Restrictions on Consonant Sequencing in Arabic Triliteral Roots

Consonant sequences in first and second position (C1C2) in Arabic triliteral roots (C1C2C3) were investigated to determine which consonants occur in C1 and C2 positions. The study also examed whether phonetic features, such as place of articulation, manner of articulation, and voice impose co-occurrence restrictions on C1C2 sequences within the root, and whether these restrictions are gradient or categorical. To that end, a total of 4,738 Arabic triliteral roots were extracted from two well-established Arabic-Arabic dictionaries; Assihah and Al-Ayn dictionaries and investigated. This study is grounded in the general synchronic (descriptive) theory of phonology which investigates sounds and their behavior at a specific stage during the development of a language with an eye on the gradient harmony of adjacent root consonants as modeled by the harmonic model of linguistic well-formedness. Roots were extracted and grouped alphabetically according to the quality of the consonant in C1 position. Then phonetic variations of the C2 consonant were compared with those of C1. Percentages of these co-occurrences were calculated and classified in tables. The data analysis showed more symmetrical restrictions on C1C2 sequences in relation to place of articulation compared to manner of articulation and voicing which had a minor role in restricting consonants co-occurrences. These restrictions were found to be gradient rather than categorical. More similarities between the consonants imposed more restriction on their occurrence in root sequences.