Labialization and Loss of Middle English /x/

This paper investigates phonological developments of /x/ which took place during the transition from Middle English to Early Modern English: labialization of [x] to [f]; the loss of [x, ç]. Specifically, this paper attempts to explain why the change of word-final [x] to [f] is accompanied by the short nucleus whereas the loss of [x, ç] before word-final [t] occurs only after a long nucleus and why [ç], unlike [x], never shows a parallel labialization. Despite their insightful account of the labialization of [x] to [f] within the framework of Articulatory Phonology, Browman and Goldstein (1990) do not deal with the other questions. Gussenhoven and van de Weijer (1990) provide a meticulous account of the labialization and the loss of [x] using a feature-geometric representation, but their claim for spontaneous labialization of [x] before Middle English Breaking seems unmotivated. Following Kwon's (2003) position that frequent interactions between vowels and /x/ throughout history of English are due to the affinity of V-Pl nodes within a rhyme, this paper attempts to propose a more natural analysis of the processes at hand using a feature-geometric representation: Middle English Breaking as a fusion of V-Pl nodes in V + /x/; labialization of [x] to [f] as the result of partial decomposition of [x]; loss of [x, ç] as the result of their total decomposition.