In Scientific Integrity at the Borderlands of Inquiry (AITSE Newsletter), Jonathan Bartlett (our own johnnyb) notes,
Science moves forward not by enshrining, but by doubting and improving on, the consensus opinion. It moves because a scientist says, “Hmm…. that’s funny” or “I wonder if…” Whole avenues of knowledge have begun this way.
Heliocentrism started because Copernicus wondered what it would look like to model the planets with the sun as the center. Throughout Copernicus’ lifetime his model of the planets was inferior to the models given by the geocentrists. However, what would astronomy be like today if Copernicus had yielded to his critics and abandoned his model because it didn’t work as well as the geocentrist view?
This question gets to the very heart of how scientific integrity works at the borderlands of scientific inquiry. New ideas should not be squelched, but neither should bad ones be perpetuated. Every offbeat idea should not be proclaimed as if it were the next Copernican revolution, nor should the fact that the greatest ideas in history often started by majorly rewriting what was thought to be scientifically true be ignored. It is a fact that scientific progress can start with data that appears to be at odds with the current understanding. Only later is it possible to reinterpret the old evidence to fit the new paradigm.
A good example is two major anomalies to do with Newton’s theory of gravity: the motions of Uranus and Mercury.