This paper investigates analogical leveling in a small set of English nouns that have irregular plural forms. In these nouns, all of which end in a voiceless fricative, the fricative standardly voices in the plural (e.g., wolf–wol[v]es, path–pa[ð]s, house–hou[z]es). Using audio data from three large spoken corpora of American English, I demonstrate that this stem-final fricative voicing is variable and conditioned by a number of factors, most notably the identity of the stem-final fricative—with /f/-final lexemes (e.g., wolf), /θ/-final lexemes (e.g., path), and the /s/-final lexeme house all patterning differently in apparent time—and the frequency of a lexeme in its plural form. I argue that the way these two factors affect the variation is reminiscent of the patterns seen in children's first language acquisition errors, providing a potential source for the variation and underscoring the importance of considering morphophonological factors when accounting for patterns of change.

This paper examines stylistic variation in the (oh), (o), (aw), and (ay) classes among native speakers of Canadian English living in or just outside either New York City or Washington, DC. Speakers show evidence of change toward US norms for all four vowels, though only (aw) shows consistent style shifting: prevoiceless (aw) is realized with higher nuclei when speakers express ambivalence about or distance from the United States, and lower nuclei when closeness to or positive affect about the United States is being conveyed. Canadians in New York also show topic- and stance-based shift in (oh): (oh)s are higher when expressing positive affect or closeness to New York City and lower when expressing negative affect or distance. These results suggest that mobile speakers continue to exploit the socioindexical links in their native dialect while learning and using new links in their adopted dialect—but only if those links are socially salient.

In this study, we identify the linguistic and social predictors that condition onset /s/ weakening in speech data from sociolinguistic interviews with 72 Salvadoran Spanish speakers. In addition, we compare and contrast the explanatory power of instrumental and traditional segmental approaches. We find that the instrumental approach, which identifies flanking segments, stress, and region of origin of the speaker as conditioners of onset /s/ shortening and lowering of center of gravity, does not account for observed social variation in the data. Contrastingly, an ordinal logistic regression based on a combination of instrumental measures and perceived phonetic categories identifies flanking segment, region of origin, sex, and age of the speaker as predictors of onset /s/ weakening. We conclude that an exclusively instrumental analysis examining variation of onset /s/ thus obscures the potential social meaning of onset /s/ weakening in El Salvador.

This paper uncovers evidence for two linked levels of morphosyntactic change occurring in Canadian English. The more ordinary is a lexical replacement: with finite subordination after seem, the complementizer like has been overtaking all the alternatives (as if, as though, that, and Ø). On top of this, there is a broader syntactic change whereby the entire finite structure (now represented primarily by like) is beginning to catch on at the expense of infinitival subordination after seem. Drawing on complementary evidence from British English and several partial precedents in the historical linguistics literature, I take this correlation to mean that like has reached sufficient rates among the finite strategy to have instigated the second level of change, to the point that it has ramifications for epistemic and evidential marking with the verb seem. I propose that the best model of these trajectories is a set of increasingly large envelopes of variation, one inside the next, and argue that the envelope might itself be an entity susceptible to change over time.

This paper explores the relation between controlled and automatic perceptions of a sociolinguistic variable that yields no metalinguistic commentary—a marker (Labov, 1972). Two experiments examine links between the backed trap vowel and its social meanings. The first, a matched guise task, measures social evaluations of the feature in a relatively controlled, introspective task. In the second, two measures are used that access different points in online processing and different degrees of listener control: (a) lexical categorization of an ambiguous stimulus, measured by a mouse click, and (b) automatic, early responses to this ambiguous stimulus, measured by eye movements. While listeners perceptually link trap-backing with social information in all three measures, specific social effects differ across the measures. Findings illustrate that the task and time course of a response influence how listeners link a linguistic marker with social information, even when this sociolinguistic knowledge is below the level of conscious awareness.