L1 Category Compactness, L1 Allophonic Targets and L2 Production

We are glad to welcome professor Christine Shea (The University of Iowa) for a visit to our Department, and cordially invite you to this lecture.

About the Presentation: (co-authors Christine Shea and Ciara Tapanes)
In this study, we present data examining how Spanish speakers transfer the abstract syllabic linking that drives the phonetic realization of /i/ in their native language to their second language, English. In Spanish, underlying /i/ may surface as a vowel, glide, fricative, affricate, or stop ([i, j, ʝ ʝɟ ɟ]), depending upon where it occurs in the syllable, whether as syllable peak, non-peak satellite member of a diphthong or onset. English has no such linking; both glides and affricates occur in word onset and can form minimal pairs in words such as “yet” [jɛt] and “jet” [dʒɛt]. Thus, Spanish speakers must learn to represent these L1 allophones as L2 phonemes.

Our predictions are guided by the Personal Articulatory Precision Hypothesis (PAPH, Kartushina & Frauenfelder, 2014; Flege & Bohn, 2020), applied to allophones. The PAPH posits a positive relationship between L1 category compactness and speakers’ personal L2 production precision. We predict that L1 Spanish speakers with more compact L1 categories will have greater precision in their L2 productions than speakers with greater variability across different allophonic positions. Participants were 14 speakers from Querétaro, México. Spanish target words had /i/ in word-initial and medial position, stressed and unstressed syllables, counterbalanced across vowel contexts. English target words had a glide or affricate in the same positions. Items varied in spelling (<ll>, <y>, <hie> in Spanish; <y> and <j> in English). There were two tasks. In the Spanish and English Delayed Repetition task, participants heard a word or a sentence, followed by a 500ms medium-frequency beep, saw a kaleidoscope (1000ms), and then repeated the word/sentence they heard. In the Spanish minimal pairs task, participants read minimal pairs with /i/ in different syllable positions. Preliminary analysis of the Spanish data indicates high variability across speakers and within certain speakers’ production of Spanish targets. English analysis is pending.

References

Flege, J. E., & Bohn, O.-S. (2021). The Revised Speech Learning Model (SLM-r). In R. Wayland ​  (Ed.), Second Language Speech Learning: Theoretical and Empirical Progress (pp. 3–83). Cambridge University Press​  

Harris, James W. and Ellen M. Kaisse. 1999. Palatal vowels, glides, and obstruents in Argentinian Spanish. Phonology 16.117–190.

Hualde, José I. 1997. Spanish /i/ and related sounds: An exercise in phonemic analysis. Studies in the Linguistic Sciences 27.61–79.

Jaggers, Z.S., 2018. Evidence and characterization of a glide-vowel distinction in American English. Laboratory phonology, 9(1).

Kaisse, Ellen M. 2019. Glides and High Vowels in Spanish. In S. Colina and F. Martínez-Gil (eds), The Handbook of Spanish Phonology, Routledge, 145-162.  

Kartushina, N., & Frauenfelder, U. H. (2014). On the effects of L2 perception and of individual differences in L1 production on L2 pronunciation. Frontiers in psychology, 5, 105122.​  

Levi, Susannah V. 2006. Phonemic vs. derived glides. Lingua 118.1956–1978.

Levi, Susannah V. 2011. Glides. In Marc van Oostendorp, Colin J. Ewen, Elizabeth Hume, and Keren Rice (eds.), The Blackwell Companion to Phonology. Oxford: Wiley-Blackwell, 341–366.

About the Presenter:
Christine Shea is an Associate Professor in the Departments of Linguistics and Spanish & Portuguese at the University of Iowa. Her research is spurred by interest in how age, experience and the native language sound system interact with the perception and production of a second language. Adult second language learners approach their second language with a first language already in place. Her work investigates how experience with a previously acquired language affects the way learners perceive and produce a second (or third) language. Her approach looks at these issues by assuming the input signal is a source of rich information that learners may or may not be able to take advantage of when perceiving and producing their non-native language and creating the representations that support further learning.