Assessing Multilingual Repertoires
This project focuses on multilingual assessment, and aims to describe the degree of bilingualism and the repertoire in each of an emergent bilingual’s languages without using the deficit model prevalent in public school standardized testing. The goal of the assessment is to describe vocabulary, grammar, fluency, and comprehension skill sets in both/all of a speaker’s languages, including under-valued language varieties, and will validate phenomena relevant to bilingual language practices such as code-switching, domain-specific language use (e.g. home vs. school), and a perceived disparity between productive and receptive skills.
This project was supported by a grant from the Advanced Research Collaborative at the CUNY Graduate Center in the fall 2015 semester.
A listener’s ability to distinguish a pair of phonemes has been shown to be based on the acoustic properties of the phonemes, and where the pairs are acoustically similar, the native language of the listener (Best, 1995; Johnson & Babel, 2010). Theories of speech perception such as the theory of Contrastive Analysis (Weinreich, 1953; Lado, 1957) and the Perceptual Assimilation Model (Best, 1995) make predictions about whether listeners will find it “easy” or “difficult” to differentiate contrasts that include non-native phonemes, but do not currently make predictions regarding the relative ability to distinguish acoustically similar phoneme pairs.
This study examines the relative degree to which listener groups from three different language backgrounds differentiate acoustically similar pairs of sibilant and anterior non-sibilant fricatives, specifically English /f, θ, s, ʃ/, Korean /s, s*, ɕ, ɕ*/, and Polish /f, s, ʂ, ɕ/. Perceptual similarity judgments from listeners from English, Korean, and Polish language backgrounds are collected using a binary forced-choice discrimination task to measure whether listeners respond that a pair is the same or different to a degree statistically significantly different from chance. In addition, a Visual Analog Scale is used with the same pairs to capture a more nuanced similarity judgment. Based on previous experiments exploring perceptual similarity (Best, 1995; Johnson & Babel, 2010), all listeners are expected to distinguish acoustically different fricative pairs to a relatively large degree, and the degree to which acoustically similar fricative pairs are distinguished is expected to be based on the native language, such that English listeners will distinguish anterior non-sibilant fricatives /f/ ~ /θ/ to a higher degree than Korean or Polish listeners; Korean listeners will distinguish tense vs. lax distinctions for the pairs /s/ ~ /s*/ and /ɕ/ ~ /ɕ*/ to a higher degree than English or Polish listeners; and Polish listeners will distinguish the pair of posterior sibilants /ʂ/ ~ /ɕ/ to a higher degree than English or Korean listeners. Additionally, listener responses will be analyzed to answer questions about which non-native contrasts listeners distinguish to a greater or lesser degree; for example, whether English listeners will distinguish tense/lax contrasts to a greater degree than posterior sibilant contrasts.
This work has been presented at:
- SUNY New Paltz Linguistics Colloquium. April 10, 2015.
- CUNY Phonology Forum – Conference on Multilingual Phonology at the CUNY Graduate Center. January 15, 2015.
 /s/ and /ɕ/ indicate lax consonants while /s*/ and /ɕ*/ indicate their respective tense counterparts.
Using static palatography, I conducted an articulatory study of the coronal obstruents of S’gaw Karen. I identified the precise place of articulation for the coronal obstruents and analyzed the data within an emergent feature theory framework. Recently, I’ve examined vowel effects on static palatography. I give workshops to graduate and undergraduate students to demonstrate the way to collect and analyze palatograms and linguograms. See my static palatography website here: https://charcoaland.commons.gc.cuny.edu.