Sometimes it's just a matter of plugging in the new parameters into the same equation. Other times, the whole calculation needs to be done again. And sometimes the whole edifice must be changed.<quoted text>
Instead of simplifying assumptions, wouldn't it be more productive to drop the assumptions and start over with a clean slate? It seems logical that if you're working with the wrong parameters it would be more productive to back up and start over. Would this approach work?
We seldom go all the way back to Aristotle when re-thinking particle physics. Instead, we tend to use certain methods that have worked well in a wide variety of situations, looking for new models with additional features.
The problem with starting over *completely* is that it ignores the information we have already learned from previous observations. It is also not nearly as easy as you might think to create a theory that is consistent with everything we know *and* is completely new. Remember that the difference between the prediction of Newton's model of gravity and Einstein's model of gravity was 43 seconds of arc *per century* in the orbit of Mercury. But that was enough to overthrow Newton. THAT is the level of accuracy that was required 100 years ago. it is more now.