How should knowledge be analyzed? Compositionally, as having constituents like belief and justification, or as an atomic concept? In making arguments for or against these perspectives, epistemologists have begun to use experimental evidence from developmental psychology and developmental linguistics. If we were to conclude that knowledge were developmentally prior to belief, then we might have a good basis to claim that belief is not a constituent of knowledge. In this review, I present a broad range of developmental evidence from the past decade and discuss some of the implications it has for the proper analysis of knowledge. The orthodox perspective from the developmental literature was one where children fail to understand belief and knowledge concepts until later in childhood (around 4–5 years of age), with typical asymmetries in belief attribution and knowledge attribution. But what emerges from both a discussion of newer findings and a contextualization of older findings is a picture of development whereby core competence with belief and knowledge concepts emerges much earlier than previously thought (in the first or second year of life) that apparent failures in later childhood may be explained by other aspects of development than conceptual development and that there is no clear evidence that knowledge attributions emerge earlier than belief attributions.