JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.Cambridge University Press is collaborating with JSTOR to digitize, preserve and extend access to Language in Society.http://www.jstor.org In a series of "reading cycles," mother and child alternate tums in a dialogue: the mother directs the child's attention to the book and/or asks what-questions and/or labels items on the page. The items to which the what-questions are directed and labels given are two-dimensional representations of threedimensional objects, so that the child has to resolve the conflict between perceiving these as two-dimensional objects and as representations of a threedimensional visual setting. The child does so "by assigning a privileged, autonomous status to pictures as visual objects" (1978: 5). The arbitrariness of the picture, its decontextualization, and its existence as something which cannot be grasped and manipulated like its "real" counterparts is learned through the routines of structured interactional dialogue in which mother and child take turns playing a labelling game. In a "scaffolding" dialogue (cf. Cazden 1979), the mother points and asks "What is x?" and the child vocalizes and/or gives a nonverbal signal of attention. The mother then provides verbal feedback and a label. Before the age of two, the child is socialized into the "'initiation-replyevaluation sequences" repeatedly described as the central structural feature of classroom lessons (e.g., Sinclair and Coulthard 1975; Griffin and Humphry 1978; Mehan 1979). Teachers ask their students questions which have answers prespecified in the mind of the teacher. Students respond, and teachers provide