Demonstrative systems in the Highland East Cushitic-Gurage contact zone

International audience Our paper is concerned with the typology of demonstrative systems in Highland East Cushitic (HEC) and Gunnän Gurage (Ethiosemitic) languages, two adjacent language groups which constitute a sub-area of the Ethiopian Linguistic Area (cf. Zaborski 1991; Crass & Meyer 2008; Crass & Meyer 2011). Based on data from published grammatical descriptions (Crass & Meyer 2007; Meyer 2010; Treis forth. a, forth. b) and own fieldwork, we investigate whether prolonged language contact with HEC could explain the relatively elaborate demonstrative system in Gunnän Gurage visà-vis the other Ethiosemitic languages outside the sub-area. For this, we compare the exophoric use of demonstratives in individual languages of the contact zone and contrast them to genetically related, but geographical more distant languages farther away. The comparison is made in terms of deictic dimensions, morphosyntactic types and ontological categories expressed by distinct demonstrative terms. Demonstrative systems are categorized, firstly, according to the number of deictic distinctions, e.g. whether a language has only a two-term system distinguishing between (speaker-)proximal and (speaker-)distal deixis, or whether the system is more elaborate. We observe here that demonstrative systems in the sub-area are often more elaborate systems than those of other related languages outside the area. Certain Ethiosemitic languages only make a two-way distinction, whereas Gunnän Gurage has an additional term for medial distance. Gunnän Gurage is thus more similar to HEC such as Hadiyya and Sidaama. Next we base our comparison on morphosyntactic criteria, and following Diessel (1999: 57–58), we distinguish (i) pronominal demonstratives, (ii) adnominal demonstratives, (iii) adverbial demonstratives, and (iv) identificational demonstratives, which are used in copular and non-verbal clauses. Certain languages use the same type of demonstrative in all four syntactic contexts, while others distinguish between several formally different demonstrative types, e.g. demonstrative pronouns vs. demonstrative determines vs. demonstrative adverbs vs. demonstrative identifiers. Finally, we also take the content dimension into account and investigate whether the contact languages have dedicated demonstratives for certain ontological types of referents. We are here especially interested in manner demonstratives (see König & Umbach 2018). Our study reveals that languages of the sub-area commonly possess one or more unanalyzable manner demonstratives, while languages outside the area tend to use periphrastic forms in this function.