Phase cycles, φ-cycles, and phonological
This paper addresses the question of how phases interact with phonology. I argue that phonological operations can affect representations that are “inactive” in the phase-theoretic sense, but only in limited ways. Specifically, I hypothesize that only <i>noncyclic</i> phonological processes can violate “phase impenetrability”. I suggest further that it is the integrated linear nature of (late) PF representations that allows for this part of phonology to override phase domains. A further hypothesis concerns phase cycles and the cycles of cyclic phonology: I argue that phase cycles and phonological cycles are distinct, and that the latter are triggered on a morpheme- or exponent-specific basis.