Based on research relating William Perry's scheme of cognitive development to foreign language proficiency, a developmental hypothesis concerning language ability is proposed. This approach suggests that lower developmental stages in the cognitive domain retard the achievement of communicative language proficiency, while higher developmental stages facilitate it. Perry's hierarchical scheme, chronicling the changing ways in which individuals explain their understanding of experience, truth, and behavior, and the American Council on the Teaching of Foreign Languages (ACTFL) oral proficiency scale, also hierarchical, show some correspondence. Certain cognitive tasks appear to underlie tasks in the proficiency scale. Empirical evidence of this relationship is drawn from a French conversation and composition course in which reading, writing, and discussion-leading were designed to encourage and challenge students at different levels of cognitive development as defined by Perry. The results help explain the fossilization of learner varieties of language occurring often at the level of 2/2+. The proposed model treats aptitude not as a fixed set of capacities but as a changing interrelationship of a variety of capacities, including personal, language, diversity, contextual. (MSE)