Linguistics

We present the results of a semantic judgment task with African American English (AAE), Appalachian English (AppE) and Mainstream US English (MUSE) users. Participants read sentences such as Nobody didn't see me with a negative subject and negated auxiliary. Sentences were presented in contexts that biased toward double negation (='everybody saw me') or negative concord (='nobody saw me') readings, and were followed by True/False judgments on statements that revealed participants' interpretation of the sentence. The results afford novel evidence for a point of microvariation in American English negation, since AppE and AAE users more readily generate both NC and DN meanings for the sentence ..

A good deal of research has established production differences in consonants depending on their prosodic position, particularly at constituent edges. For example, they tend to be longer and exhibit greater spatial magnitude at phrase edges compared to phrase-internally. This effect seems to be cumulative, with more strengthening when the consonant is initial at a higher level of the prosodic hierarchy. An ultrasound study reported here tests whether production of English /r/ varies in spatial magnitude as expected when initial across prosodic levels, comparing phrase-initial, word-initial and syllable-initial positions. Little evidence of such cumulative prosodic strengthening is found. I re..

Under some conditions, English infinitival relative clauses exhibit variable modal force that fluctuates between possibility and weak necessity. We report on a series of experiments that test two claims from the theoretical literature: (1) an infinitival relative’s force is manipulable by if-clauses that tend to bias the interpretation one way or the other and (2) infinitival relatives embedded under strong determiners have a weak necessity reading only. The results of our experiments support claim (1) but not claim (2). We close with a discussion focusing on why no effect of determiner was found, in spite of what is reported in the literature.

In this paper, I offer a detailed description of polar questions in San Sebastián del Monte Mixtec, by doing so contributing to the typology of polar questions in tonal languages. I report how polar questions are formed in three different varieties of Mixtec, looking at pitch raising and presence or absence of the question particle, and I then go into a detailed description of San Sebastián del Monte Mixtec polar questions. I analyze data produced by four native speakers with the use of a script that controls for tone production.

We build on Muwwakkil's work on the ethnography of campus conservatives and Kiesling's work on the sociolinguistic repertoires of fraternity men in this work of longitudinal, multimodal ethnography. From January 2025 to March 2026, we conducted two sociolinguistic interviews with a male Catholic student activist and member of the Newman Catholic Student Association at the Pennsylvania State University and shadowed him in a variety of discursive contexts, including his pro-life activism, podcast appearances, and public education of Catholic bioethics. We identify two lexico-prosodic assemblages, the affiliation- and expertise-signaling frames. Counterintuitively, we argue that his speech is c..

Constructed dialogue of animals has been used as an interactional resource in discourse (Tannen et al. 2004). I examine the constructed dialogue of their cats in one rescue cat cafe’s Instagram. In these texts, cats are personified through role language, a style of speech utilizing linguistic stereotypes to create recognizable characters (Kinsui 2003). Cats speak with highly marked sociolinguistic features (e.g. first-person pronouns, sentence-final particles, and sound substitutions) that index age and gender to create cat personas, and are further portrayed as unique personae, such as elderly male hosts or rouge cats. Drawing on entextualization, intertextuality (Briggs and Baumann 1990,..

Several authors have proposed that wh- adjuncts such as why can be externally merged in the left periphery of the clause they modify, rather than undergoing movement (e.g., Ko 2005; Buell 2011; Abels 2012). This paper provides another argument in favor of this view based on the distribution of negation strategies in Gweno, a Bantu language of Tanzania. Gweno has two negation strategies: a post-verbal particle strategy used in most main clauses and a post-initial prefix strategy used in most dependent clauses. Examining wh-questions, we show that the distribution is explained if the post-verbal strategy is derived via heavy pied-piping and therefore incompatible with extraction. Wh-questions ..

This paper examines the categorial status of and syntactic structures involving the morpheme gei `to give' in Mandarin Chinese, which can be a standalone ditransitive verb but also appears in the preverbal, post-verbal, and post-object positions. I challenge existing multi-category analyses as well as the unified verbal analysis of gei, the former regarding their argument for the prepositional status of the preverbal gei, and the latter on oversight regarding structural ambiguity of various gei constructions, as well as on the grounds of theoretical parsimony. I show that the post-object gei is a preposition in the prepositional dative construction, or the second verb in a serial verb constr..

This article examines the diachronic process of grammaticalization of the emphatic construction koso -e, one of the focus-concord constructions in Old Japanese. The starting point is the various forms and meanings exhibited by koso -e, especially those with strong negative implicature, as opposed to those with positive implicature. Haruaki Ishida (1939) points out that the early uses of koso -e do not conform to the KM rule, as set forth by Norinaga Moto'ori, a prominent 18th-Century Japanese linguist. I contend that the puzzle presented by Ishida has not been solved by the most recent works on this construction. I support the grammaticalization view of KM, that a focus-concord construction ..

Japanese internally-headed relative clauses (IHRCs) have long been analyzed in terms of Kuroda’s Relevancy Condition, especially its temporal component. We argue that this descriptive generalization follows from a deeper compositional mechanism. Building on a Davidsonian framework with a typed distinction between events and states, we propose that IHRCs introduce a state, either directly (when the embedded predicate is stative) or indirectly via a target-state operator that maps event predicates to state predicates. The nominalizer -no selects for state predicates, and the matrix argument must be interpreted as an entity that is in the relevant state at the time of the matrix event. On thi..

This study investigates the distribution of (in)definiteness expressed in Hindi bare nominals. Within a Neocarlsonian framework (Dayal 1992; 1999; 2004; 2018), Hindi bare nominals have kind readings like English plurals and additional indefinite and definite readings. Indefinite interpretations are reported to be more restricted for bare singulars than for bare plurals, creating a strong asymmetry between them. We demonstrate that Hindi bare singulars can receive indefinite interpretations under certain discourse, which is also observed in naturalistic corpus data. We show that these uses introduce new discourse referents and allow anaphoric continuation, demonstrating the availability of in..

Contrary to what one might assume, comprehenders may not build fully accurate representations of the linguistic input when processing language. For example, when asked “how many animals of each kind did Moses take on the Ark?”, many people respond ‘two’ despite knowing it was Noah, not Moses (Erickson & Mattson, 1981), suggesting lexical semantics of related words interferes during processing. Prior research attributes susceptibility of these illusions to representations emerging from strong top-down expectations (e.g. Ferreira & Patson, 2007; Reder & Kusbit, 1991). Building on this, we tested whether a second (non-semantic) error impacts detection of semantic illusion er..

This paper addresses a theoretical tension in accounting for the correlation between the landing site and the locality domain of movement—whether to reduce it to the A/A'-distinction (Ban on Improper Movement) or to hard-wire it to the grammar as the Williams Cycle (WC), which has recently been claimed to apply to Mandarin Chinese by Yan & Meadows (2025). By refuting their arguments, we argue that the WC does not apply to Mandarin and therefore cannot be universal. Specifically, we show that the apparent WC effects in Mandarin either follow from independently attested A/A'-distinction in the language, or simply do not hold due to incomplete generalization. The Mandarin case highlights ..

Zuraw et al. (2019) and O’Leary & Stockwell (2021) show Skills-Based Grading (SBG) to be an alternative grading method well-suited to linguistics courses. SBG asks students to show proficiency in up to 50 “skills” (number varies by course) to earn their course grade and integrates leniency and other benefits by allowing students many opportunities to show proficiency. However, in SBG, students often need many explanations of the system, and, due to the importance of repeated chances, the number of exercises on assignments can grow to unwieldy numbers for both students and instructors. This paper presents a novel implementation of SBG which significantly simplifies the system by cre..

This article demonstrates how the Functional element J-unct that is introduced in Zhang (2023, 2024a,b) can be used to build negative approximative adverbials, appropriately capture their semantics, and account for the two unique interpretations of hardly. It posits that rather than a NegP, the syntax contains an abstract negative adverbial operator ¬op. It is therefore an extension of the research on negation syntax in Newmeyer (2006), Murakami (2007), and Bruening (2025).

What spoken-language features characterize speakers from Baltimore, Maryland, and how have they changed in apparent time? Situated in the Mid-Atlantic dialect region, Baltimore is argued to lack “substantial differences” with Philadelphia, though its lexical isoglosses stretch west, and it also shows Southern influences. We reveal local distinctness among Baltimoreans living in predominantly Black East Baltimore and White working-class Hampden. We phonetically analyze three back upgliding vowels /aʊ oʊ u/, which are all synchronically (though differently) fronted by these groups. Our data come from the Baltimore Neighborhood Heritage Project, which contains 232 oral history interviews ..