Several phonologists have argued in recent years against constituent structure in phonology with a variety of arguments, claiming instead that all phonological relations are linear. Restricting itself to syllable structure, this paper claims, first, that the linearists necessarily disregard the conceptual evidence which we have in favour of constituency and headedness in phonology. Further it shows that the primary empirical pieces of evidence in favour of linearism are not valid, and that several other empirical arguments can be adduced against it. Finally, it is argued that by taking the linearist arguments seriously before rejecting them, we have improved our understanding of the nature of the syllable.
The Linguistic Review publishes high-quality papers in syntax, semantics, phonology and morphology within a framework of Generative Grammar and related disciplines, as well as critical discussions of theoretical linguistics as a branch of cognitive psychology. The journal welcomes reviews of important new monographs in these areas, dissertation abstracts and letters to the editor.