The current thesis aims at investigating the variations of the emphatic fricative contrast s~sˁ across different spoken Arabic dialects. To achieve this aim, three main studies were conducted, acoustic, perceptual, and articulatory. In terms of the acoustics, eight Arabic dialects from the IVAr corpus were utilised to examine if the differences between the emphatic contrast s~sˁ is the same or different across these eight dialects. Multiple acoustic correlates were measured for the consonants and vowel. The findings suggest that the measures that exhibited significant information to mark the emphatic contrast s~sˁ are observed on the vowel formant information; F1 tends to rise, whilst F2 is consistently lower in the context of an emphatic fricative compared to non-emphatic. The GAMM results reveal that the majority of the emphatic contrast was observable at different points in the trajectory of the vowel following the emphatic contrast s~sˁ, and dialects differ in how large the F2 difference is and how much of the vowel is affected. The findings also reveal that other variables, such as COG and fricative duration, can somehow display difference in the emphatic contrast, though these differences are statistically insignificant.
Drawing from these outcomes, the subsequent perceptual study employed a subset of the acoustic data. F2 trajectories were manipulated based on the results from three Arabic dialects, resulting in real and manipulated stimuli. Results indicate that, in real stimuli, all listeners from each dialect could identify their own native formant cues with higher accuracy in plain condition. Crucially however, all listeners can attend to the cues to emphasis of other dialects with considerable accuracy. However, for manipulated stimuli, the F2 size and trajectory shape was observed to play a significant role for listeners to be able to identify their own cues, in emphatic condition (only). This disparity between real and manipulated stimuli suggests F2 may not be the sole cue relied upon by listeners.
Subsequently, an Ultrasound imaging study was conducted, involving fifteen speakers from five dialects, to investigate any articulatory basis for this phenomenon and whether tongue position and lip movement would distinguish the emphatic contrast. Principal Component Analysis (PCA) results reveal that tongue positioning significantly contributes to distinguishing between emphatic/non-emphatic fricatives, tongue body is retracted during emphasis and advanced during non-emphatic consonants, while there was no association of lip movement observed during the articulation of the emphatic contrast except for four speakers.
