Kentish Vowel Mergers in Old English: An Optimality-Theoretic Analysis

In Kentish, one of the Old English (OE) dialects, by the 10th century front vowels /y(:)/, /ø(:)/, /æ(:)/ were merged with /e(:)/ and disappeared from the vowel inventory. The goals of this study is to explore the Kentish front vowel mergers in OE and to show how they can be explained within the framework of Optimality Theory (Prince & Smolensky 1993/2004, McCarthy & Prince 1993, 1995). In this study, I propose two analyses of the same data and compare them. One of them is based on dispersion theory (Flemming 2002, 2004, 2006). The proposed dispersion-theoretic analysis predicts that Kentish front vowel mergers would occur to increase the distinctiveness of the front vowel contrasts. This is correctly captured by the interaction of MAXIMIZE CONTRASTS and sets of MINDIST constraints. The analysis, however, fails to explain how /y(:)/ became /e(:)/ but not more faithful /i(:)/. By contrast, the other analysis treats the individual vowel mergers in terms of markedness and faithfulness constraints. It predicts the Kentish front vowel mergers would occur to eliminate the marked front vowels in the inventory. In addition, it is shown that the proposed analysis can explain how /y(:)/ became /e(:)/ by utilizing constraint conjunction. A comparison of the two analyses leads us to the conclusion that the Kentish front vowel mergers can be better explained by the interaction of the markedness and faithfulness constraints rather than by that of the constraints on contrasts.