ABSTRACT
These are some personal views on distinctive feature analysis acquired during the course of more than thirty years since my early cooperation with Roman Jakobson and Morris Halle. Most of what I have to say is neither new nor radical and has been expressed in more detail in earlier publications (e.g., Fant, 1973), containing several articles on the subject (e.g., Fant, 1969, 1970). A recent article (Fant, 1983) on Swedish vowels provides additional perspectives. The following is a summary and an attempt to relate feature theory to the topic of this volume.
