Language users'choices are shaped both by the linguistic forms used around them and by the socioindexical meanings of those forms. Teasing out these two factors is not always easy, but both offer important avenues for understanding language change at the community level. This study explores the vowel patterns of zz men living in rural Defiance County, Ohio, an understudied area nearthe boundary between the Midland and Inland North dialect regions of American English. The authors briefly document vowel patterns in the area and investigate whether travel patterns predict /ae/-raising and fronting, or /a/-, /u/-, and /o/- fronting and whether "country" identity predicts monophthongization of /u/, /o/, and/au/. Participants'vowel systems were neitherclearly Inland North norclearly Midland, showing elements of both systems. Country identity received moderate support as a factor, predicting /u/ and /au/ diphthong length, while travel patterns were less well supported, correlating only with /a/-fronting. Results support previous classifications ofthe area as a mixed ortransitional zone and offer minimal support for pure exposure and moderate support for socioindexical identity as influences on language change.
After137years and more than 30 applications to the United States government, the Lumbee American Indians of North Carolina were finallyawarded full federal recognition bythe government in an act signed by President Trump on December18, 2025. The Lumbee, with approximately 55,000 tribal members, are the largest group of Native Americans east of the Mississippi River and one of the earliest to interact with English-speaking Europeans in the sixteenth century. They lost their indigenous, ancestral language early in the process of European colonialization and have lived in a triethnic situation with African Americans and European Americans in southeastern North Carolina fora couple ofcenturies now, blurring their status as Native Americans. The author considers the contentions about their status, including their long-term roots, their persistent ethnic identity, and their peculiarsociolinguistic language situation in which they have crafted a unique dialect of English to distinguish community members from European Americans and especially Africa Americans as they lost theirancestral language early in their interaction with the European colonists.
Social media data and computational tools have become increasingly powerful alternatives to traditional methods ofdata collection and annotation forsociolinguistic research. While not without their own drawbacks, these resources can help address critical problems faced by researchers, such as sparse data, the observer's paradox, and annotating corpora at scale. This article presents a Twitter dataset of 227 million conversational messages across the U.S. and uses deep learning natural language processing (NLP) methods to analyze how 18 morphosyntactic features used by speakers of African American Language (AAL) vary geographically and across 12 demographic factors. Results demonstrate more frequent usage of AAL features in the rural South and in Mexican American communities, both of which are underrepresented in the literature. This work constitutes the first national-level description and analysis of overall morphosyntactic variation in AAL, and demonstrates how NLP tools enable the study of large-scale data to gain a more representative understandingof speech in marginalized communities.
This study investigates the speech characteristics of the American South and of Mexican American English by focusing on /ai/-monophthongization, the prenasal /i/-/s/ merger (PIN-PEN merger), and the /ae/ nasal split. The analyses are based on the speech of58 speakers who are native to South Texas. The results show that South Texas is partially Southern in its vowel production regardless of ethnicity. The study also finds that Mexican Americans in South Texas retain speech characteristics of Mexican American English, making theirspeech both Southern American and Mexican American. While ethnicity does surface as a predictorof the /ae/ nasal split, the study observes a strong apparent-time convergence between Mexican Americans and White Americans with respect to the examined variables. The results are discussed in the context of the relationship between Southern English, a dialect that is inherently regional, and Mexican American English, a varietythat is historically ethnic.
Which can behave as a"connective," marked by lacking a gap in what follows. A variety of unique discursive functions have been posited for such uses. The literature hints that American English could have more "types" of connective which than other varieties. This study uses a small, researcher-collected spoken corpus of podcast episodes to explore the distribution and discursive functions of which in American English. The corpus contains more connective whiches than British and Australian corpora, including a new type that does not precede a clause at all. Despite the diversity of which constructions, they all do similar discursive work: they allow a speakerto insert relevant information or, more often, subjective views into the discourse. This trajectory from organizingtext to also inserting expressive information is supported by novel evidence that which can end a turn in American English. Similartrajectories have been documented for though, but, and so.
In American English, the pronunciation of words like ‘button,’ with the underlying post-tonic string /tən/, is variable. Previous research has found that although the traditional pronunciation of these words is with [ʔn̩ ], as in [bʌʔn̩ ], this is not the only possible pronunciation. Recent studies have shown that /ən/ is realized as [ən] in certain speech communities, resulting in pronunciations like [bʌʔən]. Furthermore, there have been anecdotal reports that /t/ can be realized as [ɾ], resulting in pronunciations like [bʌɾən]. This article examines whether there is indeed a change underway in /ən/ and /t/ realizations, how these phonemes are interrelated, and what factors and populations are conditioning this change. To address these issues, a production experiment was carried out with participants from Long Island, NY. Results suggest that there is an early-stage change underway, such that younger speakers are more likely to realize /tən/ words with [ən] than [n̩ ]. Realization of /ən/ as [ən] is also significantly correlated with /t/ being realized as [ɾ] and a faster speech rate. There is not yet evidence for a change in progress for /t/ realizations, but [ɾ] productions are significantly correlated with the realization of /ən/ as [ən].
The Alternative Embedded Passive (AEP) occurs in the Midland region in variation with the infinitival passive. AEPs may occur with the verbs need, want, and like: The baby needs/wants/likes fed. Murray & Simon (2002) observe a hierarchy of acceptability and usage across the three AEP verbs, with need as the most acceptable and like as the least. Building on an idea in Edelstein (2014), this paper presents evidence that the hierarchy results from lexical semantic properties of need, want, and like. Experiment results reveal that the hierarchy is alive in the judgments of young Pennsylvania speakers, and also emerges partially in the judgments of speakers from New England unfamiliar with the construction. The hierarchy only fully emerges with Pennsylvania speakers when the subject of the sentence is sentient (e.g., the baby). With non-sentient subjects (e.g., the computer), want and like are similar and less acceptable than need. We attribute this to a property of the verb need, which can be either a deontic modal (which does not take a thematic subject) or a thematic subject-taking verb. We discuss how our results support the hypothesis that speakers adapt to novel dialectal constructions by forming analogies to constructions already in their grammars.
The Detroit Dialect Study (DDS) is one of the earliest foundational studies of social stratification in American English and the most expansive survey of an urban area ever undertaken in sociolinguistics. The DDS, led by Roger Shuy, systematically gathered and analyzed data to determine the linguistic patterns at play across race, sex, social class, and age, and the results of this study have informed work on educational policy regarding vernacular American English dialects. At the same time, the DDS allows for deeper insights into the development of African American Language and the spread of the Northern Cities Vowel Shift at a much greater time-depth than current digitized collections, essential for linguists interested in the mechanisms of language variation and change over time. This audio feature highlights the newly digitized and publicly available DDS recordings. Roger Shuy, Walt Wolfram, and Ralph Fasold, involved with the DDS as principal investigator, fieldworker, and researcher, respectively, reflect on the study a half-century later. Complementing their observations are selections from the 1966–67 DDS recordings that highlight linguistic patterns and important stories. The producers hope to highlight the rich data available within, as well as different ways the dataset can be used today.
A panoramic overview of French in North America (and related creoles) reveals a complex interaction ofvectors characterizing both globalization and localization. Though globalization threatens the preservation of local linguacultural integrity, it can be harnessed as a potential solution to that same threat (howbeit partial) by resorting to planned international immigration in orderto swell local ranks offranco-phones throughout Canada, by recruiting an international francophone teachingcorps to raise up the next generation offrancophones in Louisiana, or by usingthe same democratized media facilitating globalization itself to commoditize and broadcast local legacy language and culture to an unbounded audience. However, whether in Ontario, Acadia, Louisiana, or Maine, globalization used as a method for preserving French seems to come with a price, namelythe devernacularization of legacy French and its potential replacement with a different French. Conversely, though unplanned, when felicitous vectors merge, globalization becomes the agent leading to the creation ofa localized French (or Haitian Creole) space where previously there had been little awareness of one, as in New York City and Florida. The most recent figures available (at the time of writing, May 2024) for the relevant populations offrancophones and creolophones are provided.
This article investigates variation in verbal morphology following negative auxiliary contractions, most notably ain't, in past-tense and present-perfect constructions in Philadelphia African American English. Ain't is of particular interest given its tense-aspect ambiguity, which may not overtly restrict following verbal forms in the same way as other auxiliaries. It examines 774 instances of negative auxiliary contractions (hasn't/haven't, didn't, and ain't) for whether they were followed by base (e.g., give), preterit (e.g., gave), or participle (e.g., given) verbal forms in a corpus of naturalistic speech from 42 speakers ranging in age (13-81) and social histories. Analysis of 85 present-perfect constructions shows that ain't is followed by the same verbal forms (preterit, participle) as hasn't/haven't except in five rare cases (base forms) where priming by preceding forms of the same lexical verb may play a role. In the 689 past tense constructions, while the majority of verbs following ain't are found in base form, identical to forms following didn't, 25% of verbs are preterits. In these cases, speaker age is found to be the most influential conditioning factor. Overall results are examined in light of the structural relationship between negation and the expression of tense-aspect morphology in varieties of English.
