Zero Derivation in English: Base-Identity and Constraint Indexation

This paper argues that stress patterns that occur in zero derivation is best analyzed in Optimality Theory (Prince & Smolensky, 1993; McCarthy & Prince, 1995) in terms of faithfulness and lexically indexed markedness constraints. This approach is shown to capture distinctions between generality and exceptionality of stress patterns in zero derivation. Based on a variety of evidence, I argue that zero-derived forms should be faithful to their base, and that exceptions to the patterns are confined only to a restricted set of words. Assuming that a single constraint can be multiply instantiated in a constraint hierarchy, and each instantiation may be indexed to apply to a particular set of lexical items (cf. Fukuzawa, 1999; Ito & Mester, 1999, 2001; Kraska-Szelenk, 1997, 1999; Pater, 2000), I specifically claim that the lexically indexed markedness constraint WORD STRESSL outranks the faithfulness constraint Base-Identity, which in turn ranks above the general markedness constraint WORD STRESS. It is shown that with the constraint ranking, all the stress patterns that occur in zero derivation can be accounted for straightforwardly. In so doing, I also show that under the analysis adopted in this paper, the prediction that new zero-derived forms always retain the stress pattern of the base is borne out from the high ranking of the faithfulness constraint Base-Identity.