Reviewing sonority for word‐final sonorant+obstruent consonant cluster development in Turkish

The purpose of this study is to investigate the acquisition patterns of sonorant+obstruent coda clusters in Turkish to determine whether Turkish data support the prediction the Sonority Sequencing Principle (SSP) makes as to which consonant (i.e. C1 or C2) is more likely to be preserved in sonorant+obstruent clusters, and the error patterns of sonorant+obstruent coda clusters. Data from 350 typically developing monolingual Turkish speaking children (aged 2.0-5.11 years) were used in this study. Six types of sonorant+obstruent clusters were targeted for analysis of deletion, vowel lengthening, and substitution processes in C1 and C2 singletons of clusters separately. The results showed that the acquisition of Turkish clusters does not adhere to the SSP as indicated by a higher percentage of C1 errors than C2 errors. In Turkish, vowel lengthening occurred when C1 was deleted. To account for these, it is proposed that the syllable structure of Turkish plays a role in the acquisition process of sonorant+obstruent clusters. It is argued that Turkish children acquire moraic structure at an early age, as evidenced by vowel lengthening.