Douglas proposes a new ideal in which values serve an essential function throughout scientific inquiry, but where the role values play is constrained at key points, protecting the integrity and objectivity of science.