A multimodel inference approach to categorical variant choice: construction, priming and frequency effects on the choice between full and contracted forms of am, are and is

The present paper presents a multimodel inference approach to linguistic variation, expanding on prior work by Kuperman and Bresnan (2012). We argue that corpus data often present the analyst with high model selection uncertainty. This uncertainty is inevitable given that language is highly redundant: every feature is predictable from multiple other features. However, uncertainty involved in model selection is ignored by the standard method of selecting the single best model and inferring the effects of the predictors under the assumption that the best model is true. Multimodel inference avoids committing to a single model. Rather, we make predictions based on the entire set of plausible models, with contributions of models weighted by the models' predictive value. We argue that multimodel inference is superior to model selection for both the I-Language goal of inferring the mental grammars that generated the corpus, and the E-Language goal of predicting characteristics of future speech samples from the community represented by the corpus. Applying multimodel inference to the classic problem of English auxiliary contraction, we show that the choice between multimodel inference and model selection matters in practice: the best model may contain predictors that are not significant when the full set of plausible models is considered, and may omit predictors that are significant considering the full set of models. We also contribute to the study of English auxiliary contraction. We document the effects of priming, contextual predictability, and specific syntactic constructions and provide evidence against effects of phonological context.

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.

Quantitative data as a clue to auxiliary contraction processes

This paper is a quantitative corpus study of the variable contraction of English auxiliaries (e.g. John has ~ John’s been there all day). I examine auxiliaries after non-pronoun subjects, and find the following. First, auxiliaries often surface in an “intermediate” form, one which loses its initial consonant, but retains a reduced vowel (e.g. [?z] for has). After proposing an analysis of these forms, I examine the effect of subject weight on contraction, finding that contraction occurs less often after longer subjects. Finally, I discuss the implications of these findings for where contraction must be situated in the grammar.

Contraction of negatives as evidence of variance in register-specific interactive rules

ABSTRACTThis study investigates the contraction of negatives in a carefully chosen corpus of discourse and writing, to permit comparison of the relative influences of various linguistic and social parameters on contraction. Evidence is presented that negative contraction is conditioned by interactional and other register variables. The point is made that the pragmatic as well as morphological interpretation of negatives entails that negative contraction and auxiliary contraction should be distinguished from each other. Although a Cognitive Prominence Principle predicts noncontraction when the negative conveys semantically focal information, a Social Agreement Principle predicts contraction. This is because it would be face-threatening (and, therefore, in conversation analysis terms “dispreferred”) to focus on disagreement, which is most often the semantic information conveyed by negatives. This hypothesis is examined using corpora which differ along several dimensions. The most important of these (for this study) appear to be the interactional versus informational register dimensions (Finegan, 1994). Data from instructional (workshop presentations), confrontational (political debates), and casual conversational material are contrasted with comparable reading style materials. The following general results are predicted. The Cognitive Prominence Principle will take over in informational contexts when disagreement is acceptable or neutralized. The Social Agreement Principle will take over in more interactional contexts where disagreement is not acceptable. The results are of interest to the student of focus, the sociolinguist concerned with dialect, register, and style variation, and even the speech technician.