Word processing has been widely endorsed as one of the most promising uses of microcomputers in the elementary school curriculum. This article reviews the burgeoning literature that pertains to word processing and writing in elementary classrooms by constructing five major propositions that cut across individual studies and methods. Together, the propositions provide a schema for understanding and categorizing what is known and still needs to be known about word processing and young writers, a framework for probing the significant theoretical and substantive issues underlying the findings, and a point of departure for discussing the most provocative themes and questions that emerge from many fields of study. Throughout the review, I demonstrate that using word processing for writing in individual classrooms is a practice that is social as well as technical. I argue, therefore, that we cannot determine how word processing is most effectively used in classrooms apart from the ways particular teachers work in particular instructional contexts and that we cannot understand how word processing affects the quality, quantity, or processes of children's writing apart from the ways these are embedded within, and mediated by, the social systems of classrooms. I conclude that, although we know a great deal about the capacities of word processing to influence students' writing, we know much less about the ways this technology is actually introduced and used in school settings over time or the ways these introductions and settings interact with the social processes of classrooms.