Abstract
This paper focuses on the topic of coherence in the dialogue of Czech speakers. When describing coherence of texts, grammar books focus thoroughly on written texts; description of spoken texts is narrowed to the specification "less coherent".
In case of spoken and written utterances, however, the authors aim to the same goal - to bring recipients complete information, closed framework, and thus fulfill their communicative intent. Coherence of spoken texts is affected by communication situation: spontaneous dialogue is interaction between speakers and all of them can participate in developing the main topic.
This paper is based on a corpus of spoken Czech ORAL2013 and focuses on coherence in three types of situation: one speaker exposes his turn, ie. it may be viewed as a monologue; speaker is interrupted by the second one who takes the floor, then the first speaker continues to his telling; and finally, all participants of dialogue produce the text (and its topic) together. The aim of this paper is to show what resources (repetition, completion of syntactic structures, intonation, particles ...) speakers use to create a coherent text.