Numerous theories of development have utilized a stage concept which has been discussed as both an empirical and a logical construct. From the empirical perspective, the notion of stage has been attacked and largely dismissed (Gelman & Baillargeon 1983: Brainard, 1978). As a logical construct, it is difficult to explain how a simple logic can be causally related to a more complex logic (Fodor, 1980). The present paper suggests that this argument may be misguided. Ambiguity is inherent, for the stage concept is not a causal determinant. Stage, as a formal abstraction, cannot depict behavior, rather it serves to increase understanding in terms of the function it serves. Here an examination of Wittgenstein's (1953) concept of "family resemblance" is informative in understanding how the stage concept might be used. From this perspective the problem with the concept results from an attempt to view a stage as a cause of specific behavior. Instead, stage should be construed as an abstraction which explains certain similarities among behaviors without requiring a single common characteristic.