Selected papers, presentations, & posters

For a complete list, please see my CV. Feel free to email me for a copy of a paper or presentation not listed here.


Comparing constraints on contraction using Bayesian regression modeling  

This paper has three goals: (1) to document the factors shaping is-contraction in Mainstream American English; (2) to assess the extent to which these factors also shape contraction of has; (3) to use shared patterns of contraction across the two verbs to draw conclusions about how the varying forms are represented grammatically. While is has two distinct phonological forms in variation, has has three. This necessitates regression modeling which can handle non-binary response variables; I use Bayesian Markov chain Monte Carlo modeling. Through this modeling, I (1) uncover a number of novel predictors shaping contraction of is, and (2) demonstrate that many of the patterns shown by is are also in evidence for has. I also (3) argue that modeling has-variation as the product of two stages of binary choices—a common treatment of three-way variation in variationist sociolinguistics—cannot adequately explain the quantitative patterns, which are only compatible with a grammatical model under which three distinct forms vary with each other. The findings have theoretical and methodological consequences for sociolinguistic work on ternary variables.

Paper, Frontiers in Artificial Intelligence: Language and Computation 3:58.

Assessing the accuracy of existing forced alignment software on varieties of British English  

This paper presents an analysis of the performance and usability of automatic speech processing tools on six different varieties of English spoken in the British Isles. The tools used in the present study were developed for use with Mainstream American English, but we demonstrate that their forced alignment functionality nonetheless performs extremely well on a range of British varieties, encompassing both careful and casual speech. Where phone boundary placement is concerned, substantial errors in alignment occur infrequently, and although small displacements between aligner-placed and human-placed phone boundaries are found regularly, these will rarely have a significant effect on measurements of interest for the researcher. We demonstrate that gross phone boundary placement errors, when they do arise, are particularly likely to be introduced in fast speech or with varieties that are radically different from Mainstream American English (e.g. Scots). We also observe occasional problems with phonetic transcription. Overall, we advise that, although forced alignment software is highly reliable and improving continuously, human confirmation is needed to correct errors which can displace entire stretches of speech. Nevertheless, sociolinguists can be assured that the output of these tools is generally highly accurate for a wide range of varieties.

Paper with Danielle Turton, Linguistics Vanguard 6. © Walter de Gruyter GmbH, Berlin/Boston.


Perturbing the community grammar: Individual differences and community-level constraints on sociolinguistic variation  

The traditional focus of variationist sociolinguistic research is the patterning of language variation at the level of the community, which individual language users are said to learn and reproduce (Labov 1972; 2012). In this paper, I observe that, although members of a speech community may all have learned the same grammar of a sociolinguistic variable, they may nonetheless produce that variable in ways which obscure this. This "perturbation," I argue, is epiphenomenal, stemming from at least two possible sources: individual differences in mental representations, and individual differences in speech production planning. Moreover, I demonstrate that these differences are not only inter-individual; they can also be intra-individual, such that speakers may undergo age-grading which disrupts their patterning of a variable from how they previously produced it. I ask whether these individual differences may give rise to changes in constraints in the same way that individual differences can lead to sound change. The paper concludes with a call for more research that integrates sociolinguistic, formal, and psycholinguistic approaches to the study of language variation and change.

Paper, Glossa: A Journal of General Linguistics 4.

I've always spoke(n) like this, you see: Participle leveling in three corpora of English  
Poster with Alicia Chatten, Jai Pena, Kimberley Baxter, Erwanne Mas, Guy Tabachnick, and Dan Duncan, NWAV 48.


Variable stem-final fricative voicing in American English plurals: Different pa[ð ~ θ]s of change  

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.

Paper, Language Variation and Change 30:147–174. © Cambridge University Press.  

What's in a name? Teaching linguistics using onomastic data     [supplementary material]

This article describes how students can be introduced to the basics of linguistic analysis using personal, product, and place names as data. I outline several areas of linguistics that can be effectively taught at an introductory level through name data and provide examples of accompanying in-class and take-home exercises. Throughout the article, I demonstrate that the everyday familiarity of names and the ready availability of name data combine to create a class that not only engages students but also teaches them practical data-analysis skills.

Paper, Language: Teaching Linguistics 94:e1–e18.

Dialect surveys and student-led data collection  
Vignette, Data Collection in Sociolinguistics: Methods and Applications, Second Edition, Routledge.


Frequency effects over the lifespan: A case study of Attenborough's r's  

This paper uses a small-scale case study of the speech of a single speaker at two points in time to investigate the question of whether and how speakers’ mental representations change over their lives. Specifically, I test two predictions of usage-based models of phonological representation: that individuals surrounded by a changing community will show the community change in their own production, and that this individual-level change will show an effect of item frequency. The community change under study is the loss in English Received Pronunciation of [ɾ] as a realization of /ɹ/; the speaker studied is Sir David Attenborough, a well-known British nature documentary narrator. I find that Attenborough’s narrations do not show evidence of him participating in the community change away from [ɾ] over time; however, he does show a different sort of change, by which he increases his rate of [ɾ] in high-frequency collocations in later life. I propose that this result may be attributable to Attenborough’s mental representation of high-frequency collocations becoming more word-like over time. The results speak to questions about the malleability of mental representations and the role of the individual language user in cases of community change.

Paper, Linguistics Vanguard 3. © Walter de Gruyter GmbH, Berlin/Boston.

Where does the social meet the linguistic?  
Poster with Mary Robinson, NWAV 46.


The dynamics of variation in individuals  

This paper examines the factors conditioning the production of linguistic variables in real time by individual speakers: the study of what we term the dynamics of variation in individuals. We propose a framework that recognizes three types of factors conditioning variation: sociostylistic (s-), internal linguistic (i-), and psychophysiological (p-). We develop two main points against this background. The first is that sequences of variants produced by individuals display systematic patterns that can be understood in terms of s-conditioning and p-conditioning (with a focus on the latter). The second main point is that p-conditioning and i-conditioning are distinct in their mental implementations; this claim has implications for understanding the locality of the factors conditioning alternations, for the universality and language-specificity of variation, and for the general question of whether grammar and language use are distinct. Throughout the paper, questions about the dynamics of variation in individuals are set against the typical community-centered variationist perspective, with an eye towards showing how findings in the two domains, though differing in explanatory focus, can ultimately be mutually informative.

Paper with Meredith Tamminga and David Embick, Linguistic Variation 16:300–336.

Production planning effects on variable contraction in English  

This paper explores the potential role of the incremental planning of speech in interfering with the conditioning of the variable contraction of English is. Previous research has found that a variable alternation which is conditioned by the nature of the element that follows it can have this conditioning disrupted when a speaker fails to plan what that following element will be (Wagner 2011, Tanner et al. 2015). The strength of the effect of that following element on the variable alternation thus diminishes the less likely advance planning is. I extend this research, which has so far only examined following phonological elements, to look at whether this finding holds when a following element effect is localized in the syntactic domain.

Taking is-contraction as my dependent variable, I first provide a detailed account of the role of following constituent category in conditioning this variable, documenting a robust effect in Mainstream American English with a hierarchy of environments very similar to what has been found in studies of the contraction and deletion of is in African American English. I then investigate an acoustic proxy for advance planning—duration of the word following is—and find that, while it does play a role in conditioning contraction, it does not interact with the following constituent effect. I connect this finding to the proposal that advance planning scope differs for different levels of grammar (Wagner et al. 2010). More broadly, I underscore that the patterning of sociolinguistic variation may be shaped, not only by the language-internal and social factors that are familiar from decades of research, but also by constraints on the language production system.

Paper, Penn Working Papers in Linguistics 22.2, 121–130.


Doing sociolinguistics: A practical guide to data collection and analysis
Book, with Miriam Meyerhoff and Erik Schleef, Routledge.


/s/-lenition and the preservation of plurality in Modern Occitan  

This paper examines the variable weakening and deletion of /s/ in the Languedocian variety of Modern Occitan, with particular attention to how it has affected the system of plural marking in noun phrases. Using data from linguistic atlases, I demonstrate that /s/-lenition in this variety involves a stage of vocalization to [j]. I find that, where /s/ on the definite article has vocalized to [j], the immediately-preceding vowel of the definite article has undergone concomitant raising to [e]. This raising appears to preserve the difference between singular and plural despite the plural’s weakening /s/. I argue that these results support Labov’s (1994) hypothesis that the meaning of a weakening element may be transferred to a stable, co-occurrent one.

Paper, Journal of Linguistic Geography 2:59–73. © Cambridge University Press.


Variation in English auxiliary realization: A new take on contraction  

English auxiliary contraction has received much attention in the linguistic literature, but our knowledge of this variable has remained limited due to the absence of a thorough corpus study. This paper examines contraction of six auxiliaries in two corpora, considering three distinct phonological shapes in which they occur and the implications for an analysis of the grammatical processes that underlie the surface alternation in form. I argue that the data best support a two-stage analysis of contraction, one under which variation in the morphology is followed by phonetic and phonological processes. Moreover, I show that this particular analysis explains a number of patterns in the data that would otherwise be accidental. In this way, I underscore the importance of approaching the study of variable phenomena with both quantitative data and formal analysis.

Paper, Language Variation and Change 25:17–41. © Cambridge University Press.  

English auxiliary realization and the independence of morphology and phonetics  

One of the key questions in the study of language concerns the architecture of the grammar. At issue is the nature of and relationship between the systems that generate linguistic output. The present paper tests the predictions of two competing theories of grammatical architecture for a variable process of h-deletion in connected speech. A usage-based theory of grammar predicts that this phonetic lenition rule will show word-sensitivity, while a modular, feed-forward theory predicts a uniform rate of h-deletion across lexical items. Data from the Switchboard corpus supports the latter prediction.

Paper with Charles Yang, Penn Working Papers in Linguistics 19.2, 121–129.

Locating linguistic variation: A case study of English auxiliary contraction
Paper, NELS 41: Proceedings of the 41st Annual Meeting of the North East Linguistic Society.


Locating variation above the phonology  

The goal of this thesis is to develop a model of sociolinguistic variation that takes into account "variation above the phonology," namely, variable phenomena that implicate the morphology and/or the syntax. I develop a model under which intra-speaker linguistic variation is the product of two systems: (a) a grammar which derives forms and is partially probabilistic; and (b) a system of language use, distinct from the grammar, which deploys variants based on psycholinguistic and sociostylistic constraints. I illustrate this proposal using data from an in-depth corpus study of the variable contraction of six English auxiliaries. Two sets of findings from the corpus study support the partially probabilistic derivational grammar. First, I show that the patterning of auxiliary forms in spontaneous speech provides evidence in favor of a two-stage model of contraction, under which variation in the morphosyntax is followed by variable phonological processes. This analysis explains a number of patterns in the data which would otherwise be accidental. Second, I examine the linguistic conditions on contraction and argue that they are incompatible with an analysis under which string frequency predicts the occurrence of contraction. Accordingly, internal conditions on contraction are best treated as being encoded in the grammar, rather than as emerging from language use. Evidence in favor of a system of language use distinct from the grammar comes from the finding that contraction shows a strong effect of the number of words in an auxiliary's noun phrase subject, with contraction becoming less and less likely as a subject increases in length. I argue that this effect displays a kind of non-locality which is uncharacteristic of alternations that are the purview of the grammar, and that it should instead be interpreted as stemming from extra-grammatical, memory-based constraints on the system of language production. I localize these constraints, along with sociostylistic constraints on language variation, to a grammar-external system of language use. The dissertation thus provides evidence that variation in surface forms may be attributable to more than one underlying locus, and opens up new lines of research into conditions on variation that have their source in extra-grammatical systems.

Doctoral dissertation, University of Pennsylvania.

English auxiliary contraction as a two-stage process: Evidence from corpus data  

This paper describes the results of a corpus study of English auxiliary contraction and their implications for the place of variation in a grammatical architecture. Specifically, I argue that the quantitative patterns displayed by this variable lend themselves to a two-stage analysis of contraction, as follows: a variable process of morphosyntactic adjunction conditions insertion of either a full or a contracted form; then, low-level phonetic and phonological processes alter the phonological shape of that form. I support this analysis with data showing an effect of subject length on contracted forms, such that contracted forms are disfavored after longer subjects, with an apparent eight-word cut-off after which contracted forms categorically do not surface. I conclude by addressing the implications of this finding for a grammatical architecture, proposing that this eight-word cut-off may be best localized in a system of "usage" distinct from the grammar.

Paper, Proceedings of WCCFL 29, 152–160.


A quantitative analysis of diphthongization in Montreal French  

In Montreal French, a process of diphthongization affects long vowels: those that are inherently long due to historical compensatory lengthening (Yaeger-Dror and Kemp, 1992), and those that are allophonically lengthened before voiced fricatives and /R/ (Dumas, 1981; Santerre and Millo, 1978). Our quantitative analysis of diphthongization in real time examines both the trajectory of this change through the community as well as individual speakers’ participation in it across their lifespans. Our study also provides acoustic measurements of the Montreal French vowel system. We tracked individuals’ vowel trajectories across a 24-year span for a panel of six speakers of diverse social classes. Matched trend samples from the 1971 and 1984 Montreal corpora, with four speakers sampled per year, provide a picture of the community as a whole. We find that four vowels show significant lowering and/or backing in the community, and that all long vowels show decreased diphthongization. Some panel speakers’ longitudinal movements mirror these changes, while other speakers are stable across their lifespans and still others show apparently anomalous movements. We discuss these results and their interpretation.

Paper with Gillian Sankoff, Penn Working Papers in Linguistics 15.2, 91–100.


The acoustic and visual phonetic basis of place of articulation in excrescent nasals  

One common historical development in languages with distinctively nasalized vowels is the excrescence of coda velar nasals in place of nasalized vowels. For example, the dialect of French spoken in the southwestern part of France (Midi French) is characterized by words ending in the velar nasal [ŋ] where Parisian French has nasalized vowels and no final nasal consonant ([savɔ̃]~[savɔŋ] "soap"). More generally, there is a cross-linguistic tendency for the unmarked place of articulation for coda nasals, and perhaps also for stops, to be velar. In four experiments, we explored why the cross-linguistically unmarked place for the excrescent nasal is velar. The experiments test Ohala's (1975) acoustic explanation: that velar nasals, having no oral antiformants, are acoustically more similar to nasalized vowels than are bilabial or alveolar nasals. The experiments also tested an explanation based on the visual phonetics of nasalized vowels and velar nasals: velar nasals having no visible consonant articulation are visually more similar to nasalized vowels than are bilabial or alveolar nasals. American English listeners gave place of articulation judgments for audio-only and audio-visual tokens ending in nasal consonants or nasalized vowels. In the first and second experiments, we embedded recorded tokens of CVN (N = /m/, /n/, or /ŋ/) words in masking noise and presented them in audio-only and audio-visual trials. We also synthesized "placeless" nasals by repeating pitch periods from the nasalized vowel to replace the final consonant in CVm with nasalized vowel. These stimuli provide a direct test of Ohala's acoustic explanation of coda velarity in nasals. The third and fourth experiments extended these results with tokens in which the last portion of CVN (N = /m/, /n/, or /ŋ/) and Cx̃ syllables were obscured with masking noise. These experiments were designed to force listeners to assume the existence of a final consonant and to rely primarily on visual cues in a more direct test of the visual similarity of nasalized vowels and velar nasals. Taken together, the results of these four experiments suggest that excrescent coda nasals tend to be velar because nasalized vowels are both acoustically and visually similar to velar nasals.

Paper with Keith Johnson and Christian DiCanio, UC Berkeley Phonology Lab Annual Report, 529–561.

  (212) 998-7958

Room 305
  10 Washington Pl
  New York
  NY 10003