Friday, May 13, 2011

Oh, That's Just a Theory!

It's often the case that people unfamiliar with or resistant to scientific undertakings dismiss peer-reviewed research by saying: "that's just a theory." As if it were an unsupported guess.

I have even said something like that myself: "I have a theory" about something. What I mean to say is, "I have a hypothesis."

A hypothesis is more than a guess. It is a supposition based on familiarity with the subject, experience, or deep thought. A proper hypothesis must be testable.

The point of testing a hypothesis is to disprove it. No hypothesis can be proven. It can only be disproven. If a proper test fails to disprove a hypothesis, the next step is to try another test. Collect more data. Give the problem more thought. Examine whether we have a case of coincidence or one of cause and effect.

Then take all the data collected, observations made, and develop a theory. The theory must be compatible with all the observed data. The theory should also be testable. If the tests fail to disprove the theory, then it may be adopted as the best explanation available, but no theory can ever be proven. It is the job of scientists to reexamine accepted theory in light of new knowledge, new methods of measurement and observation.

Theory is the best you get. There is never final certainty.

No comments: