Friday, January 05, 2007

Off-topic: A scientific investigation into Richard Dawkins

I haven't posted much in the last fortnight. I've been on holiday!

But I've also been working on what follows. It's not directly related to the BCSE, except in so far as we've seen that the core of the BCSE leadership is generally following the script which Richard Dawkins writes (see here and here).

Here is is:

"Does Richard Dawkins Exist? A Parable"

If you like it, you can show your appreciation by linking it from any websites or blogs which you have.

David Anderson


Iain said...

I'll attempt a reply, David.

It's an amusing and well-written article. I was a amused by the reference to the nerdy type in glasses. I hope you weren't referring to anyone in particular in the ID movement! :-) As it happens I wear glasses and I also work on algorithms that detect patterns in data ... but I no longer consider myself an ID supporter.

I think you are trying by this humorous parable to show the absurdity of any argument that states Dawkins can't exist, and evidence of Design (i.e. that the book shows evidence of having been written by an intelligent author), shows that Dawkins clearly exists and it's absurd to suggest that the book arose by spontaneous mutations, or that it's only at a reductionist level just a collection of atoms.

This is (I think) little different from Paley's Argument from Design for the existence of a Designer. I was for a long time persuaded that this argument (and yours by the same token) is strong and persuasive. However, I am now not convinced at all.

In order to make an inference that some object (be it a book or a biological mechanism) was created by an intelligent being, there must be two different conditions that must be satisfied.

(1) The designer must exist.
(2) The designer must have designed the object in question.

My problem with the Paley argument (and yours over Dawkins's book) is that one's a priori probability of the existence of the watchmaker/Richard Dawkins is one. You know about watchmakers and you can get hard empirical evidence that they make watches like the one described in Paley's article. You've only to go to a watch factory and see it happen. Equally you've only got to watch a TV program on Richard Dawkins to show that a person called Richard Dawkins exists (or you could even track him down to his home in Oxford, as some Australian creationists once did) to prove he exists. You also know that he writes books and talks about them. Therefore it's not at all difficult to reach the conclusion that "The God Delusion" was written by Richard Dawkins, or Paley's watch was made by a watchmaker.

The problem with applying this to biological structures is that you don't know about the a priori probability of the existence of a designer, and you don't have any independent evidence of it (like you could just go and visit Richard Dawkins and verify that he exists). The only evidence you have is the evidence of design. How do you assign an a priori probability to God? This is simply not possible without independent evidence. A Christian such as myself would assign a probability (according to my beliefs) of close to one. But an atheist like Dawkins will give an a priori probability of close to zero (actually I had a colleague once who was even more extreme than Dawkins and gave a probability of zero, to which another colleague who gave 10^(-100) as the probability said "Now that's what I call faith!").

If your prior probability of God's existence is close to zero, then alternative explanations of design (such as evolution), even if they don't seem likely at first, become the only game in town.

Furthermore, we don't know the probability that evolution could succeed, and can't even estimate it, because we don't have sufficiently good models of DNA and what the space of organisms spanned by DNA sequences actually looks like. To say that "This couldn't have happened because it's too unlikely" is often described as an argument from ignorance. It's easy to take that as intended as an insult, but I don't mean it insultingly. That ignorance applies to me just as much as to anyone, and a more humble approach is to suggest that we don't know, so we have to accept the possibility that evolution could indeed have occurred naturally.

To summarise. Despite the clever way you've argued for the existence of Richard Dawkins by "Design" (or pretended to argue for his non-existence), the fact remains that we have independent, and scientifically testable evidence for the existence of Richard Dawkins. That evidence is needed before you can make a "design" conclusion.

By contrast, one can't present Design by itself as evidence of the Designer - to assume that you KNOW it couldn't have arisen naturally is lacking in scientific humility.

David Anderson said...

Hello Iain,

Wow - quite a response!

Parables are never allowed to be explained, as they are told only "for those who have ears to hear"... (at least in public, anyway!).

Nevertheless, there are two things I'd like to say in response to some of your points:

1) In reply to what you say about something being too unlikely, and arguing from ignorance... you surely can't mean this absolutely? All scientific reconstructions of the past are tentative in some sense and rely on a whole set of inter-related judgments about likelihood.... I don't think any IDer would admit to making the mistake you talk about in your last paragraph.

2) You have assumed that the designer is God. I don't believe that science can say anything about the question of the designer's identity. As such, I'd say that your whole discussion about needing to have an a priori opinion on the existence of God is moot. To use the watch-maker analogy, discovering the watch only tells me that there is a watch-maker: it doesn't tell me that it's definitely H Samuel from the High Street, and hence my a priori opinion of the existence of H Samuel does not come into it.