What is wholeness? To answer this question, it is helpful to present a specific setting. Imagine someone not yet recognizing it asking, “what is roundness?” We might try to answer him by giving a number of instances, such as “the moon is round,” “the plate is round,” “the coin is round,” and so on. Of course “round” is none of these things, but by adducing a number of such instances we may hope to provoke in him the recognition of roundness. This happens when his perception of the specific instances is reorganised, so that they now become like mirrors in which roundness is seen reflected. In spite of what many people might think, this process does not involve empirical generalization — i. e., abstracting what is common from a number of cases. The belief that concepts are derived directly from sensory experiences is like believing that conjurors really do produce rabbits out of hats. Just as the conjuror puts the rabbit into the hat beforehand, so the attempt to deduce the concept by abstraction in the empiricist manner presupposes the very concept it pretends to produce.