Calendrier

• Keren Rice (University of Toronto)

dans le cadre de l'axe Description, Typologie, Variation

‘Dismantling’ in a polysynthetic language? A Dene (Slavey) perspective

In Athabaskan languages, the verb word is complex, with prefixes generally divided into two types, conjunct (largely functional) and disjunct (largely lexical). It is agreed that the former are older additions to the verb word than the latter; it is also agreed that the latter as well as the former are part of the verb word.
Disjunct and conjunct prefixes are distinguished in several ways in the Dene (Slavey) language complex, but, in most varieties, there are ways in which their patterning is not distinct. The Fort Good Hope (FGH) variety exhibits an interesting and unexpected pattern, with developments that render the disjunct prefixes more distinct from the conjunct prefixes than in related varieties of the language. In particular, in other varieties, there are conditions under which a nasal conjunct prefix can be realized as nasalization on a disjunct prefix; this is not found in FGH. In addition, in FGH, verb stem tones appear on the syllable preceding the stem, with a new tone arising when the stem tone is displaced onto a conjunct prefix but not when it is displaced onto a disjunct prefix. Finally, vowel hiatus is resolved in FGH in one way if one of the vowels is in a disjunct and the other in a conjunct prefix and in a different way when both are in conjunct prefixes; other varieties do not clearly distinguish between the source of the vowels and allow for hiatus in environments where it is not found in FGH.
I ask why, given the tendency towards incorporation across the family and, in addition, the general lack of evidence for “dismantling” in polysynthetic languages, there is a tendency in FGH to strengthen the distinction between conjunct and disjunct prefixes, with a kind of dismantling. I explore two avenues. The first is linguistic. Given phonological developments in FGH, nasalization crossing the disjunct boundary would lead to allomorphy with some disjunct prefixes; by disallowing conjunct nasals to be realized on disjunct prefixes, such allomorphy is reduced. Different types of hiatus resolution and tone creation depending on the type of prefix also distinguish these prefix types more clearly. In FGH, there is a bias towards greater phonological distinction between disjunct and conjunct prefixes, at the expense of the integration into the verb word found in other varieties. That the “dismantling” affects disjunct prefixes alone points to the importance of structure, and the awareness of structure in the verb word. The second perspective is sociocultural. There is a long history of relationships between the Dene of FGH and the Gwich’in. While Dene varieties are similar, Gwich’in underwent many changes that make it quite distinctive. From a socio-cultural perspective, perhaps contact between FGH Dene and the Gwich’in was a factor in leading to the further differentiation of the disjunct prefixes from the conjunct prefixes, making lexical material more highly salient.

Bénédicte Pivot & Michel Bert : « Is orthography creation unavoidable for postvernacular languages ? The case studies of francoprovençal and rama languages revitalisation. »

Suite à une double communication donnée en juillet 2014 lors de la 4e Conférence sur les Langues En Danger de Cambridge, nous présenterons la trame de l’article que nous préparons pour l’ouvrage à paraître 'Creating Orthographies for Endangered Languages', Cambridge University Press (M. Jones ed.). Notre réflexion s’appuie sur l’analyse de deux situations de revitalisation de langues en danger, celle du francoprovençal en France et celle du rama au Nicaragua. A partir d’exemples concrets issus de nos travaux de terrain, nous questionnerons la nécessité d'établir une graphie standard pour la revitalisation de telles langues, que nous considérons comme « postvernaculaires ». Nous décrirons les pratiques d’écriture des différents acteurs et leurs choix en matière de graphie, en les situant dans leur contexte partagé d’obsolescence linguistique, mais se différenciant nettement par les représentations sociales liées à l’écrit et la norme. Ces deux contextes s’inscrivent dans des logiques idéologiques et pragmatiques qui influencent les choix et les stratégies de graphie/standardisation pour lesquels il nous est parfois demandé de nous positionner.

J’introduirai la régression linéaire d’une façon générale, ce qu’elle cherche à établir et les éléments auxquels prêter attention quand on utilise cet outil statistique (normalité des résidus etc.). Dans un 2ème temps, je montrerai comment utiliser le logiciel R pour réaliser des régressions et vérifier leurs conditions d’application. Et si on a le temps, je parlerai un peu des effets fixes et des effets aléatoires (pour les modèles de régression avec effets mixtes)

Discussion des séances et travaux à venir dans le cadre de l'Atelier Typologie Sémantique, animée par Maïa Ponsonnet et Anetta Kopecka

Jorge ROSES LABRADA: "Extended subject agreement on the Piaroa verb and its origin" Animate subjects are usually cross-referenced on the Piaroa verb by means of a set of word-final suffixes. In future verb forms, however, subjects are additionally cross-referenced using either a set of prefixes or a set of inner suffixes that immediately follow the root. This realization of subject marking by more than one morpheme constitutes an example of extended--in the sense of Matthews (1972)--agreement, a phenomenon that is cross-linguistically rare. In this talk, I attempt to show how this pattern of extended subject agreement results from a diachronic change whereby a non-verbal predication construction entered the verbal predication domain and came to replace the inherited subject marking strategy (i.e., two sets of affixes, one set prefixal, the other suffixal) in all tenses except for the future.