Content area
Abstract
Metathesis, a sound change involving a permutation of segments, is usually treated in the literature dealing with descriptive or historical linguistics as a sporadic sound change. By contrast this study argues: first, that metathesis is a natural phonological process arising from the phonotactic constraints of languages; and second, that in the case study of Gascon metathesis is a systematic and predictable sound change which is rule governed.
Gascon, a dialect of Occitan, is selected as the focal point of the study due to its numerous examples of metathesis and assimilation. A data base for the investigation of these sound changes is gathered from the writings of Luchaire (1881), Grammont (1909), Millardet (1920), Baldinger (1962), Bec (1968), and Rohlfs (1970). Then Grammont's analysis of metathesis in the gascon of Bagneres de Luchon is presented and criticized. Similarly the phonological theory of Chomsky and Halle (1968) as well as Foley's approach (1977) are found to have weaknesses when they are applied to the analysis of the linguistic data in Gascon. Consequently the 'new paradigm' theory advocated by Bailey (1973) is introduced as the most promising approach for the systematic treatment of metathesis.
I adapt Bailey's system in order to obtain a method that explains how, after the constraints favoring assimilation are reordered, metathesis takes place each time assimilation does not occur. The proposed method, which emphasizes the differences in syllable weight between two contiguous syllables, is found to apply to all instances and absences of metathesis in Gascon. Moreover, it produces a single rule applicable to all possibly relevant circumstances capable of predicting the occurrence of metathesis. As a result, the study terminates with strong support for the general thesis that metathesis is indeed a phonological process amenable to the methods of systematic explanation.
