A focus on "beauty" or "elegance" is, for starters, just an updating of Occam's Razor. The simplest theory that fits the facts is assumed best, and people pick

A focus on "beauty" or "elegance" is, for starters, just an updating of Occam's Razor. The simplest theory that fits the facts is assumed best, and people pick different ways of expressing the notion of "simple". There's nothing wrong with that as a starting point. Problems can arise, however, when you take that further and make simplifying assumptions. Economics has been particularly prone to that problem; I believe the famous can opener joke dates to the 1960s. (http://netec.mcc.ac.uk/JokEc.html http://netec.mcc.ac.uk/JokEc.html gives the simplest :) form of the can-opener joke I know: A physicist, a chemist and an economist are stranded on an island, with nothing to eat. A can of soup washes ashore. The physicist says, "Lets smash the can open with a rock." The chemist says, "Lets build a fire and heat the can first." The economist says, "Lets assume that we have a can-opener...") By way of contrast, simplifying assumptions in physics have usually gone better. In particular, relaxing them has gone fairly smoothly ("Assume all the mass is concentrated at a point of zero diameter ... oops, I mean in a small sphere of finite diameter ...") On the other hand, one of the core critiques of string theory is based on the suspicion that a key simplifying assumption (only the early terms of certain Taylor series matter) is badly mistaken. And in climate science, predictions are necessarily based on simplified models; debate can then ensue as to whether or not the models are misleading.