The Old English Letter Wynn <ƿ> as the Labial Approximant [ʋ]: A Diachronic and Synchronic Analysis from the Perspective of Optimality Theory

Old English (ca. AD 450-1100) constitutes the only stage in the historical development of the English language, at which the standard spelling conventions prescribed a graphemic differentiation between the back vowel [u] and the back approximant [w]/[ʋ]. While the former has always been spelled with versions of the Latin letter <u>, the scribes of Old English manuscripts introduced the originally runic character wynn < w>/<ƿ> to mark the latter. The present dissertation examines the issue in a transdisciplinary fashion, combining in a novel way the research tools offered by current models of generative phonology, statistical analysis in the programming environment R, and Latin and runic paleography. By the same token, the dissertation introduces insightful developments into the methodology of diachronic phonology, which has been somewhat neglected hitherto, in comparison with synchronic phonology. The phonological representations are couched in Feature Geometry (Sagey 1986; Halle 1992, 1995), while the derivations follow the practice of Optimality Theory (Prince & Smolensky 1993; McCarthy & Prince 1995). The combination of the two models allows for an explanatorily adequate reconstruction of a [+consonantal] labial approximant [ʋ] in Old English. The results of the phonological analysis are confirmed and further enriched by means of a statistical analysis, conducted in two models: Multiple Correspondence Analysis, and Generalised Linear Model. The statistical analysis demonstrates how generative phonology constitutes a highly accurate research methodology, and one that can be used for corpus studies – despite the increasingly common accusation that it is excessively abstract. Furthermore, statistics reintroduces time-depth into the diachronic phonological considerations by providing precise dating of the particular stages in the consonantisation of Old English [ʋ]. Finally, the results are verified by a paleographical study of how the rune wynn was introduced into the Latin scripts used by the Anglo-Saxons. The paleographical part of the dissertation is pioneering, because the letter wynn has not yet attracted scholarly interest, which traditionally focuses on letters used for writing Latin. The dissertation shows how paleography can ground diachronic phonology in historical reality by reflecting details of phonological development in minute characteristics of letter shapes.