Challenging the syllabic model of ‘syntax-as-it-is’

In this paper I respond to recent claims by Carstairs-McCarthy (1998, 1999, 2000) that the structure of the syllable formed an evolutionary model for the structure of the clause. I focus on the possible parallels between syllable structure and sentence structure, and consider the plausibility of syllabic origins for the clause on the basis of how successful the proposed parallels are. I suggest that the properties of modern syntax proposed by CM as evidence for the syllabic model do not in fact support it, whilst other properties can be seen as counterevidence. Many crucial features of modern syntax, such as complement structure and subcategorization, embedding, and movement of constituents, cannot, in any case, be accounted for under the ‘syllabic model’, and notions which are central in descriptions of the syllable, such as the sonority hierarchy and phonotactic restrictions, have no counterpart in the syntax. I further argue that the proposed similarity in hierarchical structure between syllable and clause – which is central to the syllabic model – appears purely superficial when examined in detail.