A multi-dimensional, multi-functional and multilingual account of discourse marker variation

Discourse markers (henceforth DMs) are the focus of a very rich field of study, investigating their many forms and functions in various languages. However, they are still rarely studied onomasiologically, especially in spoken multilingual data, as opposed to the bulk of contrastive case studies. This presentation aims to analyze the variation in use and functions of a broad bottom-up selection of DMs across three languages from different typological families, namely French (Romance), English (Germanic) and Polish (Slavic). Such an endeavor requires not only to overcome issues of definition and delimitation of the DM category, accounting for the diversity of their forms in different languages through an operational tertium comparationis (Krzeszowski 1981), but also to design an annotation model encompassing their full functional spectrum, in the perspective of spoken discourse analysis. Our study follows a corpus-based methodology based on Crible & Degand’s (in press) multilingual annotation scheme for functions of (spoken) DMs. The functional taxonomy distinguishes between four domains (ideational, rhetorical, sequential, interpersonal) that may be combined with eleven functions (e.g. cause, contrast, topic-shift). This taxonomy with two independent levels has been applied to spoken unplanned dialogues in the three languages (approx. ca. 30 minutes; between 5000 and 6000 words), resulting in the identification of 286 DMs in English (30 types), 442 DMs in French (35 types), and 847 DMs in Polish (48 types). The annotations were extracted for contrastive analyses of distribution and variation of DMs and their functions. The results indicate that the multilingual annotation scheme may be validly applied to the three different languages, demonstrating that semantic equivalence of DMs attested in different languages does not necessarily lead to functional and distributional similarities between them (e.g. in the case of English you know, French tu vois and Polish wiesz). Currently the annotation scheme is tested on additional languages (Slovenian, Spanish, L2 English, Brazilian Portuguese)