Occam's Razor and Scientific Certainty

A Theory with Fewer Variables that Explains More is Better

Aug 27, 2009

The simpler the explanation the better. This theory is otherwise know as KISS. It is also the basis of Occam's razor. This is a theory about evidence and method.

What makes a good scientific theory? Most importantly, it is the ability to explain and predict. But what if two competing theories both explain and predict, which is better? That's where Occam's razor comes in.

First discussed by William of Occam, a 14th century Logician, he simply stated that "Entities should not be multiplied unnecessarily." This was a logical construct of how theories should behave. Later the philosopher Leibniz with his theory of observables and the Scientist Issac Newton, explained scientific theories as "when you have two competing theories that make exactly the same predictions, the simpler one is the better."

Let's look at an example of this at work. Suppose a commercial airliner blows up in mid flight over the continental US. Immediately, the department of Homeland Security, the FBI, and the National Transportation Safety Board is called in. Homeland Security and the FBI are convinced that this was not a terrorist bombing. The NTSB after months of investigation concludes that there was an electrical spark over an exposed fuel line that caused the explosion.

Fine. Some one says. But the person adds, "The reason that there was an explosion in the first place was because there was a fight between Gremlins and Leprechauns and they ruptured the fuel line. And that was the real cause of the explosion."

Let's look at this from Occam's point of view. First of all, one has to prove that there are such creatures as Gremlins and Leprechauns. One also has to prove that they were present in the fuel line area and one has to prove that there was a fight there, and finally that this fight actually caused the rip in the fuel line. Anyone by themselves would be hard to believe, but when combined, it is impossible to believe in the Gremlin/Leprechaun theory.

If you compare the two theories, the NTSB theory and the Gremlin/Leprechaun theory you see the difference in variables. The NTSB theory talks about fuel lines and sparks. The other theory talks about creatures, a presence in the area of the explosion, a fight in the area of the explosion, and that the fight caused rip which caused the explosion. That is two variables vs. four variables. Occam would say the theory with two variables is the better one.

A variation of Occam's razor is KISS. Keep It Simple Stupid. Another is to use the simplest explanation to fit the facts.

What does one need to have workable scientific theory and how does Occam’s Razor fit it?

2. Create a description, or hypothesis, that is consistent with what was observed. (This is where Occam’s Razor comes in.)

4. Use experiments to test or verify those predictions or observations and modify the hypothesis if necessary based on the results found. Here, can one verify the predictions or can one prove that the predictions are wrong; that the theory is falsifiable.

5. Finally repeat steps 3 and 4 to remove discrepancies between a theory and experiment and/or observation.

With this in mind let us look at the theory of gravity. Which theory is better; Newton’s or Einstein’s? Newton’s and Einstein’s both explain gravity but from a different perspective. Newton explains what happens to objects on earth and near earth: how gravity affects objects in space and time. But Einstein explains gravity from a cosmological point of view; how gravity affects space and time.

In the former, Newton’s case, it does not explain how gravity works, only how gravity affects objects. Einstein’s theory is more universal, it tries to explain how gravity works. From Occam’s perspective, Einstein’s theory explains more, and it can also be used to create Newton’s theory of gravity, so it is the better theory.

As one can see, the development of a scientific explanation or theory has several steps; the critical ones are observation, prediction, and verification. But while a theory may be good if there are other theories that explain more than others, then that theory would be better and hence preferred.

The copyright of the article Occam's Razor and Scientific Certainty in Scientific Research Methods is owned by George Garza. Permission to republish Occam's Razor and Scientific Certainty in print or online must be granted by the author in writing.