Path dependence, a model first advanced to explain puzzles in the diffusion of technology, has lately won allegiance among analysts of the politics of public policy, including health care policy. Though the central premise of the model-that past events and decisions shape options for innovation in the present and future-is indisputable ( indeed path dependence is, so to speak, too shallow to be false), the approach, at least as applied to health policy, suffers from ambiguities that undercut its claims to illuminate policy projects such as managed care, on which this article focuses. Because path dependence adds little more than marginal value to familiar images of the politics of policy-incrementalism, for one-analysts might do well to put it on the back burner and pursue instead "thick descriptions" that help them to distinguish different degrees of openness to exogenous change among diverse policy arenas.
