Naomi Nagy at ILLS

Posted on March 18, 2013 by

Next Monday (note the odd day!) there will be a special ILLS seminar, at the usual time and place (Ellen Wilkinson A2.16, 4.15pm), with Naomi Nagy (Toronto). The title: Looking for contact effects in Heritage Languages. Here’s the abstract:

Heritage languages sit in a place between established, codified languages and learner languages, exhibiting features of both. As such, they are a rich area for improving our understanding of how languages develop, both in the individual and in the community. Variationist sociolinguistic methods can be fruitfully applied to this task in order to analyze, interpret and generalize from patterns of variation across groups of speakers. The groups in contrast include different generations of heritage speakers, speakers with different acquisition trajectories, speakers with different linguistic attitudes and language use patterns, and speakers of different heritage languages. Examples of the application of variationist methodology to a corpus of Heritage Language data from Toronto will be presented.

We investigate how social factors (language use, ethnic identity, and linguistic attitudes at the individual level, and demographics at the community level) relate to type and degree of language change by recording sociolinguistic interviews that include an Ethnic Orientation Questionnaire. Responses are coded to develop indices of language usage and attitudes. Multivariate analyses of linguistic variables incorporate these social factors along with relevant linguistic constraints. When complete, the corpus will contain conversational data from 40 speakers, spanning three generations since immigration, of a range of heritage languages: Cantonese, Faetar, Hungarian, Korean, Italian, Polish, Russian, Ukrainian. To date, >200 speakers have been recorded and transcribed. In this paper, outcomes for several variables will be contrasted.

One is subject pronoun presence (“pro-drop”). Multivariate regression analyses contrasting three generations of speakers, in Cantonese, Italian and Russian, show cross-generational stability both in rates of pronoun use and in the linguistic factors constraining the variation. Individuals’ rates do not correlate strongly to any measures of Ethnic Orientation. This stability contrasts with reports on USA Spanish pro-drop (cf. Otheguy et al. 2007).

A second variable is Voice Onset Time (VOT) in voiceless stops. While bilinguals maintain phonetic distinctions in similar sounds in two languages, the languages influence each other such that there are distinctions from monolingual speech (Flege 1987). Acoustic measurements extracted from conversational speech and subjected to ANOVA show that the VOT of bilinguals in the HL drifts away from the monolingual short lag of Russian and Ukrainian, and the much longer lag of Cantonese, toward the long lag of English, while the Italian sample shows no such variability. Ethnic orientation correlates to VOT drift.

The contrast between inter-generational change for VOT and stability for pro-drop underscores the importance of examining multiple variables in multiple languages, and examining multiple facets of multilinguals’ performance.