Life of a Lab Rat

Funny thing, science. Most of us do experiments to show that something is so, but in reality we're excluding other possibilities. What we're doing, if we think about it, is trying to disprove hypotheses so that a theory is strengthened. Now that sounds a little odd, but it is the basis of the scientific method.

An observation leads to a hypothesis explaining that observation. We test that hypothesis by doing an experiment. We think of another hypothesis, and test that one. And we keep doing this, until we run out of testable hypotheses, to formulate a theory that is not contradicted by evidence, one that we say is 'supported' by experiment. Philosophically, it is difficult to be certain that any explanation of a phenomenon really represents reality, that is it is difficult to prove anything This is because someone with a good imagination can always come up with an explanation that fits the evidence but does not agree with your pet theory. It is much easier to disprove something.

We have Ockham, of course, but that is a philosophical concept and it although it might say something is likely or not, it does not tell us that something is. This is why good experiment design is important.

And yes, this is also why proponents of Intelligent Design do not make good scientists and and why it is bloody difficult to reason with them. They do not have any hypotheses in support of their position that are falsifiable. You can not prove them wrong, because everytime they are in a corner, puff of smoke! Dear old William can just go whistle.

But I did not want to argue about that today. I want to talk about proof, and faith.

A lot of people, and especially scientists and other rationalists, get very confused when talking about faith, and indeed proof. Is faith really a belief in the the improbable, the illogical or even the irrational? If you have evidence, does it stop being faith? If you have evidence is it necesarily proof? Surely you can only have faith in things that can not be proved, or that you know are not true? A few weeks ago someone opined that anyone who could have faith in the intangible was stupid. I wondered aloud where that left scientists, because let's face it friends, we all have faith in things we can not prove. Our experiments, our theories; we can only support from evidence, we can only prove, so much.

I came across an article from the other side of the Divide, as it were, on absolutism and relativism. This caught my eye:

A distinction needs to be made here between proof and evidence. When we state that a certain position cannot be absolutely proven, and therefore must to some degree be accepted on faith, it does not mean that there is no evidence for that position. Faith is not necessarily blind and unsupported by evidence, nor is it, as H. L. Mencken described it, 'an illogical belief in the occurrence of the improbable.' If this were true, this would be a very short book, for this is a book about evidence. Faith can be supported by evidence.

To say that we cannot absolutely prove that the universe started in a tremendous explosion called the Big Bang (which we cannot) does not mean that there is no evidence to support this theory. There is a great deal of evidence to support the Big Bang theory. This evidence, while it is strong enough to convince the scientific community, does not absolutely prove that the Big Bang occurred. Some faith is required to believe in the Big Bang. It may not require very much, but it does require some.

The entire article is worth a read, if you're into comparative religion. But if you are not, then maybe you especially should take this message home:

Much of comparative religion consists of loosely correlating shallow views of various faiths, views that would generally be unrecognizable to their adherents.

Because it is easy to set up and knock down straw men. It is much more difficult, yet so much more constructive, to find out what the other guys really think and their reasons for doing so before going off half-cocked. It applies to rival scientists arguing over the same problem as much to evolutionist versus creationist debates. Perhaps then we can move away from the same old, tried arguments that do not convince anyone. And perhaps we can realize when something is a lost cause and not worth bothering with. That works, too.

You never know, knowing the limits of proof and the limits of faith might make you a better scientist. Finding out what is known (reading the Literature! Agh!), more importantly what is unknown; discovering how far the current evidence takes us and where the authors let faith push them the rest of the way . . . why, I bet you can come up with some experiments straight away. Turn that faith into proof, or as near as you can.

Happy Christmas.