This paper critically explores 'involvement' as an analytic category in sociolinguistics and discourse analysis, identifying problems associated with its nature and locus and presenting a number of ethnographic studies assessing the usefulness of the notion in understanding the relationship between linguistic practices and their cultural and social contexts. Involvement as based on Western views of interaction, emotionality and personhood doesn't adequately account for the reported interactional dynamics, as these require an additional examination of cultural and social dynamics.