| This dissertation investigates the academic speech of Humanities and Natural Science instructors and students in 36 German and 32 American lectures and interactional classes. It examines how English and German structural markers, questions, question tags, backchannel signals, and receipt markers contribute to variations of style in response to academic contexts, disciplines, gender, conversational roles, and varying linguistic practices in American and German academic discourse. The data are drawn from the MICASE corpus of academic speech compiled at the University of Michigan and a smaller corpus of German academic speech assembled by the author. The data analysis couples qualitative, discourse-analytic methods with a quantitative sociolinguistic analysis.; Among instructors in both cultures the factors of conversational role, academic discipline, and conversational mode - not gender - are most influential in the use of structures investigated in lectures. In the interactional class format, however, gender trends could be detected for instructors, particularly in relation to receipt markers. Students' use of the structures under investigation is influenced by their less powerful conversational role, by demands of information exchange in their discipline, and also by gender. It is argued that these results arise from discourse restrictions in academic speech, such as turn type pre-allocation, exchange and speech length restrictions, as well as knowledge building and teaching strategies in the Humanities and the Natural Sciences.; The cross-cultural comparison shows remarkable similarities when it comes to a link of several structures to conversational role and discipline. However, it also indicates several differences between American and German academic discourse. Among others, these concern differences in male gender-preferential language in the two academic cultures and a more formal and less interactive German academic style - as reflected in the use of group vocatives, read out lectures, and a lower frequency of discourse markers and questions. This dissertation thus positions the language of academia in relation to a variety of factors and shows that the instructors' and students' discourse style is dependent on complex constellations of social, contextual, disciplinary, and cultural conditions. |