Recently I discussed the concept of a hypothesis and gave an example of two competing economic theories. And I explained that two different theories for the same group of phenomena might just be two acceptable ways of explaining the same thing. But often there are multiple, contradictory hypotheses or theories. How do you sort things out and find the much coveted TRUTH?

In every day usage we often ask for someone to prove their statement. But the word, proof, has multiple, related meanings. Originally, I think, “proof” meant simply to test. So 80 proof alcohol was “tested” to have so much alcohol. Chain and rope were “proof” tested to support a certain load. Engineers still use the word in this way.

In mathematics the word, proof, is used much more strictly. To prove a theorem, for instance, is to use a set of axioms, the results of other proofs and allowed rules and operations to derive a mathematical result or statement.

In science “proof” is probably too strong a word outside of theoretical or mathematical calculations. Usually a better word would be “evidence.” But we often use “proof” in a loose way, more like the original word meaning “test,” especially in those heated debates where we demand proof of a conclusion, hypothesis or even a theory. So let me use the word in that original, less precise way, as a synonym for test,  just for convenience.

How do we prove the things we believe? In other words, how do we convince others? The crux of this question is a big deal and explored in the branch of philosophy called epistemology. Modern scientists have generally accepted a limited number of ways of proving ideas, considered the “scientific method.”

But let’s examine some of the ways we might prove our beliefs. For most of human history we didn’t seem to need to prove anything. If a stranger came to town and someone got sick the stranger must have been a witch. This is quaint but we use the same “logic” today. If my wife uses a cell phone and my wife gets cancer it must have been the cell phone. If you worked at a brake shop and you got lung cancer it must have been the brake shop. If you walk into the cold weather and you get the flu it must have been that you didn’t wear a hat.

By the way, those conclusions are all examples of a common logical fallacy: post hoc ergo propter hoc, or, “after this therefore because of this.” It leads us to wrong conclusions almost every time it’s tried.

Or maybe you can read tea leaves or chicken bones to see if you should marry someone. Maybe your star chart would help. Perhaps a better approach would be to pray. Once upon a time men would fight a duel to give God an opportunity to determine who was lying and who was truthful. I’d be willing to use that method with some people today, in fact.

Proving, that is, testing a hypothesis is very difficult because it’s so very easy to fool yourself. What methods work and what methods are faulty?

In the natural sciences, physics, chemistry, astronomy, biology etc. it’s often possible to set up a controlled experiment and test a hypothesis. Experimental proof is, I think, the strongest proof. But even experimental proof is fraught with difficulty. You may do experiment after experiment with results that you think support your hypothesis. But, it’s easy to have an experiment contaminated or to misinterpret a result. It’s also easy to unconsciously reject data that does not support our pet hypothesis. That’s one reason we want experiments to be repeatable, so the hypothesis can be tested over and over and by others too.

In other sciences, particularly  the social sciences, proving a hypothesis can be even more difficult. And, often, what people consider evidence is very misleading. Moreover, because the conclusions of social sciences often matter very much to people, there is a big temptation to accept shoddy or incomplete research or misleading evidence as proof. That’s why we sense that popular health recommendations change back and forth over time.

We also want to state a hypothesis in such a way so that not only can it be supported with evidence, but also so that it can be proven wrong. As an example, there are individuals who go around telling you that they can cure disease or move objects by supernatural powers. Typically, if you put them to the test, the proof, and they fail to produce a cure or move an object or predict the future, whatever, they will explain that their powers don’t work when there is mental negativity in the vicinity. So the hypothesis that this person can cure disease can produce positive results if the patient is cured. But if the patient is not cured the hypothesis is not invalidated. A hypothesis incapable of being invalidated is not testable and not a part of science. It may be a part of religion or some other belief system but it is not science, which has to limit itself to testable hypotheses, that is, hypothesis that can be proven wrong. Someone once said that science does not tell you what is true, only what is not true. That’s because you can do 1000 experiments that support your hypothesis but it only takes one to demolish it.

In the next installment, I’ll concentrate on the multiple ways we mislead ourselves.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: