Substitution Patterns of the English Voiced Interdental Fricative by L1 Costa Rican Spanish Speakers

L1 Spanish speakers learning English struggle to produce the English voiced interdental fricative ð. This is surprising as ð occurs naturally in Spanish as an intervocalic allophone of /d/. To investigate Spanish speakers’ production of ð in English, I conducted a study with ten L1 Costa Rican Spanish/L2 English speakers. The results show that English voiced interdental fricatives were produced with target pronunciation 43.4% (214/494) of the time and were substituted in the remaining 56.6% (280/494). It was determined that target-sound substitution patterns depend on the phonological contexts in which the target sound occurs (word-initial vs. intervocalic), and the type of speaking task (reading vs. spontaneous). I interpret these results as partially due to phonological transfer: L1 allophonic information affects L2 speech production. The results also show that the realization of the segments is affected by the different experimental tasks: speakers have more target-like production in tasks where there is an opportunity for more careful pronunciation (reading task) than in more spontaneous ones. Finally, I argue that the unexpected appearance of the alveolar tap ɾ is due to an interaction between the English /d/ tapping allophone rule and the status of d and ð in the learner’s interlanguage.