Monday, September 26, 2011

Was Einstein Wrong?

Einstein's famous equation E=MC2 is among the fundamental aspects of modern physics that would have to be revised if it turns out other scientists can replicate the recent finding of a sub-atomic particle that travels faster than the speed of light. But if C is off, it means that all nuclear physics has to be recalibrated, says scientist Michio Kaku, theoretical physics professor at City College of New York.

Modern physics is based on two theories, relativity and the quantum theory, so half of modern physics would have to be replaced by a new theory.

One almost hesitates to mention it, but a similar spirit of strict adherence to the limits of all scientific theories applies elsewhere. "Scientific" claims not based strictly on the limits of experimental evidence are matters of dogma, not science.

Dramatic "climate change" has happened often in Earth's geologic history, and only recently would it even have been possible for human activity to affect, much less cause it. Critical skepticism about the amount of human-caused climate change is not non-scientific or anti-scientific, but a recognition that we do not have clear or ample data to measure changes occurring over geologic time.

Similarly, one can accept random evidence of genetic mutation, the geologic age of the Earth and many other biological occurrences as compatible with science. But one might question the completeness of the theory that one can explain the evolution of all life on the planet, occurring only by random processes. That is not anti-scientific or non-scientific. It is only skepticism about a theory that claims complete explanations on partial evidence.

It remains to be seen whether other independent tests can confirm the CERN findings. But if the findings are confirmed, we will have a clear example of the fact that "science" is bounded by those claims which are confirmed by repeatable experimental evidence. Where experimentation is not possible, we ought to treat our theories as that: theories. When new evidence doesn't fit our theories, we must change our theories.

No comments:

Costs of Creating Machine Learning Models is Up Sharply

With the caveat that we must be careful about making linear extrapolations into the future, training costs of state-of-the-art AI models hav...