Phonotactic blocking through structural immunity

These are all phenomena in which phonological generalizations are defied. Quite different approaches have been taken to them in the past. For strict cycle effects, conditions such as the Strict Cycle Condition (Mascaró 1976, Kiparsky 1982) have been proposed; lexical exceptionality has been handled by morphemic exception features (e.g. Chomsky & Halle 1968); geminate inalterability has inspired the Linking Condition (Hayes 1986) and Uniform Applicability Condition (Schein & Steriade 1986). Each of these various theoretical devices is targeted at specific types of phonotactic immunity. None extends to either of the other two types. By contrast, Structural Immunity is a general approach that has been applied to all three of these phenomena. Structural Immunity is the method of prespecifying (either lexically, or via the grammar) strings with structure that blocks the application of phonological alternations or phonotactic conditions. The idea of handling nonderived environment blocking using structural immunity is due to Kiparsky (1993). It has since been applied to geminate inalterability (Inkelas & Cho 1993) and lexical exceptionality (Inkelas, Orgun & Zoll 1997). The goal of this paper is to show that Structural Immunity, a time-tested technique, fares better than recent proposals couched within Optimality Theory (OT; Prince & Smolensky 1993) in handling nonderived environment blocking and in connecting that phenomenon to other types of blocking. In addition, we shall see that the structural approach leads to a new perspective on the “Richness of the Base” hypothesis in OT.