Graphic geminates as diacritics in the Orrmulum (12th cent.) and in Thomas Sheridan's General Dictionary of the English Language (1780)

International audience ; This poster compares the representational logic and phonological principles behind the doubling of consonants in Orrm's manuscript and in Thomas Sheridan's “system for respelling”. In the Orrmulum, graphic geminates do not represent phonological geminates, but act, in certain contexts, as a means “to reflect/project a short vowel for the vowel graph that immediately precedes” (Anderson & Britton 1997:300), unambiguously fulfilling a “diacritic function” (Murray 1995:127). Similarly, Sheridan makes use of ostensibly ambisyllabic consonants in his transcriptions in a manner that is clearly not linearly segmental (Pouillon & Ballier 2013). In both cases, information is given about the preceding vowel, namely, that it is short; see for instance Orrm's and Sheridan's for affability.This comparison sheds light on the evolution of the cognitive status of the short vowel, from epilinguistic intuition to proto-phonological representation. Though he applies it more consistently than his predecessors (Mokrowiecki 2012), Orrm was not the first to make use of such a device – perhaps in a complex relation to still recent “sound changes [.] such as Homorganic Cluster Lengthening and Closed Syllable Shortening” (Mailhammer 2007:37). By the fifteenth century, consonant doubling after short vowels became the orthographic norm (Scragg 1974:50). Sheridan applies the same logic in his respellings; a key distinction, however, is that he explicitly refers to syllable boundaries, and extends the doubling of consonants in his transcription system to words that do not have double consonants in the spelling. We aim to describe this (pre)-conceptualization of CVC.