EDGE features and French liaison

The sandhi phenomenon of consonant liaison in French has received a great deal of attention in pedagogical grammars as well as in more theoretical linguistic work. For the most part, theoretical studies have addressed the phonological aspects of liaison, focusing on the issue of syllabification at word boundaries or various mechanisms of deletion, insertion, or suppletion. See Klausenburger (1984) and Encreve (1988) (ch. 3) for a chronological overview of this research. Liaison is not a purely phonological phenomenon, however. The elements that trigger liaison cannot always be identified based on their phonology, and the elements targeted by liaison do not always have phonologically predictable forms. Furthermore, liaison is not necessarily realized at every word boundary where it is phonologically possible. It is subject to a wide range of lexical, syntactic, and stylistic conditions, as well as to the influence of speakers’ conscious metalinguistic knowledge about the phenomenon. This combination of factors gives rise to a very diverse and variable set of facts, a situation not fully acknowledged in most (normative) descriptions of French. This paper presents a descriptive overview of liaison, giving an idea of the scope of the phenomenon and possible approaches to its analysis. As for the contextual conditions on liaison, in many cases, the traditional notions of obligatory and prohibited liaison do not reflect speakers’ actual behavior. It turns out that general syntactic constraints cannot determine the systematic presence or absence of liaison at a given word boundary (contrary to the proposals of Selkirk (1974), for example). At best, specific constraints can be formulated to target particular classes of constructions. To express such constraints, I propose a system of