Uncertainty surely plays an important role in promoting and directing cognitive growth, but most investigations of children’s reasoning and problem solving have relied on procedures that are poorly suited to reveal the degree to which individuals are certain of the answers they endorse. We examine research on children’s reasoning in the context of obvious indeterminacy in order to assess whether simple modifications of the standard assessment procedures would allow us to monitor subjective certainty more reliably. It is concluded that when we set as our goal the measurement of judgment certainty, major changes in our testing methods are required. Two promising alternatives are considered, and it is hypothesized that once these new procedures are perfected and applied, we will discover that uncertainties are both prevalent and prolonged – that most of what seems to be known at any one time is actually understood only partially and therefore with some degree of uncertainty. It is argued that this is a natural and acceptable situation. Progress requires only a relative degree of confidence and not complete certainty. Persistent, vague uncertainties, on the other hand, provide continuous motivation and sensitization for further discoveries, growth, and development.