Ancient Greek science is one of the most productive topics for thinking about what science actually is and how it works. The standard narrative — empirical observation plus deductive logic equals early science — is both true and inadequate.

The Milesian school (Thales, Anaximander, Anaximenes) in the 6th century BC made the fundamental move of seeking natural explanations for natural phenomena rather than mythological ones. This is genuinely important. But the natural explanations they proposed — everything is water, everything is the apeiron, everything is air — were not empirically tested in any meaningful sense. They were plausible stories, not experiments.

Aristotle is the exception who proves the rule. His biological works are based on extensive direct observation, and the History of Animals is a genuinely empirical work. He described and classified over 500 species of animals, many confirmed by modern taxonomy. His physics, by contrast, was almost entirely a priori.

Hellenistic science — Eratosthenes measuring the earth, Aristarchus proposing heliocentrism, Hipparchus building the first star catalogue, the Antikythera mechanism — was more consistently empirical and technically sophisticated. The question of why Greek science didn't generate a scientific revolution is one of the most interesting in the history of thought.