atlas of North American English benchmarks (Labov, Ash & Boberg 2006). For each attribute, we used the Boruta algorithm (Kursa, Jankowski & Rudnicki 2010), via the Boruta package in R (Kursa & Rudnicki 2010), to determine which features influenced scale ratings.

This reanalysis revealed that vowels that are changing (or have changed) in California English did not necessarily impact social meanings more than those with marginal or no evidence of change. The most impactful variable, FOOT, is rarely investigated in California despite being structurally related to well-attested GOOSE and GOAT fronting; FLEECE, which is almost completely unattested as undergoing change, outranked several well-attested California English variables. In addition, despite the historical ordering of Low-Back-Merger Shift (Becker 2019) sound changes (TRAP then DRESS then KIT), TRAP and KIT impacted social meanings more than DRESS.

I argue that these findings reveal a need for greater attention to variable co-occurrence in modeling language variation. While variable co-occurrence is a known problem in sociolinguistics, actually accounting for it in practice is challenging. I suggest that bottom-up approaches like that described here can account for variable co-occurrence while mitigating the potential bias introduced by researcher choice.

Check back for more info. 

Where: Zoom https://uky.zoom.us/j/89119143772

When: Friday April 14, 2023 @ 1pm

Title: The Phonomaton: A new browser-based tool for implementing linguistic analyses

Abstract:

In this talk we present the Phonomaton, a freely available browser-based program that allows users to implement derivational analysis of phonological and morphological phenomena using standard notation. The program serves to ensure that complex analyses yield the expected results and provides a valuable method of exploring alternatives in real time. The Phonomaton handles all aspects of SPE style notation (with a modified SPE feature system) as well as autosegmental representations and can even make the full trip from underlying representation to sound wave. It is designed for both experts and as a teaching tool for students, and thus facilitates problem solving. Finally, it contains a library of phonological phenomena that serves as an interactive reference and theoretical sandbox. We hope to offer a hands-on presentation where everyone can explore the program's features as we present them

Studying grammatical variables in Spanish: New approaches and insights for sociolinguistics

Investigating grammatical (i.e., morphological, syntactic, discourse-pragmatic) variables poses some well-known challenges. To begin with, the approach pioneered for phonic variables (Labov 1963) must be modified in order to successfully apply variationist methods to the study of grammatical variables. As well, there is a long-standing assumption (see Schwenter 2011) that the social or stylistic significance of grammatical variables is harder to capture than it is for phonic variables, particularly because grammatical variants are typically less frequent in discourse. In this talk, I will draw on my research with Spanish speakers in distinct settings to show how the study of grammatical variables can lead to fruitful areas of discovery for sociolinguistics, similar to what has been shown for phonic variables. The range of phenomena to be covered include variation of the simple present and present progressive among classroom learners of Spanish, subject pronoun expression by bilingual Latino children in the U.S. South, and double possession in Peruvian Amazonian Spanish.

https://modernlanguages.olemiss.edu/stephen-fafulas/

Title: What’s in a name?: Research questions and Researcher questions about Creole New Orleanians

Abstract:

As part of a broader project examining language variation and change in post-Katrina Greater New Orleans, I report on preliminary results of linguistic analysis with a focus on the methodological conundrum of how to study ethnic identity in New Orleans. Ethnicity in New Orleans has always been fluid and complex, tracing back to colonial times when a tripartite social division of major ethnic groups existed – Free Europeans, Enslaved Africans, and Free People of Color (Campanella 2006). And indeed, New Orleans historically had a large population of Free People of Color compared to elsewhere in the American South. Over time, this population acquired intergenerational wealth and privilege that set them aside as an elite—and sometimes insular—ethno-cultural group which came to be known as Creoles (Brasseaux 2005). In current times, the term ‘Creole’ is contested. Some locals define Creoleness based on phenotypical features such as skin tone or hair texture, while others consider it a linguistic or cultural label, and still others refuse the legitimacy of this label entirely. Via quantitative and qualitative analysis, I shed light on a situation in which language change and social change appear to be progressing in tandem, posing questions about how to encode varied definitions – as well as varying individual stances towards those definitions – of ethnic identities in variationist research. In doing so, I build upon prior research on the linguistic expression of complex and multiracial identities (cf Holliday 2019; Bissell & Wolfram 2022) as well as work centered on methodological considerations in coding ethnicity (cf. Hall-Lew & Wong 2014; Nagy, Chociej, & Hoffman 2014).

https://liberalarts.vt.edu/departments-and-schools/department-of-englis…

Dr. Kelly Berkson will give our first colloquium talk of the semester regarding the Chin Languages Research project. Join us on Friday, September 16 from 1 to 2 PM in the Cornerstone Esports Theater. The talk will be broadcast live in the venue with the availability of realtime interaction among guests and the speaker. 

https://linguistics.indiana.edu/about/faculty/berkson-kelly.html

Abstract:

This talk provides an overview of the Chin Languages Research Project, a collaboration between speech scientists and members of Indiana’s large Burmese refugee community. Indianapolis, just an hour north of the IU Bloomington campus, is home to a community of more than 20,000 Burmese refugees. Many are originally from Chin State in western Myanmar and speak under- and un-documented Tibeto-Burman languages from the Kuki-Chin branch. Our best estimate is that upwards of 30 Kuki Chin languages are being spoken in Indianapolis. For the most widely spoken of these—Hakha Lai, also called Laiholh or Hakha Chin—syntactic and morphological work exists, but phonetic and phonological work is minimal. Other languages, such as Zophei and Lutuv, are completely undocumented: their names have been mentioned by previous scholars, but no prior linguistic work exists. Yet Zophei, spoken by 20,000 people worldwide (Eberhard, Simons, & Fennig 2019), is spoken by 4,000 people in Indiana, including more than a dozen students on the IU campus. Lutuv, with 15,000 speakers worldwide (Eberhard, Simons, & Fennig 2019), is spoken by perhaps 1,000 people in Indiana and several students on the IU campus.

Against this backdrop, at least two observations can be made: (1) IU linguists have an invaluable opportunity to engage in intensive fieldwork with many speakers of under-resourced languages just an hour’s drive from campus; (2) Chin community members regularly face communication challenges in both urgent and daily settings which speech scientists are uniquely positioned to help address. I describe how our team of IU speech scientists, student members of the Indianapolis Chin community, and community partners from churches, social organizations, and local businesses is working to engage in scholarship and address practical needs in tandem. I discuss several ongoing initiatives, including translation of Covid-19 information, documentation of traditional knowledge through in-language ethnographic interviews, development of Lutuv literacy materials, and an NSF-funded collaboration that seeks to develop health-related materials for linguistically underserved communities. I also touch on several of the typologically rare phenomena exhibited by Chin languages, in order to highlight the sheer volume of linguistic research that can be done right here on the ground in Indiana. Throughout, I highlight ways in which our undergraduate Chin students are the MVPs on our team—they are incorporated into the research process from the ground up, serving as a bridge between the university and the community and helping to determine new research and service directions.

Title: Speaker Identity and Syntactic Expectations

Abstract:

A lot of research has been devoted to understanding how speakers of multiple languages, multilinguals, use their knowledge of multiple languages to interpret language in context. But less is known about how speakers who know different varieties of a single language (‘dialects’) accomplish this. In this current project, the researchers investigate African American Language (AAL), a variety of English spoken primarily, but not exclusively, by African Americans. The current work plans to investigate how speakers of AAL understand sentences that use linguistic features that are present in AAL but not present in Mainstream American English (MAE), the ‘standard’ dialect spoken across the US. AAL and MAE overlap significantly, but there are specific grammatical constructions where the two diverge. This project investigates when and how AAL speakers use these variety-specific constructions in different contexts. Research into how AAL speakers use context to do this will shed light on how language processing proceeds in more naturalistic contexts, and will also shed light on the usage of a variety of English spoken by a significant portion of the US population.

https://people.umass.edu/anneal/

https://linguistics.as.uky.edu/users/kbmc229

This event has been deferred to a later date. More information to come.