Aristotle denies (as we have seen) that the principles of a scientific demonstration are themselves to be reached by way of demonstration (APo. 72b19–20; 93b16–18). But a demonstration (sullogismos) is a strong form of argument, in which conclusions follow necessarily from the premises (APr. 24b18–20). So some weaker form of reasoning could be involved, such as “induction” in our modern sense: a generalization over a collection of particulars. The latter gives rise to the notorious “problem of i