Abstract
Many researchers in composition instruction assume that free and journal writing exclusively and necessarily produce “meaningful” writing. This is not substantiated in their limited case study research, or in the research of anyone else. We need to establish a precise definition of “meaningful” writing, determine its place in the curriculum, and determine better means of designing instruction that produces writing that is both meaningful and of high quality. The meta-analysis of Hillocks (1984) indicates that structured composition assignments produce better writing than nondirectional writing experiences. This article explores the reasons for this, and establishes hypotheses based on these reasons for developing a theory of composition instruction. The hypotheses support a need for structured instruction, rather than student-generated direction.