A reliable diagnosis of speech sound disorders (SSDs) in any language requires the availability of assessment tools following international linguistic and psychometric guidelines (Hua, 2006) and normative data on speech development for that language (Fabiano-Smith, 2019). To date, these criteria still need to be fulfilled for Saudi Hejazi Arabic (SHA). Normative data on phonological development that is theoretically and developmentally reliable is still missing for SHA-speaking children.
This thesis aimed to describe the typical phonetic and phonological development in monolingual SHA- speaking pre-schoolers. Speech samples were collected cross-sectionally using a newly designed, linguistically controlled picture naming test (SHAPA) and a phone–imitation (stimulability) task from 235 SHA-speaking children in Saudi Arabia, aged 2;6–5;11 (7x6–month age bands). SHAPA includes 151 words to elicit urban Hejazi Arabic (UHA) and Modern Standard Arabic (MSA) phoneme inventory. All samples were phonetically transcribed using Phon (Rose & MacWhinney, 2014).
SHAPA was evaluated regarding its construction criteria, items’ familiarities and suitability as a clinical assessment tool. The speech data were analysed regarding 1) the age of SHA-speaking children acquire and master the Arabic (UHA and MSA) phonetic system (PCC, PCC-R, PVC, PCCC, C and CC structures’ age of acquisition), and 2) the occurrence (Tokens and Types) of phonological patterns typically for SHA-speaking children, and those phonological variants that occur in children’s speech but not frequently enough to be counted as phonological patterns (InfVar). Two cut-off criteria (≥4 and ≥6) were applied to distinguish InfVar from phonological patterns.
SHAPA successfully elicited a sample for phonetic and phonological analysis since its construction requirements nearly met the international construction criteria. However, it was a lengthy test with a low incidence of spontaneous naming. All quantitative and qualitative measurements revealed a general developmental tendency, with an acquisition rate almost identical to prior Arabic research.
Phonetic analysis indicated that SHA-speaking children achieved an average score of >80% for PCC and PCC-R by age 4;0, and they approached 90% by age 5;0. They acquired all shared UHA and MSA consonants (i.e. 75% accuracy) by age 3;11 and mastered them by age 4;5 (i.e. 90% accuracy), except for postalveolar /ʃ/ and /ʒ/. The stimulability task revealed that the phonetic inventory of the children in the oldest age group was not completed (for both UHA and MSA phones) because some phones were difficult to be imitated (e.g. the emphatic consonant and the trill /r/).
The number of Tokens, Types, and InfrVar decreased steadily with age. Twenty-seven patterns were identified in children’s speech, with only four phonological patterns and one phonetic distortion existing across all age groups using both cut-off criteria. While all structural simplifications were overcome by 4;11, certain systemic simplifications and one phonetic distortion were still found when the upper cut-off criterion was used (i.e., fronting of /ʃ, ʒ/, devoicing, vowel substitution, and sibilants distortion /s, z, ʃ, ʒ/).
These findings highlighted that children’s phonetic and phonological development was not completed by age 5;11. The theoretical and clinical implications of the results were discussed.
