This chapter first considers the functional and semantic overlap between discourse connectives and discourse markers, where the latter is presented as including the former. Since the two categories share most protypical features, the fuzzy boundaries between the two categories are explained in terms of partial overlap.

We then show that the description of connectives’ meanings and functions can proceed following an onomasiological as well as a semasiological approach. The latter has given rise to numerous case studies in a variety of languages aiming to come to a fine-grained semantic description of specific connectives. Strenghts and weaknesses of such studies are presented. Onomasiological approaches focus on a set connectives that are categorized together on the basis of shared semantic properties, trying to disentangle their similarities and differences, within and across languages. In the final section, we turn to one of the most described features of discourse connectives namely its polysemy and polyfunctionality, and how contextual cues may help solve this ambiguity, and how polysemy is a key explaining factor in the (frequency) distribution of connectives.