Despite the validated EIL status, native English accents are commonly regarded as the most appropriate pronunciation-teaching norms, while nonnative accents are still considered inappropriate. This study attempts to explore Korean-speaking elementary teachers’ perceptions of native and nonnative English accents and find out the extent to which the teachers’ ratings of the accents show variance and correlations as regards likeability, familiarity, intelligibility, comprehensibility, and instructional model suitability. For these, 53 in-service Korean-speaking elementary teachers were asked to evaluate two native (American and British) and two nonnative (Filipino and Korean) English accents by the five variables. The results were as follows: 1) The AmE accent was rated most positively, while the FiE accent was least favorable in all the variables; 2) The KoE accent was evaluated to be more familiar, intelligible, and comprehensible than the BrE accent at a significant level. In contrast, the two accents exhibited an insignificant difference in likeability and suitability; 3) Significant correlations were obtained among the variables except for comparisons of likeability-intelligibility, likeability-comprehensibility, and familiarity-comprehensibility; 4) The instructional model suitability had a significant correlation with the four remaining variables. Based on these results, pedagogical implications and implications for teacher education were suggested, followed by this study’s limitations.

L2 learners tend to encounter and use morphological derivatives more frequently as their L2 skills develop. To pronounce the morphological derivatives correctly, L2 learners have to be aware of the phonological changes occurring in the derivatives with affixation. The four phonological rules (stress shift, vowel laxing, consonant coalescence, and vowel reduction) in English, applying to morphological derivatives, are examined with respect to the entities that the rules affect (stress, vowels, and consonants); the number of rules (only one and more than one rule); and rule interaction (interacting and non-interacting) in L2 phonological acquisition. This study reveals that 1) the accuracy of consonant coalescence is highest although its significance with respect to the accuracy of stress shift and vowel laxing varies according to the number of rules; 2) stress shift and vowel laxing show different order in accuracy depending on the number of rules; 3) the three rules (stress shift, vowel laxing, and consonant coalescence) exhibit distinct behaviors with respect to the number of rules; 4) in the interacting rules, the accuracy of the fed rule (vowel reduction) is significantly lower than the feeding rule (stress shift) while in the non-interacting rules, the accuracy of one rule appears to have no influence on that of the other rule. Based on the results, implications for teaching pronunciation of the derivatives are suggested.

A foreign accent in a second language (L2) may be caused by nonnative-like pronunciation of suprasegmental elements such as stress as well as segmental elements of the L2 phonological system. In the literature of L2 phonological acquisition, however, most studies have focused on the segmental features and only a few studies have investigated the L2 acquisition of suprasegmental elements. This study examines the acquisition of English word stress by adult speakers of Korean in an attempt to see how they learn English stress patterns, particularly if they treat nouns and verbs differently and show sensitivity to the internal syllable structure with respect to stress assignment or if they treat English word stress entirely as a lexical phenomenon. For this, 51 Korean university students were assigned the production and perception tasks in which they were instructed to produce 35 monomorphemic nouns and verbs classified into 7 classes according to stress patterns and listen to them to mark on which syllable they perceived stress to be. It was revealed in this study that although they had no knowledge of stress placement associated with lexical category, vowel weight and the extramatricality of word-final consonants in English, they showed sensitivity to the effect of coda consonants, treating open and closed syllables differently. Based on the results of this research, pedagogical implications are suggested for the teaching of English word stress.