Bibliographic information and abstracts for publications. Please consult the original periodical or book for the complete text.

Van Hofwegen, Janneke and Walt Wolfram. (2017). On the utility of composite indices in longitudinal language study. In Suzanne Evans Wagner and Isabelle Buchstaller (eds.), Using Panel Data in the Sociolinguistic Study of Variation and Change Routledge Studies in Language Change Series. London: Routledge.

Although most longitudinal studies focus on a single structural variable or a limited subset of variables, the potential of composite indices in the assessment of language varieties is relatively unexplored in sociolinguistics. For valid reasons, linguists have been cautious about assessing dialect use by means of an overall dialect measure, particularly with respect to a variety such as African American Language (AAL). Notwithstanding the limitations, there appears to be some heuristic, analytical, and descriptive justification for objectifying dialect use in terms of an overall score. Scalar indices of language proficiency are commonly used in allied fields where it is expedient to assess speakers in terms of overall language proficiency. Accordingly, we propose that for some analytical and descriptive purposes, cautiously constructed unitary indices may provide valid information about the relative use of a designated language variety, including how its use correlates with various social and educational factors.

This paper considers the utility of composite dialect indices in longitudinal studies by comparing different approaches to the assessment of AAL development, using data from a unique, longitudinal language study of 70 African American subjects at 7 temporal points during their early lifespan (from 48 months through age 20). Notwithstanding the impressive assortment of descriptive and applied research conducted to date, many questions remain about the use of AAL over the lifespan. When do speakers optimize their use of AAL? Are there different trajectories of AAL use during the lifespan? What primary social, psychological, and educational factors correlate with AAL use at different stages in the lifespan and how do different linguistic structures participate in change through the life cycle? Answers to these questions may be addressed most effectively by using composite measures of dialect density rather than limiting investigation to singular structural traits.

D'Onofrio, Annette, Penelope Eckert, Robert J. Podesva, Teresa Pratt, and Janneke Van Hofwegen (2017). Low vowel variation in California's Central Valley. In Betsy Evans, Valerie Fridland, Tyler Kendall, and Alicia Wassink (eds.), Speech of the West, Vol. 1. Publication of the American Dialect Society. Durham, NC: Duke UP.

Most work on the California Vowel Shift (CVS) has focused on coastal cities. This study, by contrast, investigates the CVS in three non-urban inland communities: Bakersfield, Merced, and Redding. For 54 speakers, we analyze the low vowels TRAP, LOT, and THOUGHT. We show that TRAP is backing, and LOT and THOUGHT are becoming more merged in all three sites, with male speakers and speakers from Merced, at the center of the Central Valley, remaining the most conservative. These processes reconfigure the vowel space from a trapezoidal to a triangular shape. Contrary to findings from California's coastal cities, we find LOT and THOUGHT merging via LOT raising (not THOUGHT lowering) in two of the three sites. Findings further reveal continued raising/backing of the merged LOT/THOUGHT vowel over time. We suggest that while all three non-urban inland communities are moving toward a California-shifted system, the process by which this is achieved may differ by location.

Van Hofwegen, Janneke (2015). Dyadic analysis: Factors affecting African American English usage and accommodation in adolescent peer dyads. Language and Communication 48: 28-45

This study analyzes relative accommodation of African American English (AAE) usage in peer interactions for 201 African American adolescent dyads at Grades 6, 8, and 10. A Dialect Density Measure (DDM: Van Hofwegen and Wolfram, 2010) assessed speakers' overall AAE use. Then, dyadic analysis revealed that variation between dyads is greater than variation within dyads. That is, children are significantly similar (accommodate) to their peers at all three time points, regardless of whether their peers were self-selected or assigned. However, girls' and boys' accommodation varied by how well they knew their peers and in terms of ethnically- or formality-salient features.

Van Hofwegen, Janneke (2015). The development of African American English through childhood and adolescence. In Sonja Lanehart (ed.), Oxford Handbook of African American Language (pp. 454-474). Oxford UP.

After more than a half-century of rigorous research on many diverse facets of African American English (AAE), many questions about the use of AAE over the lifespan have been posed. What are the primary influences on a child’s acquisition of AAE? At which point in the lifespan do speakers optimize or minimize their use of AAE? Which elements of language (morphosyntactic, phonetic, pragmatic) and which variables are most salient in the acquisition and use of AAE in childhood? How do AAE speakers negotiate the use of their home dialect in the face of a mainstream American English-speaking majority? This chapter reviews the spate of research that has attempted to answer these questions. It begins with earlier theories about AAE acquisition and use (e.g., those by Stewart, 1965; 1968; Dillard, 1972; and Labov, 1965), and studies examining the relationship between general- and dialect-specific- language acquisition (e.g., Steffenson 1974; Kovac 1980; Kovac & Adamson 1981; Stockman & Vaughn-Cooke 1982, 1989) and change across the early lifespan to more recent studies tracing the relationship between dialect use and academic performance and social/family measures (e.g., Craig & Washington, 2006; Washington & Craig, 2002). Specifically, the chapter highlights a wide-ranging longitudinal study reported elsewhere (e.g., Van Hofwegen & Wolfram, 2010; Van Hofwegen, forthcoming; Renn & Terry, 2009) in which a cohort of ~80 African American children, followed from birth to post-high school, were examined across a range of linguistic and social factors. From this series of studies, a picture has emerged of an AAE speech community that utilizes a roller coaster of AAE use (but with rather stable vocalic structures), peaking and trough-ing at consistently different points in the life span; a community that gradually learns patterns of style-shifting with mainstream American English (to varying degrees); with important impacts of family, educational, and social factors at varying points along the way.

Podesva, Robert J., Annette D'Onofrio, Janneke Van Hofwegen, and Seung Kyung Kim. (2015). Country ideology and the California Vowel Shift. Language Variation and Change 48: 28-45.

Addressing the dearth of variation research in nonurban, noncoastal regions of California, this study examines the extent to which speakers in Redding, an inland community just north of the Central Valley, participate in the California Vowel Shift (CVS). We acoustically analyze the fronting of the back vowels BOOT and BOAT, the raising of BAN and backing of BAT, and the merger of BOT and BOUGHT, in sociolinguistic interviews with 30 white lifelong residents. Results reveal a change in apparent time for all analyzed variables, indicating the CVS's progression through the community, though not as robust as in urban, coastal areas. Additionally, we provide evidence that shifting patterns for different vowels are structured by the ideological divide between town and country. Thus, as the CVS spreads through Redding, speakers utilize particular features of the shift differently, negotiating identities relevant in California's nonurban locales.

Podesva, Robert J. and Janneke Van Hofwegen. (2015). /s/exuality in small-town California: Gender normativity and the acoustic realization of /s/. In Erez Levon and Ronald Mendes (eds.), Language, Sexuality, and Power: Studies in Intersectional Sociolinguistics. Oxford UP.

Previous sociophonetic research has established that, in some communities, the realization of /s/ may correlate with a speaker's gender (Fuchs & Toda, 2010; Hazenberg, 2012; Levon & Holmes-Elliott, 2013; Pharao, Maegaard, M°ller, & Kristiansen, 2014; Strand, 1999; Stuart-Smith, Timmins, & Tweedie, 2007; Stuart- Smith, 2007; Zimman, 2012), sexuality (Campbell-Kibler, 2011; Hazenberg, 2012; Levon, 2007; Linville, 1998; Munson, McDonald, DeBoe, & White, 2006; Pharao et al., 2014; Zimman, 2013), or rurality (Campbell-Kibler, 2011). While past research has tended to zero in on one of these dimensions, this paper examines how ideologies about the country, gender, sexuality, and their inter-relations play out in the same community. We find it fruitful to consider these connections because communities where rurality constitutes an important axis of social distinction are likely to subscribe to traditional norms regarding gender and sexuality. This paper examines the acoustic realization of /s/ among residents of Shasta County, a community just north of California's San Joaquin Valley. Drawing on a series of comparisons within the local community, we show that country-identified speakers exhibit different patterns from speakers who orient to the town, and also that members of a tight- knit LGBT community exhibit somewhat different patterns from community members who do not identify as sexual minorities. We further contrast our findings with patterns previously observed for speakers in urban areas and find that in spite of significant differences between straight and LGBT speakers in Redding, both groups of speakers produce more normatively gendered patterns than their counterparts in cities. Our primary claim is that the realization of /s/ is heavily constrained by dominant local ideologies about gender and sexuality. Specifically the sociopolitical conservativism characteristic of this country community polarizes gender distinction and pressures sexual minorities to adhere to normative gender patterns.

Podesva, Robert J. and Janneke Van Hofwegen. (2014). How conservatism and normative gender constrain variation in inland California: The case of /s/. University of Pennsylvania Working Papers in Linguistics 20.2: 129-137.

In this paper, we analyze the realization of /s/ in a non-urban inland community in far Northern California. Residents of this community, especially those who identify as LGBT, harbor complex feelings about and relationships with their neighbors, fellow Californians to the South, and to the land. Our primary claim is that these complex local ideologies significantly shape the realization of /s/ in this community, primarily ideologies about how gender and sexuality should be expressed. Specifically, the sociopolitical conservatism characterizing this non-urban community effectively polarizes gender distinctions and heavily pressures sexual minorities to adhere to normative gen- der patterns.

Thomas, Erik and Janneke Van Hofwegen. (2014). Consonantal variation in the English of a Spanish-Substrate Community. In Proceedings in Bamberg Series, Peter Lang Editions, Methods in Dialectology XIV.

One important means by which new dialects can be formed is from linguistic substratal influence. However, scholarly attempts to watch this process in action are mostly recent phenomena and they often cover only one or two generations. Observation of contact communities of longer standing is needed. We have analyzed such a situation for several years. Our community lies in southern Texas, 54 miles (87 km) southwest of San Antonio, composed of two long-standing Anglo and Mexican American populations. In this paper we are looking at variation of the consonants /l/, /­/, /r/, as well as voice-onset time of stop,s in apparent time in the Mexican American community as they compare with their Anglo neighbors. Our oldest generation, born before 1940, was Spanish-dominant, but the youngest generation, those born since 1980, is English-dominant, often with only a passive knowledge of Spanish. Our analysis reveals that of these consonantal features, two of them, /l/ and /­/, display ethnically-salient variation patterns. The other two, /r/ and VOT, are not ethnically relevant in this community, but are most associated with educated speech. It may be that in this community rhoticity, bunched /r/s, and positive VOTs are associated with “standard” unmarked, mainstream speech. These results shed light on what variables may or may not be ethnically distinctive to Mexican American English, and which features are regionally salient as well.

Van Hofwegen, Janneke. (2011). The evolution of /l/ across three generations of African Americans. Language Variation and Change 22: 373-396.

In the wake of numerous analyses of vowels in African American English (AAE), this study examines acoustically the phonetic production of a consonant—the word-initial lateral /l/— across several generations of AAE speakers from an enclave community in central North Carolina. Traditionally, /l/ has been dichotomized into alveolar (‘light’) and velarized or alveo- velar (‘dark’) allophones (Giles & Moll, 1975); recent acoustic and articulatory analyses, however, have shown that light/dark /l/ variants are gradient according to context (Sproat & Fujimura, 1993). The results of the study, analyzing speakers from a long-standing African American community, show that /l/ is darker in younger AAE speakers than in older ones, independent of phonetic context. Comparisons with ex-slaves suggest that a light variant of /l/ may be a substrate feature of AAE that has changed in recent decades. Additional comparisons with regional European Americans suggest that the darkening may be due to convergence with majority American English dialects.

Van Hofwegen, Janneke and Walt Wolfram. (2010). Coming of age in African American English: A longitudinal study. Journal of Sociolinguistics 14: 427-455.

This study examines trajectories of development in the use of African American English (AAE) for 32 speakers through the first 17 years of their lives based on a unique, longitudinal database. Temporal data points in the analysis include 48 months, Grade 1 (about age 6), Grade 4 (about age 9), Grade 6 (about age 11), Grade 8 (about age 13), and Grade 10 (about age 15). Complementary methods of analysis for assessing AAE include a token-based Dialect Density Measure (DDM), a type-based vernacular diversity index, and frequency-based variation analysis. The study reveals different trajectories and peak periods for the use of AAE, including a ‘roller coaster’ and a curvilinear trajectory; at the same time, there is a common dip among speakers in the overall use of vernacular AAE from Grade 1 through Grade 4. Examination of a selective set of demographic and self-regard measures shows no significant differences for gender, school racial density, racial peer contacts, and measures of Afro-centricity, but does show a significant correlation between mothers’ and child use of AAE as well as age/grade.

Van Hofwegen, Janneke. (2009). Cross-generational change in /l/ in Chicano English. English World-wide 30: 302-325.

The acoustic study of consonants has lagged considerably behind that of vowels. While a robust literature exists about vowel shifting, vowel quality, and the sociolinguistic significance of vowels, comparable literature is lacking for the acoustic quality of liquids. This study seeks to supplement the acoustic studies of vowels by analyzing characteristics of the liquid /l/ in its word-initial context. Traditionally, phonologists have subdivided /l/ into two allophones: dark and light, although current analysis has characterized these distinctions as gradient, not discrete. Word-initial /l/ is thought to be the canonically lightest variant of the phoneme, but cross- dialectal research has shown great acoustic variance in its phonetic realization. This case study aims to trace the phoneme through three generations of Chicano English speakers from south Texas, and to draw conclusions about how its variation among speakers and generations can shed light on other sociolinguistic phenomena, such as the persistence of substrate features from Spanish (with its characteristically light /l/s) or assimilation into mainstream American English dialects (with their characteristically dark /l/s). The study shows that there is indeed significant shift in the lightness of /l/—independent of phonetic context—across the generations of speakers under examination. This result supports other studies that show notable assimilation with Anglo English varieties in earlier generations, but robust use of ethnically-marked phonological features among recent generations.