The Principle of Continuity
Dewey's principle of continuity, articulated most forcefully in Experience and Education (1938), is not a pedagogical recommendation but a description of how experience actually works. The child who burns her hand carries the consequence into every subsequent encounter with heated surfaces. The consequence is not a stored memory but a modification of the organism's entire orientation. Applied to AI-augmented building, the principle raises a question no productivity metric can address: what kind of experiential chain does the practice create? Does each session take up the results of previous sessions in a way that produces cumulative growth, or does each session exist in relative isolation, producing artifacts without depositing the kind of understanding that transforms future practice?
In the AI Story
The principle has two directions. Experience is cumulative forward — each session modifies the builder's capacity for the next. Experience is cumulative backward — the quality of each session is shaped by the history of sessions that preceded it. Together these give experience what Dewey called its longitudinal quality: the fact that the educational value of any present experience cannot be assessed in isolation from the chain in which it occurs.
The distinction between domain-continuous and model-continuous experience that the Dewey volume develops is a direct application of this principle. The traditional software developer's experiential chain ran through the domain of software itself — each encounter with the domain's resistance deposited understanding transferable across tools. The AI-augmented builder's chain may run primarily through the model — each session deposits understanding about the tool, valuable and genuine, but contingent on the tool's continued existence and current behavior.
This is not a pronouncement against AI. It is a diagnostic. The same builder may operate domain-continuously on some sessions and model-continuously on others, depending on conditions: whether she examines the code the AI produces, whether she tests its behavior against her own predictions, whether she encounters domain expertise in her community of practice, whether the problems she works on require her to bring understanding that the model cannot supply. Conditions shape continuity, and conditions are designable.
Dewey insisted that every educational arrangement must be evaluated by what it contributes to the longitudinal quality of experience — what it prepares the learner to do next, not merely what it enables her to do now. A productivity gain that produces a shipped artifact but deposits no transferable understanding is, in the longitudinal view, a kind of debt: the builder's future self pays the cost of the present session's shortcut.
Origin
The principle received its canonical formulation in Experience and Education (1938), written near the end of Dewey's life as a corrective to progressive educators who had taken his earlier work as license to abandon structure. Dewey insisted that not every experience is educative — some experiences are mis-educative, in that they narrow the organism's capacity for future experience rather than expanding it. The principle of continuity is the diagnostic for distinguishing the two.
Key Ideas
Every experience leaves a deposit. The deposit shapes what the organism brings to the next experience, for better or worse.
Mis-educative experiences exist. Not every activity that produces an outcome produces growth; some experiences narrow rather than expand the capacity for further experience.
The chain matters more than the link. A single session of AI-augmented work cannot be evaluated in isolation; its educational value is a function of the chain it belongs to.
Conditions shape continuity. Whether AI-augmented practice produces cumulative growth depends on designable features of the work, not on the tool alone.
Appears in the Orange Pill Cycle
Further reading
- John Dewey, Experience and Education (1938).
- Jim Garrison, Dewey and Eros (1997).
- Gert Biesta and Nicholas Burbules, Pragmatism and Educational Research (2003).