Covert contrasts in the acquisition of English high front vowels by native speakers of Korean, Portuguese, and Spanish

ABSTRACT

The purpose of this article is to report results of an investigation into the production of a covert contrast by native speakers of Korean, Portuguese, and Spanish in the acquisition of the English distinction between the high front vowels /i/ and /ɪ/. A covert contrast is a statistically reliable acoustic distinction made by a language learner between target-language phonemes that is nevertheless not perceived by native speakers of the target language. We gathered data on both the production and perception of this vowel contrast from a total of 36 second-language learners from the three native-language backgrounds, along with data from an additional 22 native-speaker controls. The findings are that about two-thirds of our second-language learners produced a covert contrast between the targeted vowel phonemes. In light of the results from research on child language acquisition and disordered speech, our data suggest that a covert contrast is not limited to the domain of the acquisition of a first language, either by normally developing children or by those with speech disorders.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ Note that the total number of target words actually presented and recorded during the experiment was more than what we report in the present study. In addition to the /i/ – /ɪ/ contrast examined in the current study, we initially elicited productions and perceptions for another vowel contrast (/æ/ – /ε/ as in pat and pet) as well. However, we decided to report findings for the other contrast in a separate paper, as the size of the data made it impossible for us to discuss all the findings and implications in depth in one paper.

² One of the anonymous reviewers questioned our criterion of requiring listeners’ above-chance identification of both categories /i/ and /ɪ/ to determine whether a participant was producing a contrast. The reviewer suggested that the requirement that the transcribers must identify the two vowel sounds, rather than simply distinguish them, may have underestimated the extent to which a native speaker could differentiate the two sounds. This, in turn, would overestimate the number of cases of covert contrast. Our motivation for requiring the above-chance identification of the two vowels is that this necessitated not only that the transcribers hear a difference between the two sounds but also that this differentiation is in line with the target language distinction. We believe that this criterion is consistent with our view that a covert contrast represents an intermediate stage of acquisition.

³ An anonymous reviewer asked to what extent the observed covert contrasts would be the result of the training. We believe that the effect of training was minimal in the present study, contrary to our expectation. The majority of our L2 participants did not change their contrast status between sessions. That is, most of the participants who produced a covert contrast in session 1 (before receiving training) continued to produce a covert contrast across sessions. Likewise, most of the participants who produced no contrast in session 1 continued to produce no contrast across sessions, although they received training sessions. Therefore, we expect to see as many covert contrasts regardless of the type of training we offered.