The Imperfect Endings in Romance

The historical development of imperfect indicative and conditional inflection in Pyrenean Romance*

“…Within the north‐western region where thematic labials occur for non‐first‐conjugation verbs, a gradient is observed (see also Posner 1961: 31–2, 38). Westerly varieties, bordering Aragón, tend to exhibit a three‐way contrast of the ‐ava ‐iva ‐eva type described in sections 2.2 and 3.1.…”

“…For Gascon varieties, the prevailing view in the literature is that three‐way contrast in the imperfect is an innovative, analogical feature (see e.g. Ronjat 1937: 173–5; Posner 1961, Coromines 1991:137, Massourre 2012: 232); Rohlfs (1935: 148–9) is unusual in treating it as conservative and etymological.…”

“…Thus, in contrast to [ɛ] in Gascon varieties, there is no immediate historical precedent for the presence of /e/ in third‐conjugation imperfect forms. A more plausible assumption, initially advanced by Posner (1961), is that modern ‐ iva /− eva systems develop from earlier ‐ iva systems via analogical introduction of /e/ from other TAM categories of third‐conjugation verbs. For Aragonese, where [e] tends to raise to [i], the reintroduction of [e] may also be favoured by a process of hypercorrection, i.e.…”

“…19 The exact vowel quality of forms with orthographic <e>, <ie> and <eve> is not specified: as the orthographic systems used during the mediaeval period do not explicitly indicate either stress placement or vowel aperture, orthographic <e> may represent [e], [e] or in some cases [ə]. 17 A notable example is that of Posner (1961), who focuses solely on 'Bearnese' varieties, i.e. those spoken in the historical province of B earn, corresponding to the eastern part of the modern d epartement Pyr en ees-Atlantiques.…”

“…Reflexes of *essere 'be' typically present a unique series of imperfect forms continuing Latin ERAM, etc., the development of which falls outside the scope of this study.3 Labelling conventions vary across the grammatical traditions of the varieties discussed in the study: for Catalan and Aragonese, the e-conjugation is traditionally labelled II, and the i-conjugation III; for Occitan, while some authors adopt the Catalan convention, most literature follows the French convention in which the i-conjugation is labelled II and the e-conjugation III.4 This explanation, though consistently repeated in historical grammars, is known to be problematic(Posner 1961: 18, Togeby 1964, Esher 2018). Based on observed crosslinguistic tendencies, dissimilation would be expected to affect the first /b/ of HAB EBAM, etc.…”