Infixation and allomorphy have long been investigated as independent phenomena—see, for example, Ultan 1975, Moravcsik 1977, and Yu 2007 on infixation, and Carstairs 1987, Paster 2006, Veselinova 2006, and Bobaljik 2012 on allomorphy. But relatively little is known about what happens when infixation and allomorphy coincide. This article presents the results of the first crosslinguistic study of allomorphy involving infixation, considering fifty-one case studies from forty-two languages (fifteen language families). Allomorphy and infixation interact systematically, with distinct sets of behaviors characterizing suppletive and nonsuppletive allomorphy involving an infix. Perhaps most notably, suppletive allomorphy is conditioned only at/from the stem edge, while nonsuppletive allomorphy is conditioned only in the surface (infixed) environment. The robustness of these and related findings supports a universal serial architecture of the morphosyntax-phonology interface where: (i) infixation is indirect, involving displacement from a stem-edge position to a stem-internal one, counter to several influential theories of infixation (see especially McCarthy & Prince 1993a and Yu 2007); (ii) suppletive exponent choice is prior to (i.e. not regulated by) the phonological grammar (in line with Paster 2006, Pak 2016, Kalin 2020, Rolle 2021, and Stanton 2023, inter alia); and (iii) realization—including exponent choice and infixation—proceeds from the bottom of the morphosyntactic structure upward (à la Bobaljik 2000, Embick 2010, Myler 2017).

In many cases of stress-dependent harmony the trigger is associated with a morpheme. We examine two instances of morphemic harmony where the triggering morpheme is mixed, that is, it consists of segmental material and floating features. The floating features cause stepwise raising of the stressed vowel, /a/ → [e], and /e/ → [i], /o/ → [u]. We examine in particular Felechosa Asturian, where the triggering masculine singular count morpheme usually has the exponent /-o/ and the floating features [+high] and [–low]. When the stressed vowel is mid, /-o/ raises it to high (/negɾ-o/ → [‘niƔɾ-o]). When the stressed vowel is /a/, the suffix raises this vowel to [e] as predicted, but at the same time the triggering morph /-o/ raises to [u] (/blank-o/ → [‘bleŋk-u]). This phenomenon, which we call harmony in situ, derives from the fact that, because raising is stepwise, one of the floating features cannot link to the stressed vowel, and thus it has to be realized on the trigger itself. Felechosa Asturian is compared to Ḷḷena Asturian, which does not present harmony in situ, and an optimality-theoretic analysis is provided.

Hindi and several other Indo-Aryan languages contain a discourse marker that has been described as having a wide range of functions, including topic marking, intensive, emphatic, contrastive, and assertive. In Hindi, this function is realized by the enclitic =to. Possible translational equivalents for =to include expressions like in fact, sure, you know, well, as for, at least, finally, and but. This article investigates the diverse uses of =to and argues that the full range can be uniformly accounted for only if =to is taken to be a particle that signals that the question resolved by its prejacent is weak. The analysis treats =to as a generalized downtoner that comments on the strength of the question the prejacent addresses, relative to the speaker's information state, prior discourse moves, and assumptions about the common ground.

Social category information plays a central role in speech perception and processing. To date, research on this topic has struggled to model how social category perceptions evolve over the time course of an interaction. In this article, we build on recent methodological developments to investigate trajectories of listener perceptions, focusing on how impressions change as linguistic, social, and contextual details emerge. We base our arguments on an analysis of listeners' real-time evaluations of the perceived competence of speakers of two British regional accents during a mock interview for a job in a law firm. Results indicate a need to move away from the view, predominant in sociolinguistics, of category perception as a discrete phenomenon and toward a model of perception as inference under uncertainty. We discuss implications for theories of sociolinguistic cognition and for understandings of accent bias.

I argue that the phenomenon of ‘extra be’ (e.g. That's the thing, is we lost) can be analyzed in terms of specifying coordination. Specifically, ‘extra be’ derives from a ‘colon phrase’ (:P) structure, where Spec-:P is a host sentence and Comp-:P is a pseudocleft adding information to the host. ‘Extra be’ arises when the head T of the pseudocleft raises to :0, an operation that is possible only under specific circumstances involving ellipsis in the pseudocleft. I motivate this analysis by first considering a set of syntactic, prosodic, and semantic properties exhibited by extra be sentences, including properties of ‘extra be’ itself, properties of the post-copular specificational phrase, and locality conditions in the construction. I then develop the analysis described above, emphasizing in particular the assignment of a uniform structure to both extra be sentences and their ‘extra be'-less counterparts (compare: That's the thing {: / is} we lost). Finally, I compare key features of the new analysis with those of previous proposals.

Since Aristotle first set out rules of natural priority, rhetoricians and linguists have sought to establish the ‘natural order’ of words, phrases, and clauses. Accounts of constituent order by classical rhetoricians and philologists and by modern linguists and psychologists have addressed word order within phrases and phrasal order within clauses. However, they have not tended to investigate clausal order within sentences, with the important but limited exception of NARRATION sequences (They had a baby and they got married), which—as recognized from Dionysius ('What is prior in time should also be prior in word order') to Grice ('Be orderly')—exhibit a robust but defeasible iconic link between order of events and order of mention. For clauses exhibiting the rhetorical relation of CONTRAST rather than narration, the literature is less perspicuous. It is on such cases that I focus here, inspired by BEHAGHEL'S SECOND LAW (1932:4): ‘That which is less important (or already known to the listener) is placed before that which is more important (or unknown) … Old concepts are placed before new'.

Recent research has shown that a distinct variety of American Sign Language, known as Black ASL, developed in the segregated schools for deaf African Americans in the US South during the pre-civil rights era. Research has also shown that in some respects Black ASL is closer than most white varieties to the standard taught in ASL classes and found in ASL dictionaries. This article explores the circumstances that resulted in the creation of a distinct ASL variety, with attention to the role of language in education policy in both the white and Black Southern schools for the deaf. Archival research shows that while white deaf students were long subjected to oral instruction and forbidden to sign in class, Black students, although their severely underfunded schools provided only basic vocational education, continued to receive their education in ASL, with classes often taught by deaf teachers. The differences in language education policy explain the difficulties Black students experienced in understanding their teachers and white classmates after integration occurred, despite great resistance, in the 1960s and 1970s, as well as the fact that Black signers from the South, particularly older Black signers, are more likely than their white counterparts to use traditional features.

This article describes a pedagogical innovation implemented in our introductory linguistics course. We supplement classic theory building with a series of labs, deployed through a co-requisite ‘lab’ course that meets weekly. This builds on two previously established teaching strategies: the implementation of hands-on activities in linguistics classrooms, and the lab sections traditionally utilized in the natural sciences. The labs aim to fulfill three goals: (i) to better represent the field of linguistics in our introductory course, (ii) to help students solidify theories and connect them to the real world, and (iii) to teach practical skills for linguistics research and more broadly.

Despite the long tradition of using the scientific method to study language, there is a widespread, if largely anecdotally based, feeling among language scientists that the general public does not perceive language to be a classic object of scientific study. The goal of the current study was to investigate this notion. We report the results of three experiments conducted in informal science learning environments that: (i) confirm the public thinks of science and language as fundamentally different objects, and (ii) show there are some areas of language science that are more readily accepted as science than others. Our results also suggest that high-impact demonstrations of core linguistic phenomena may be used to encourage people to recognize that language can be, and often is, an object of scientific study. Although the public has an incomplete understanding of the study of language, we argue that the strong humanistic approach with which the public associates the study of language can be seen as an opportunity to broaden participation in science.