الفهرس | Only 14 pages are availabe for public view |
Abstract The present study encompasses three main aims: first, it presents a standardized acoustic database of the basic allophones of colloquial Egyptian Arabic CEA and establishes the vowel space for each group of the allophones, including 43360 measurements for F0, F1, F2, F3 presented with two scales: Hertz and Bark, accompanied by audio files and images of the natural speech sample. Second, it fits in Arabic within the universal map based on acoustic data in the context of Dispersion Focalization Theory DFT (Schwartz et al, 1997) which joins both, Quantal theory QT (Stevens, 1972, 1989) and Adaptive Dispersion Theory ADT (Liljencrants and Lindblom 1972; Lindblom, 1986) to establish the convenient structure of CEA according to Crothers’ typology (1978). Third, it provides acoustic evidence for some of the controversial issues concerning the vowel system of CEA. The assumptions posed by DFT and its implications are investigated by having evidence from the present study. Hence, five hypotheses summarizing the DFT implications are discussed: The first hypothesis supposes that ”large vowel inventory results in a large acoustic vowel space and vice versa”. The second hypothesis states that ”acoustic vowel space is inversely promotional with vowel inventory size ”; the smaller the vowel inventory is, the greater the acoustic dispersion per vowel type is, and vice-versa. The third hypothesis proposes that” Point (focal) vowels are resistant to contextual effects”. The fourth hypothesis suggests that ”Point vowels occur at stable regions in vowel space across languages”. The fifth hypothesis proposes that” Point vowels have less intra-category variability with respect to the non-point vowels”. To investigate these hypotheses, four experiments were conducted. Each experiment investigates one type of vowel allophones ordered as plain, emphasized, pharyngealized, and unstressed vowels; this is in addition to cross-dialectal and language comparisons between the present study and six other studies with small and large vowel inventories, via calculating vowel space areas VSAs for each study. Dispersion per vowel type also is calculated using phonR language package. One hundred and forty speakers (50 men, 50 women, and 40 children) were asked to read 51 words including the context hVd in the first and fourth experiment and CVd in the second and third experiments, where C is an emphatic or pharyngeal consonant in the second and third experiments respectively. Most of the major trends related to CEA vowel system are well captured by the DFT, where the current study approves of three hypotheses implicated in DFT and disagrees with two of them. More cross-dialect and cross-language comparisons are recommended, as the researchers can use the data of the present study in the future to compare it with different languages having different vowel inventories and therefore, attaining extra explanations to vowel universals tendencies for estimating theories that explain vowel inventories. |