In composition studies, the most influential statement of what Brian Street calls the autonomous model of literacy is the work of Walter Ong. Ong bases the foundation of this model on the research on cognitive development done by Alexander Luria, a student of Lev Vygotsky. Ong finds in this research, carried out among the Islamic people of Uzbekistan in the Soviet Union during the 1930s, empirical evidence for his contention that literacy actually causes fundamental changes in human cognition. The weakness of this interpretation is that it fails to consider the context. Ong's reading ignores not only the ethnocentrism and the propaganda value of Luria's research but also the political ideology incorporated into the experiment. In addition, Ong's autonomous reading allows him to dismiss as "elaborate Marxist scaffolding" the theory underlying Luria's project. This theory of the relationship of language, society, and individual thought leads to a richer explanation of the Uzbeki data than either Ong's neutral autonomous reading or Luria's rudimentary Marxist interpretation. This alternative explanation has implications for our own literacy instruction, for it is consistent with David Bartholomae's findings about the written language of freshmen attempting to move into the academic discourse community. (Twelve references are appended.) (MS)