Poster: The acquisition of the English /s-z/ voice contrast by learners of three L1s

Introduction

English has a high functional load voice contrast between /s/ and /z/, which is active word-initially (sing /s/ - zing /z/), word-medially (fussy /s/ - fuzzy /z/) and word-finally (rice /s/ - rise /z/). In this study we investigate the acquisition of this contrast /s/ - /z/ by FrenchL1 (FR), Northern ItalianL1 (NI) and American SpanishL1 (AS) learners.

Given the status of [s] and [z] in their L1s, and based on predictions of current L2 phonology acquisition models (SLM ¹ and MDH ²), we expect that the three groups will show different voicing patterns for these sounds.

/s/ and /z/ are phonemes in FR

casse → /kas/

case → /kaz/

obligatory allophones in varieties of NI: [z] before voiced sounds, [s] in front of voiceless consonants (cf. ³), yet voicing contributes to oppose minimal pairs such as

casa → /ˈkasa/ → [ˈkaza]

cassa → /ˈkasːa/ → [ˈkasːa] (where the primary distinctive feature is length)

AS only has /s/, but partial or total voicing can happen in various contexts (esp. syllable coda) due to non-obligatory voice assimilation with the following consonant, esp. in casual speech [4].

los gatos → /los ˈgatos/ → [los̬ ˈgatos]

mismo → /‘mismo/ → [‘mis̬mo]

So, based on the SLM and despite potential difficulties encountered by all learners, we expect that (i) FR and NI learners will exhibit no difficulties in producing distinct realizations for /s/ and /z/; (ii) AS learners will exhibit difficulties in producing distinct realisations for this contrast. Additionally, based on MDH, we expect that (iii) AS learners will have difficulties producing /z/ (universally and cross-linguistically more marked) rather than /s/, and that (iv) AS and NI learners will show difficulties with /z/ word-finally (more marked than word-initially, and non-existent in NI).