Note 34 — Chapter I · THE CONTRIBUTION OF THE ANCIENT WORLD

These proofs were worked out by Aristotle in two ways (De caelo, II, 14). First he explained that physical laws require that the earth must be spherical; then he demonstrated that observation shows that it actually is a globe. Aristotle’s physics were built upon a theory that superficially has been compared with the Newtonian theory of gravitation, although fundamentally it is entirely different. A principal law of Aristotelian physics is that all heavy bodies seek the center of the universe, whereas Newton’s law is that all bodies, whether heavy or light, attract each other (see Duhem, Système, vol. i, p. 210). Aristotle (De caelo, II, 4) showed by mathematical argument that water, in obedience to his physical law, will, if unhindered, become a perfect sphere, with the center of the universe as its center, and that land, though it cannot become a perfect sphere owing to its rigidity, will tend to assume such a form.

That the earth actually is a globe, the Stagirite maintained, is revealed by the circular shadow it casts upon the moon in an eclipse. Furthermore, a traveler journeying from north to south sees new constellations appear above the southern horizon and vice versa, constellations that could only be hidden from him at his starting point by the curvature of a spherical earth (Duhem, Système, vol. i, pp. 211–215).

Adrastias of Aphrodisias, one of the Peripatetic school, adduced proofs similar to those of Aristotle (Duhem, op. cit., vol. i, pp. 473–474), although he presented them with greater clarity. He showed by the argument of the appearance of new constellations to a traveler journeying north or south that the earth is convex from north to south. That it is 369also convex from east to west he proved from the observation that the same celestial body rises sooner in the eastern parts of the habitable world than it does in the western. This could be demonstrated by any eclipse of the moon: the eclipse appears at a later hour of the night and higher in the heavens to an observer in the east than it does to one in the west. As both observers see the same eclipse, it follows that the moon must in reality rise in the east before it rises farther west. If the earth were flat both observers would necessarily see the eclipse at the same hour of local time.