Adopting the methods of multimodal conversation analysis, this study demonstrates that adult second language (L2) users with emergent literacy formulate requests as 'complex multimodal Gestalts' consisting prototypically of deictic or depictive gestures combined with gaze, situationally relevant material, and vocal or linguistic resources. The data obtained for this study are from classroom interactions and out-of-class service encounters that were conducted as part of pedagogical tasks in the context of integration training for adult L2 students with emergent literacy. The analysis shows how co-constructed request sequences typically involve multimodal negotiation in which the co-participants support L2 users to formulate requests that are understandable and that contribute to the progressivity of the interaction. The findings contribute to an understanding of practical requests as multimodal, context-sensitive, collaborative actions and underline the importance of this perspective especially from the view of L2 users with emergent literacy.