Korean-English Bilinguals Research Articles

Interaction of Native- and Second-Language Vowel System(s) in Early and Late Bilinguals

The objective of this study was to determine how bilinguals' age at the time of language acquisition influenced the organization of their phonetic system(s). The productions of six English and five Korean vowels by English and Korean monolinguals were compared to the productions of the same vowels by early and late Korean-English bilinguals varying in amount of exposure to their second language. Results indicated that bilinguals' age profoundly influenced both the degree and the direction of the interaction between the phonetic systems of their native (L1) and second (L2) languages. In particular, early bilinguals manifested a bidirectional L1-L2 influence and produced distinct acoustic realizations of L1 and L2 vowels. Late bilinguals, however, showed evidence of a unidirectional influence of the L1 on the L2 and produced L2 vowels that were "colored" by acoustic properties of their L1. The degree and direction of L1-L2 influences in early and late bilinguals appeared to depend on the degree of acoustic similarity between L1 and L2 vowels and the length of their exposure to the L2. Overall, the findings underscored the complex nature of the restructuring of the L1-L2 phonetic system(s) in bilinguals.

“I know the word, but ...”

Abstract Bilinguals sometimes report on difficulties in finding words while speaking in the first language (L1) or the second language (L2), which is frequently attributed to the negative influence of one language onto the other. This paper addresses this issue by investigating the relationship between L1 and L2 vocabulary knowledge of Korean-English bilinguals who have been exposed to an L2 learning environment since the age of 12 years. Their vocabulary knowledge in L1 and L2 is measured by using a standardised vocabulary test in each language and their L1 vocabulary measures are compared with that of 12 year old Korean monolinguals. The findings show that there is a significant positive correlation between the scores from the L1 and L2 vocabulary tests among the participants. The findings suggest (1) that L1 vocabulary learning continues in the L2 learning environment, and (2) that the extent of the vocabulary knowledge in one language is the good predictor of the other, highlighting the positive role of L1 vocabulary knowledge in L2 learning. Implications for the support for L1 development of young immigrants are also discussed.

Acquisition of English word stress patterns in early and late bilinguals

Given early acquisition of prosodic knowledge as demonstrated by infants’ sensitivity to native language accentual patterns, the question of whether learners can acquire new prosodic patterns across the life span arises. Acquisition of English stress by early and late Spanish–English and Korean–English bilinguals was investigated. In a production task, two-syllable nonwords were produced in noun and verb sentence frames. In a perception task, preference for first or last syllable stress on the nonwords was indicated. Also, real words that were phonologically similar to the nonwords were collected. Logistic regression analyses and ANOVAs were conducted to determine the effect of three factors (syllable structure, lexical class, and stress patterns of phonologically similar words) on the production and perception responses. In all three groups, stress patterns of phonologically similar real words predicted stress on nonwords. For the two other factors, early bilinguals patterned similarly to the native-English participants. Late Spanish–English bilinguals demonstrated less learning of stress patterns based on syllabic structure, and late Korean–English bilinguals demonstrated less learning of stress patterns based on lexical class than native-English speakers. Thus, compared to native speakers, late bilinguals’ ability to abstract stress patterns is reduced and affected by the first language. [Work supported by NIH.]

Effects of word class differences on L2 pronunciation accuracy

This study examined the effect of word class (nouns vs. verbs) on L2 pronunciation accuracy of Korean–English bilinguals. The participants were 192 adult immigrants whose age of arrival (AOA) in the United States ranged from 6 to 23 years. Transcriptions of their productions of English sentences indicated that they were more accurate in pronouncing verbs than nouns. Similarly, the results of a grammaticality judgment test revealed that they were more accurate in detecting incorrect formulations of verbs than nouns. These effects were significant in late L2 learners (AOA 12–23). The results were interpreted to reflect the influence of the linguistic structure of Korean (where the verb is more prominent than the noun) on learning English as an L2.

Pronunciation proficiency in the first and second languages of Korean–English bilinguals

This study examined pronunciation proficiency in both the first (Korean) and second (English) languages of bilinguals. The participants were adult immigrants whose age of arrival in the USA ranged from 1–23 years. English and Korean sentences were rated by native listeners to obtain measures of pronunciation proficiency. English pronunciation of participants with ages of arrival of 1–5 years was close to monolinguals, heavier accents were noted as ages of arrival increased from 6 to 23 years. Korean pronunciation of participants with ages of arrival of 1–7 years was distinctly accented, while those with ages of arrival of 12–23 years were rated the same as monolinguals. Participants with ages of arrival of 1–9 years pronounced English better than Korean, whereas the reverse was true for ages of arrival of 12–23 years. Overall, the results were more consistent with the view that deviations from native pronunciation result from interactions between the languages of bilinguals rather than with the view of a maturationally defined critical period for language learning.

The effect of L1 use on pronunciation in Quichua–Spanish bilinguals

This study investigated the interaction of the L1 and L2 systems in bilinguals by assessing the effect of L1 use on L1 and L2 production accuracy. A novel design feature of this study is that it examined bilinguals who used their L1 on a regular basis in a bilingual setting: Otavalo, Ecuador. Thirty native Quichua speakers who were matched for age of Spanish acquisition were recruited to form three groups differing in self-reported L1 use. The three groups repeated aurally presented sentences from their L1 and L2. Monolingual listeners from each language rated the blocked, randomly presented sentences for degree of foreign accent. For the Spanish sentences, the group with the highest L1 use had stronger Quichua accents than the group with the lowest L1 use. On the other hand, L1 use had no effect on the ratings of the Quichua sentences. Results from an analysis of Korean–English bilinguals are also reported. These results replicate the finding that L1 use affects L2, but not L1 production. These findings indicate that the interaction of the L1 and L2 systems affects the success of L2 acquisition, providing evidence that factors other than neurological maturation influence L2 acquisition.

Pronunciation proficiency in L1 and L2 among Korean–English bilinguals: The effect of age of arrival in the U.S

Pronunciation proficiency in the first (L1=Korean) and the second (L2=English) languages of bilinguals who had emigrated to the U.S. was examined. The focus of this presentation is on the effect of age of arrival (AOA) in the U.S. on pronunciation in L1 and in L2. The subject pool consisted of 240 Korean–English bilinguals classified into ten subgroups (ages 2–22) based on AOA. Two groups of monolingual subjects were also tested: 24 Korean and 24 English speakers. The bilingual subjects produced sentences in their L1 and L2 and the monolingual subjects produced sentences in their L1’s. Native listeners from each language group rated the speakers’ pronunciation on a 9-point scale. The results show a positive correlation between AOA and Korean pronunciation (0.74) and a negative correlation between AOA and English pronunciation (−0.85). In each language group and regardless of AOA, the monolingual subjects were rated better in pronunciation than the bilingual subjects. The implications of these findings to second language acquisition and first language retention will be discussed. [Work supported by NIH.]

A case study of fluent Korean-English bilingual speakers: Group membership and code choices

This paper presents a case study of code choices for six first generation Korean-English ‘full-bilinguals’. The study was designed to explore factors that would account for the code choices in specific situations (i.e. conversations with first generation ingroup and outgroup members, and second generation ingroup and outgroup members) by these Korean-English bilinguals. A review of linguistic characteristics ingrained in Korean and American English languages and cultures leads to a hypothesis that code choice and switching has to do with the type of social relationship between interlocutors. The results of the study reveal that the group membership is an important factor which determines the degree of use of one language or the other; the more distant the conversation partners are, the more they use Korean language during conversation. Conversely, the closer or more intimate the relationship they have, the more they use English. Furthermore, the results imply that the association of each language with each group membership has to do with two aspects of politeness (i.e. discernment and volition). The more distant the bilingual speakers are, the more they show the discernment aspect of politeness by producing more small-size switches respecting the Korean morpho-syntactic patterns. On the other hand, the closer or more intimate the relationship they have, the more they show the volition aspect of politeness by using more large-size switches involving English clauses.

Maturational limitations to pronunciation accuracy in a second language

This study examines pronunciation proficiency in a second language (L2) as a function of the age of (L2) acquisition. Accuracy of pronunciation of words and sentences produced by Korean–English bilinguals who began learning their L2 (English) between 2 and 24 years of age is reported. All subjects were born in Korea and had resided in the U.S. for at least 8 years. Their average age at the time of testing was about 24 years. The results are based on the subjects’ accuracy of pronunciation of vowels and consonants in English words and sentences. Preliminary findings, regarding segmental phonemes, indicate no major difficulties up to about 11 years for L2 acquisition. Those who learned English (L2) at or later than 12 had significantly lower pronunciation accuracy scores. These difficulties were more evident in vowels than in consonants. The results support the critical period for language learning hypothesis. The discussion will address the problem of specifying the optimum conditions for second language acquisition and first language retention.

Effects of Concreteness on Cross-Language Priming in Lexical Decisions

Korean-English bilinguals were shown a prime word in one language for 150 msec, followed by a word or nonword in the other language which required a lexical decision. Target words which were translations of the preceding word showed a substantial (96 msec.) priming effect compared to unrelated target words; targets semantically associated to the translation also showed a significant though smaller priming (56 msec). For both types of target and for both target languages, the priming effect was greater for concrete word-pairs than for abstract word-pairs. In contrast, a second experiment with monolinguals gave equivalent priming for concrete and abstract pairs. The results support the hypothesis of Saegert and Young and of Paivio and Desrochers that the two lexicons of the bilinguals are most closely integrated via a common, nonverbal representation.

EFFECTS OF CONCRETENESS ON CROSS-LANGUAGE PRIMING IN LEXICAL DECISIONS

Korean-English bilinguals were shown a prime word in one language for 150 msec, followed by a word or nonword in the other language which required a lexical decision. Target words which were translations of the preceding word showed a substantial (96 msec.) priming effect compared to unrelated target words; targets semantically associated to the translation also showed a significant though smaller priming (56 msec). For both types of target and for both target languages, the priming effect was greater for concrete word-pairs than for abstract word-pairs. In contrast, a second experiment with monolinguals gave equivalent priming for concrete and abstract pairs. The results support the hypothesis of Saegert and Young and of Paivio and Desrochers that the two lexicons of the bilinguals are most closely integrated via a common, nonverbal representation.