An Investigation of Language-Specific and Orthographic Effects in L2 Arabic geminate production by Advanced Japanese- and English-speaking learners

Research has indicated that second-language learners have difficulty producing geminates accurately. Previous studies have also shown an effect of orthography on second-language speech production. We tested whether the existence of a contrast in the first language phonology for length aids the second-language production of the same contrast. Furthermore, we examined the effect of exposure to orthographic input on geminate consonant production in a cross-script context. We tested the production of Arabic geminate-singleton stop consonants [/b(sic)/-/b/, /t(sic)/-/t/, /d(sic)/-/d/, and /k(sic)/-/k/], a nasal stop consonant /m(sic)/-/m/, and an emphatic stop consonant /t(sic)(sic)/-/t(sic)/, as well as the effect of the diacritic used in Arabic to mark gemination in a delayed imitation task and two reading tasks (ortho-with diacritics and ortho-without diacritics). A comparison of the productions of advanced Japanese-speaking learners, English-speaking learners, and an Arabic control group showed that both learner groups were able to produce Arabic geminate stops; however, the Japanese-speaking learners exhibited an advantage over the English-speaking learners in the auditory-only task and in the presence of diacritics, highlighting the fact that orthographic effects may occur in some cross-script contexts.