Two styles of theorizing:

From Eros to Gaia, Freeman Dyson, 1992:

>There is one fairly obvious way of getting a new theory. Keep close to the experimental results, hear about all the latest information that the experimenters obtain, and then proceed to set up a theory to account for them. That is a more or less straightforward procedure and there are many physicists working on such lines, competing with one another, and it might develop somewhat into a rat-race. Of course it needs rather intelligent rats to take part in it. But I don’t want to speak about this method of procedure.

There is another way in which a theoretical physicist may work which is slower and more sedate and may lead to more profound results. It does not depend very closely on experimental work. This consists in having some basic beliefs and trying to incorporate them into one theory. Now why should one have basic beliefs? I don’t know that I can explain that. It’s just that one feels that nature is constructed in a certain way and one hangs onto the idea rather like one might hang onto a religious belief. One feels that things simply have to be on these lines and one must devise a mathematical theory for incorporating the basic belief.

These two styles of theorizing are well known in the history of science. Historians call the first style Baconian and the second Cartesian. Our young colleagues today, with less awareness of their place in history, are accustomed to call the two styles “bottom-up” and “top-down.”<