The argument from intelligent design

Dr Charles Townes was on National Public Radio yesterday, because he has been awarded the Templeton Prize for progress or research in spiritual matters. Dr Townes won the Nobel Prize for physics for inventing the laser, and is Professor here at Berkeley.

What shocked me was that Dr Townes said that he thought that complexity of the universe, arranged in a way that permits life, while not conclusive, was at least suggestive of the existence of God. This is known as the argument from intelligent design. The argument runs as follows. A vast array of different things need to be in place for life to exist on earth, ranging from the Earth being just the right distance from the Sun, the existence of water, to the immense complexity of our biochemical structure. The chances of each of these things is very small; the chance of them all coming together are very small indeed. So, the argument goes, it seems more likely that some "intelligent design" is behind the this set of circumstances.

This argument may sound convincing at first, but it is stastical nonsense. Suprisingly, a quick trawl of Google did not turn up any simple or clear critiques of the argument.

To see why this is a fallacy, keep in mind that for evidence to be useful in distinguishing two theories, an observation must be more likely under one theory than the other. For example, if you drop a heavy object and a light object from the Leaning Tower of Pisa, then if you believe Galileo they should hit the ground at roughly the same time. If you believe an alternative theory – say, that heavier objects fall faster – then they should hit the ground at different times. If we observe two objects landing at the same time, then we can say that this is evidence against the alternative theory, because that theory suggests that this observation is very unlikely. This would strengthen your belief in Galileo’s theory, which predicts that this is likely. But if two theories both predict that the objects would hit the ground at the same time, but differ in some other way (eg in their prediction of how fast the objects would be going at that time), then your observation would not help you to distinguish between those two theories.

Let us apply this to the observation that the universe has the characteristics necessary to support life. If the theory of intelligent design were true, what are the chances that we would look around us and observe a universe with the complex characteristics needed to support life? The answer is 100%. If the theory of intelligent design were not true, what are the chances that we would look around us and observe a universe with the complex characteristics needed to support life? The answer is 100% again. The point is, if we are able to look around us, we MUST be in a universe that is capable of supporting life. So both theories predict this observation with exactly the same probability. The observation provides absolutely no evidence for or against the theory of intelligent design.

Another way to think of this is that it is a bit like being dealt a good hand in a game of cards. In Bridge, the odds against being dealt any particular hand are about 13 billion to one. But when you are dealt your hand and you look at your cards, do you conclude that the dealer is cheating because the chances of having been dealt just that hand are so infinitesimally small?

Just to be clear: I am not saying that the argument from intelligent design is merely weak because the chances of life developing are less unlikely than they might first appear (though this may also be true).

The argument is logically invalid: because the observation is equally likely under either hypothesis, it tells us absolutely nothing about which we should believe. It seems strange to me that an eminent academic physicist should fall into this trap.

3 thoughts on “The argument from intelligent design”

  1. You capture the statistical shenanigans around ID perfectly. I also like John Derbyshire’s example: from the point of his conception to a working-class mother in England, what were the chances he’s meet his future wife in a village in north-west China…but it happened…and so on. Post hoc arguments around stats are of course common errors – but not, like you say, usually made by Berkeley profs.

    I would just point out another couple of problems I have with the ID ‘theory’:
    1. It has no predictive power at all. Unlike evoluton, ID has nothing to say about what might happen tomorrow, or in the next ten thousand millenia. That makes it an assertion not a scientific theory.
    2. It has no explanatory power. It invokes a ‘creator’ or ‘intelligence’ as the answer to the ‘how’ question about life on earth, but that just begs another: ‘how’ the creator? This, too, makes ID an assertion not a theory.

Leave a Reply

Your email address will not be published. Required fields are marked *