Lexical and morphological conditioning of paradigm gaps

Although paradigm gaps are an analytical puzzle for all linguistic theories, until recently they have received only sporadic discussion in the literature. Since the advent of Optimality Theory, however, gaps have gradually become the subject of more systematic attention. This is certainly due in part to the particular mechanical challenge that gaps pose for OT; hence, a major focus has been on providing a mechanism for all overt candidates to be eliminated, rather than the more usual outcome of one emerging as optimal (Prince and Smolensky 2004, p. 57; Orgun and Sprouse 1999; Fanselow and Féry 2002, Raffelsiefen 2004; Rice 2005, 2006; McCarthy and Wolf [this volume]). Equally important, though, is the fact that OT provides a natural way to formalize an intuition expressed in some earlier discussions, that gaps often appear when the expected form would violate a surface-true phonotactic constraint and that silence is simply one part of a larger conspiracy to avoid illegal configurations (Hetzron 1975; Iverson 1981). For this reason, discussions in the OT literature have focused primarily on cases that appear to have clear phonotactic motivation, in that the expected faithful candidates would involve illegal configurations like stress lapse, OCP violations, sonority sequencing violations, and so on (Orgun and Sprouse 1999; Raffelsiefen 2004; Rice 2005).