Contrary to assertions by Walley (Bruck and Genesee 1995) that bilinguals have heightened sensitivity to individual phonemes and are therefore better able to analyze language, our results do not indicate a significant overall difference between monolingual and bilingual performance on the phoneme tests. The data does not support the assertion that bilinguals are more advantaged than monolinguals in phoneme awareness. The data also does not support Bruck and Genesee's (1995) finding that monolinguals had an advantage. In short, in this study, monolinguals and bilinguals seem equivalent.
Durgunoglu et al. (1993) found general transfer of phoneme awareness in one direction, from Spanish to English. We also found a general phoneme awareness transfer between languages; however, according to our data, transfer is in both directions from Spanish to English and vice versa. The transfer was significant amongst the bilinguals. Transfer also tended to be evident for the monolingual, English-only and Spanish-only groups. Even though these children knew only one language, they were aware of the phonemes in words from another language, implying that one need not know meaning to support sound structure. Thus, level of performance in one language was predictive of performance in the other. Upon closer inspection of the data, some subjects who a) were monolingual and thus had little or no exposure to the second language tested and b) had high phonological awareness in their native language, scored highly on phoneme awareness tasks in the other language. This suggests a language independent effect of the linguistic units involved in phoneme awareness: individuals scoring high in the phoneme awareness tests of one language did not necessarily need to have any experience with the other language to score high in that second language.
Bruck and Genesee (1995) contended that reading instruction in a language does affect the phoneme awareness of that language, thus facilitating stronger grapheme-phoneme associations within the language. The data in this study shows language of instruction and performance on corresponding language versions of the tests are related. Children taught in Spanish performed better on Spanish versions of the test; children taught in English performed better on English versions.
However, upon closer inspection of the data, it appears that transfer occurs regardless of the language of instruction. This finding is consistent with a very recent study by J. F. Carlisle, M. Beemen, C. H. Davis, and G. Sphraim (1998), who also found cross-language transfer of metalinguistic skills between Spanish and English, independent of language of instruction.
The implications for the education of bilingual children are varied. The data indicates that in either language of instruction, English or Spanish, the children who perform the best on phoneme awareness tasks in the language of instruction also tend to perform well on phoneme awareness tasks in the other language. Furthermore, the data in this study suggests that regardless of language of instruction, phoneme awareness transfer occurs. Therefore, in principal, regardless of the language of instruction, the children who develop high phonological awareness will acquire reading with greater facility. Phonological awareness in both languages should be equivalent.
In practice, however, this does not appear to be the case. The data also shows that the children taught in Spanish-instruction classrooms do not attain as much phoneme awareness as the monolingual and bilingual children taught in English-instruction classrooms. The children in English-instruction classrooms perform better on the Spanish versions of the test than the Spanish-instruction students do on the English versions of the test.
A number of factors may contribute to this performance disparity between the children taught in English- and Spanish-language instruction classrooms. This un-equal transfer between languages may be caused by outdated curricula or methodology in the Spanish-only classrooms, which are not as effective as those in the English-only classrooms at promoting phoneme awareness. In addition, the author noted that not all teachers in the Spanish-language instruction classrooms were native speakers of Spanish. Two teachers, in particular, did appear to speak heavily accented Spanish and did not appear to have firm command of Spanish grammar or vocabulary.
Our results indicate that, because reading ability and phoneme awareness are so heavily correlated, of central concern is the facilitation of the development of phoneme awareness, regardless of language of instruction. Only when phoneme awareness is nurtured will children develop reading competency in alphabetic language (McBride-Chang 1995; Yopp 1988). It appears that the current debate in the media about how to most effectively educate bilingual children, whether in English or in Spanish, is a political one. In the case of reading, it appears that children will develop phoneme awareness regardless of the language of instruction. However, the literature states that most children, especially those children who are likely to have reading problems, need some instruction in order to develop phoneme awareness (Mann and Liberman 1984).