Abstract

The Obligatory Contour Principle (OCP) has generated much literature in phonology, however its history includes unjustified extensions. While applications of the OCP involve calculating segmental identity, the formulation of the OCP itself obfuscates the notion of segmental identity, thus allows unprincipled analysis. Alternative to the appeal to the OCP, two attempts have been made to tackle identity calculation. Feature Algebra by Reiss (2003) introduces quantificational statements in structural descriptions, enabling phonology to refer to arbitrarily similar segments. Bakovic's (2005) Optimality theoretic constraint dynamics focuses on the interaction between NoGem and Agree and treats partial/sufficient identity to be an emergent property.
Bakovic's model makes a typological prediction that languages that exhibit avoidance of partial geminates must have independent assimilation processes with respect to the features that are irrelevant to segmental identity. However, this predication is shown to be untenable. Formalization of NoGem brings to light that quantificational statements will be required in effect for the calculation of segmental identity.