Production of gutturals by non-native speakers of Arabic

Abstract

This paper investigates the production of Arabic gutturals by native (NSs) and non-native speakers (NNSs) of Arabic. A total of 40 participants, 20 NSs and 20 NNSs were recruited. 240 tokens were collected using two major methods: free speech and nonsense word testing. Using PRAAT software (version 6.1.01), the tokens were analyzed acoustically to measure F1 and F2 and to signal the (non)significance of the difference between the target groups and auditorily to rate gutturals’ production accuracy by NNSs. F1 and F2 of the vowels neighbouring the gutturals were normalized using the speaker extrinsic Labov ANAE method (NORM version 1.1) to eliminate the effects of gender and age. The study demonstrates some important findings: in terms of quality, the F1-F2 approximation varies by nativeness in that NNSs were unable to make enough coarticulatory effects associated with Arabic gutturals. This result indicates that NNSs do not make a sufficient primary constriction in the posterior regions of the vocal tract. Relying on auditory judgments of accuracy, the most accurately produced gutturals by NNSs were the voiceless glottal fricative /h/ followed by the voiceless glottal plosive /ʔ/, and the lowest ranked gutturals were the voiced velar fricative /ɣ/ and the voiced pharyngeal fricative /ʕ/. The study concludes that non-temporal cues especially F1 and F2 are essential correlates to Arabic gutturals’ production. Because such factors are language-specific, they should be taken into consideration in the teaching of Arabic as a second/foreign language.