Maggie's Farm

Which shows that academic conservatism, the resistance to overthrowing established paradigms, has been intrinsic to scholarship for centuries. If you are to overthrow an existing paradigm, you have to show why your own theories can explain the old observations, while also predicting new observations that distinguish the new theory. Indeed, the first rational step of most any research project is finding out what has already been learned.

Consider Einstein's Theory of Relativity. When it was shown that the orbit of Uranus was anomalous (i.e. different than what was predicted by Newtonian Theory), an attempt was made to explain it by looking for a planetary body that might be causing the effect. That search led to the discovery of Neptune—further confirmation of Newtonian Theory. When it was shown that the precession of the perihelion of Mercury was anomalous, a similar attempt was made to explain it by looking for a planetary body that might be causing the effect. But no such body could be found! So, did they dispense with Newtonian Theory? Of course not. Newtonian Theory explained so many phenomena, that they simply modified the theory to be Newtonian Theory with an anomaly of unknown origin.

Later, when Einstein proposed his General Theory of Relativity, he showed how his Theory could explain not just the Mercury's orbital anomaly, but also all the other observations consistent with Newtonian Theory. Then, he made a prediction from his Theory that light passing near the Sun's limb would be bent more than expected by Newtonian effects alone. Since then, the predicted bending of light around massive objects has been repeatedly confirmed.

All models are wrong, but some are useful. — George E. P. Box