Leucippus, Democritus, et al get to Atomism by argument, not experiment:
- Recall Zeno: if distances or bodies can be divided infinitely, get paradoxical conclusion.
- The Atomist response: once reach atoms, can't keep cutting.
- Recall Melissus on motion: motion requires empty space, which by (PPMI) is ruled out;
- The Atomist response: since plainly motion is possible, there must be nothingness, i.e. a void.
Details of the Atomist picture:
- Something like our intuitive picture of 'microscopic ball bearings' bouncing off each other and conglomerating.
- But complications: some have different shapes, to "hook together" in different ways.
- Explain vision by sheets of atoms being shed from outer layers of bodies
Argument for infinity of atoms:
- Principle of Sufficient Reason (PSR): no reason for any finite number of atoms, similarly no reason any shape should be lacking.
- If there are infinite number, and by PSR no reason for any particular number, size, shape, etc., then some atoms should be visible.
- Such atoms would be invisible anyway, by theory of vision above, since they can't send off sheets of atoms;
- collisions and entanglements of atoms of different degrees of heaviness explains, e.g., planets, which are conglomerations of the heavier stuff.
Note that Democritus gets to this kind of skepticisim by virtue of the Eleatic rational argument approach to cosmology. What we get by rational argument undermines what we get by sensory experience. That's as opposed to contemporary atomic theory which explains and vindicates sensory experience, and thus puts no pressure on the reality of, e.g., mid-sized dry objects.