Argument

The past five years have seen revisions of what had been regarded as the single definitive reconstruction of Eudoxus’ homocentric sphere theory. Crucial to these revisions, which present alternative constructions for the moon (Mendell 1998a, 1998b) and the five planets (Yavetz 1998), is a careful assessment of Simplicius’ testimony regarding the Eudoxan theory. The purpose of this paper is to show that even though we can account for lunar motion by means of two distinct homocentric models, both models involve forced and ambiguous readings of Simplicius’ account. Both reconstructions adequately accommodate the lunar month and the 18.5-year precession of the lunar nodes in a homocentric model that fits the general spirit of Eudoxus’ approach to the problem. Both are also in full accord with Aristotle’s short, general account of the Eudoxan system. However, since neither of them can be naturally reconciled with Simplicius’ text, and since we know of no other appropriate alternatives, it seems to follow that Simplicius’ account may suffer from serious technical inadequacies, and that only limited historical weight can be given to his testimony regarding the details of the system.