This paper offers supporting evidence for the true, but unstylish, thesis that Western science entered Athens in the mid-fifth century b.c. like a tinker's cart, hung about with jangling and gleaming hardware. That thesis, in turn, is a vital link in an inductive proof that Western science has never actually developed apart from close interaction with measures, models, and machines.

The traditional view, common to histories of science and philosophy and to more general interpretations of classical culture, made the use of models, observation, and apparatus the crucial difference between ‘science’ in its classical and in its modern form. This did not quite match the actual documents and achievements of classical work from Archimedes through Ptolemy, and more detailed study led to a new and different consensus. Since 1951, the empirical character of post-Aristotelian science has been recognized, and explained as the result of a confluence of Near Eastern computational, observational technique and a more austere classical qualitative, speculative fondness for theory construction. The date of fusion is put at about 323 b.c., the death of Alexander the Great. On this recent view, apart from an incidental early coincidence, science in Greece before 323 would still have been the austere, anti-empirical enterprise which earlier historians thought typical of classical science in general.