[grey_whiskers]

People were predicting solar eclipses long before and in the absence of knowledge of Newton's laws of motion or universal gravitation. The empirical observations were what enabled the laws to be formulated.

Perhaps it would soothe your ruffled feathers if I suggest people can predict eclipses from a dynamical model as RWP pointed out, but they can also predict them by categorization (e.g. "almanacs" ?) :-)

Cheers!

[aruanan]

Perhaps it would soothe your ruffled feathers if I suggest people can predict eclipses from a dynamical model as RWP pointed out, but they can also predict them by categorization (e.g. "almanacs" ?) :-)

You overlooked the point of my original comment. "Categorization" is the observational data from which models are derived, something I never disputed.