Five MSU linguists (including Sociolinguistics Lab members Mofart Ayiega & Leah Nodar) will be presenting their work at the Linguistics Society of America’s annual conference, taking place this year in New Orleans from January 8th – 11th!
Please find their talk titles below:
- Mofart Ayiega (co-authored with Dr. Suzanne Wagner) – Morphological Leveling of Noun Class Agreement in Urban Swahili
- Dr. Leah Nodar – Twelve Variables of Africatown English in 1927 and 1979
- Jingying Xu (co-authored with Dr. Cristina Schmitt):
- Beyond Truth Conditions: Context Modulates Telicity Interpretation
- Learning Telicity in Context: Developmental Evidence from Mandarin Children
- Ellie Xia (co-authored with Dr. Alan Hezao Ke) – Tonal Marking of Telicity in Hakka: An Agree-Based Analysis
- John Ryan (co-authored with Yaxuan Wang) – Reference, Aspect, and Event Completion in Mandarin Sentence Judgments
Click ‘Continue Reading’ for abstracts.
ABSTRACTS
Mofart Ayiega (co-authored with Dr. Suzanne Wagner) – Morphological Leveling of Noun Class Agreement in Urban Swahili
This study reports initial findings on morphological leveling and noun class simplification in Swahili subject-verb agreement, based on a picture description task with six young adult speakers in Nairobi. Swahili typically aligns animate nouns with subject markers a-/wa- and inanimate ones with i-/zi-. However, speakers increasingly extend i-/zi- across animate nouns, regardless of traditional rules. Some also distinguish agreement based on whether an animal is alive or dead. Variation was especially notable with zero-prefix nouns like samaki ‘fish’. These patterns point to structural simplification shaped by urban multilingualism and peer-group identity, warranting deeper analysis with a larger speaker sample.
Jingying Xu (co-authored with Dr. Cristina Schmitt) – Beyond Truth Conditions: Context Modulates Telicity Interpretation
This study examines how context modulates telicity interpretation, drawing on Piñón’s degree-based semantics and Pietroski’s instruction-based framework. English speakers judged telic sentences with numeral or demonstrative DPs across two incomplete-event types and two presentation orders. In 3rd Object Incomplete scenes, numeral sentences were rejected more often than demonstratives—but only when the video was presented first. We argue that demonstratives support contextual accommodation of contrast sets. These findings suggest that telicity judgments reflect not only compositional semantics, but also how linguistic instructions are pragmatically implemented in real-time interaction with non-linguistic context.
Jingying Xu (co-authored with Dr. Cristina Schmitt) – Learning Telicity in Context: Developmental Evidence from Mandarin Children
This study explores how Mandarin-speaking children interpret telic sentences in incomplete event contexts. We argue that developmental differences arise not from core aspectual meanings but from how children pragmatically implement telicity instructions. Two experiments compared Mandarin-speaking children (ages 4–6) and adults using truth-value judgment tasks. Adults showed flexible interpretation based on determiner type and visual contrast. Younger children relied on object quantization and perceptual cues, while 6-year-olds showed emerging sensitivity to determiner type. When embedded in after-clauses, 6-year-olds showed more adult-like judgments, suggesting that the relevant pragmatic knowledge is present but requires scaffolding for effective integration.
Ellie Xia (co-authored with Dr. Alan Hezao Ke) – Tonal Marking of Telicity in Hakka: An Agree-Based Analysis
In Hakka, telicity is not marked segmentally (e.g., via affixation), but rather prosodically, through tonal morphology. Specifically, predicates that occur in perfective contexts undergo a tonal shift from their lexical tone to a rising Tone 35. We analyze this as the phonological realization of a syntactic feature: the head AgrO, which carries an [uTELIC] feature, is spelled out prosodically as a rising tone on the next head via Agree. Building on Schmitt (1996), we propose that AgrO immediately c-commands the maximal projection that contains the internal argument, thereby ensuring locality for feature checking and operator binding in telicity computation.
John Ryan (co-authored with Yaxuan Wang) – Reference, Aspect, and Event Completion in Mandarin Sentence Judgments
This study investigates how grammatical aspect and referential form interact in Mandarin event interpretation. Participants rated the naturalness of sentences describing LEGO puzzle-making, varying by aspect (perfective le vs. progressive zai) and referential form (indefinite yi+CL vs. definite zhe+CL), paired with images showing complete or incomplete puzzles. Results show that perfective aspect aligns with completed events regardless of referential form, while progressive interpretations depend on definiteness: definites license both complete and incomplete readings, but indefinites disfavor incomplete contexts. These findings suggest that referential specificity modulates aspectual interpretation, with demonstratives supporting pragmatic accommodation in progressive contexts more readily than indefinites.
Dr. Leah Nodar – Twelve Variables of Africatown English in 1927 and 1979
In 1860, the last known slave ship to land in America arrived in Mobile Bay, Alabama, USA. After the Civil War, the shipmates from this vessel founded Africatown, a semi-isolated community north of Mobile (Diouf 2007). Contemporary sources describe this group as speaking a unique variety of English (Diouf 2007, Allen Jones 1979). This variety is portrayed in Zora Neale Hurston’s Barracoon (2018), based on interviews from 1927-8 she conducted with Oluale Kossula (aka Cudjo Lewis), then the last shipmate living in Africatown.
Africatown, Alabama is a unique part of American history worthy of study in itself. In addition, information from the history of this variety holds potential insights for the historical development of Black English in America more broadly. Barracoon (2018) is a rare instance of a long, detailed portrayal of the English of a man born in Africa who survived the Middle Passage. This text offers evidence relevant to one strand of American Black English’s multifaceted origins (Kautzsch 2008, Mufwene 2000): features of a native Yoruba speaker who as an adult learned Southern American English from Black and White Mobilians. In previous research (Nodar 2023), I compared Barracoon to another Hurston work, Mules and Men ([1935] 2008), an anthropological text portraying speakers of Black Southern English from Eatonville, Florida; Polk County, Florida; and New Orleans, Louisiana. This analysis found 28 written representations of phonetic/phonological or morphosyntactic features that differed significantly between the two books, suggesting a difference heard and portrayed by Hurston between these Southern Black Englishes and the English of Kossula of Africatown.
In this work, I analyze 12 out of the 28 written representations as linguistic variables in 11 audio recordings of Black residents from Africatown and Mobile that were collected in the 1978-9 Mobile Public Library Oral History Project. Of these 11 speakers, two were raised in Africatown and are descendants of the shipmates; two were were raised in Africatown and are not descendants of the shipmates; and the final seven are from other Black communities around Mobile. For each interview I code the 12 variables and determine the rate of the variant that Hurston suggests was distinctive in early Africatown (henceforth, the Barracoon-like variant). I then ask: for each feature, does the rate of use of the Barracoon-like variant distinguish (A) shipmate descendants from Black Mobilians (including Africatown residents who are not descendants); or (B) Africatown residents (descendants and non-descendants) from Black Mobilians from other communities? In other words, do the Barracoon-like variants continue to distinguish either (A) shipmate descendant speech or (B) Africatown-area speech?
Logistic regression, with either (A) descendant status or (B) Africatown status as the dependent variable, finds no statistically significant difference for any variable. However, secondary testing found that for certain variables, inclusion improves fit of one of the models, as shown by a lower AICc, AIC, and BIC; has a significant likelihood ratio test; and has a large effect size. These together suggests it is plausible, though speculative, that with a larger sample size they would achieve significance in a regression. Model (A) shows this improvement with rates of [ð]-stopping and infinitive to absence. Model (B) is improved with rates of past tense ain’t, relativizer what, is/are copula absence, and was/were copula absence. In all of these cases the difference was in the expected direction, i.e. Africatown speakers using more of the Barracoon-attested variant.
