Thursday 12 March 2009

Do Scientists Dream of Black Sheep?

Recently Tom Harkin, a US senator who founded the National Center of Complementary and Alternative Medicine (NCCAM), proclaimed that the Center had failed at one of its aims:


One of the purposes of this center was to investigate and validate alternative approaches. Quite frankly, I must say publicly that it has fallen short. I think quite frankly that in this center and in the office previously before it, most of its focus has been on disproving things rather than seeking out and approving.


Now, this statement betrays a very fundamental lack of understanding of the most basic tenets of science. The idea of science is to try to falsify things. By complaining that this is what has been done, Harkin is displaying the true dishonest reasons for the Center's existence.

This is a complaint that can be heard a lot among pseudo-scientists or paranormal practitioners. In a documentary shown on Channel 5 last year, Derek Ogilvie, the Baby Psychic, made a similar complaint after failing a test set up by James Randi (with a protocol to which he had unreservedly agreed beforehand, I might add). Ogilvie laments, "these tests are built for people to fail," as though he is revealing the methods of a crooked coconut shy.

Of course the truth is he is right. All scientific tests are built to be failed, because it is by these tests that something can be learnt. Science isn't designed to confirm whatever beliefs or biases are previously held by the tester, it is designed to find out whether a hypothesis is true; whether it conforms to reality. If a hypothesis fails a test that was designed to be failed, then it can't have been true. Whereas if a hypothesis passes a test you already knew it would pass, you have learnt nothing.

Allow me to illustrate this with a parable. Imagine two men go for a walk. One of them is a sceptical scientist, let's call him Karl (after Karl Popper). The other is a homeopath, let's call him Tom (after Mr. Harkin).

Our two men both see a white sheep in a field (for some reason, neither has seen a sheep before, but let's not worry about why that might be). Having seen this sheep, both of them formulate a sheep-based hypothesis - that all sheep are white - and they both decide to go their separate ways to test this.

Tom, our alternative medicine practitioner, wanders around looking for other white sheep, and with each one feels more confident about his position. He knows where he can find more white sheep, so he goes and looks at those, and by doing so further bolsters up his position in his mind.

Karl, on the other hand, heads off to unknown places looking around with genuine curiosity to see if there are any examples of sheep that are not white. Though he may see many white sheep, he keeps in mind that every new example he sees does not tell him anything new. He already knows white sheep exist, so seeing another white sheep means nothing. Eventually, he comes across a field wherein a black sheep quietly grazes.

Karl has disproved his hypothesis. Not all sheep are white. He now knows more than Tom. While Tom is confidently giving talks on his theory and backing himself up with hundreds of examples of white sheep, Karl has discovered that his original hypothesis was not true. He is now able to ditch it and come up with a new hypothesis that better matches reality, and start testing that.

By setting out to falsify their hypotheses, rather than confirm them, scientists are able to learn more about the world. Creationists, CAM proponents and pseudo-scientists look for confirming evidence and ignore or reject any evidence that contradicts their position. In this way, they leave themselves in the dark about the truth.

4 comments:

  1. So was this post inspired by N.N Talebs Book - The Black Swan?

    An interesting post.

    ReplyDelete
  2. I haven't read that, but the black swan is a common example used in describing Popperian falsificationism.
    I just decided the whole thing would be a bit cuddlier with sheep.

    ReplyDelete
  3. Thanks for the great blanket statement in the last paragraph, protecting all the totally unbiased and objective evolutionists who turn a blind eye to fossil and mathematical evidence

    ReplyDelete
  4. Thanks for commenting. Would you like to furnish me with any examples of overlooked evidence?

    ReplyDelete

Please make any comments constructive and interesting. Spam and abuse will be removed.