Re: [asa] Dawkins, religion, and children

From: Iain Strachan <>
Date: Wed May 02 2007 - 17:36:25 EDT

On 5/2/07, David Campbell <> wrote:
> > I think that Dawkins position is a bit more subtle than that. Dawkins'
> > position is not one of a-priori non-existence of God but rather a more
> > subtle, and thus much harder to reject, argument based on
> > improbabilities.
> In other words, the same arguments as ID, but making up different
> numbers for the unknown probabilities in order to generate opposite
> conclusions.

I should clarify to Pim, that a priori probabilities can be extremely small,
but still a measure of BELIEF. The foundations of statistics are based on
Bayes' theorem, whereby Prior probabilities (often termed "beliefs") are
modified to posterior probabilities in the light of observatons (evidence).
A common class of probabilistic model is called a "Bayesian BELIEF network".
So I think the distinction you want to make between a very small probability
and zero is splitting hairs. It's still about belief.

However I think it's rather a silly argument to assign a probability to
God's existence anyway. A probability is defined as the number of times an
event is expected to happen in a large number of trials, divided by the
number of trials, e.g. what is the probability you'll get a double six when
you throw two dice. If you repeat it for 3600 trials, you can expect to get
a double six about 100 times.

Similarly with Russel's teapot orbiting the sun, which Dawkins is fond of
quoting. You know that teapots exist, you can calculate the probability
density function of teapots as a they are distributed over space by
observing when you find a teapot, and there are many teapots. You'll find
that the probability density function is higher in cupboards, and on tables,
lower on floors, very low on the street, or in the middle of a river
(someone might have chucked a teapot in the river on a picnic), and orbiting
the sun in outer space, vanishingly unlikely. But you can do this because
you know about teapots - you can physically measure when they are there.

But there is only one God, and only one "event" being assessed of two
possibilities. Either God exists or God does not exist. There aren't
millions of trials you can conduct. You can't visit millions of universes
and see which ones God exists in and which ones God does not. So I think
it's an absolutely ludicrous concept to assign a probability to God's

As an amusing aside, in the theoretical physics department of where I
worked, people were asked in a questionnaire to assign probabilities to
various things (like England winning the world cup football). One of them
was "God". Two guys, both staunch atheists, wrote down respectively
10^(-100) and 0. When the 10^(-100) guy saw what the 0 guy had written, he
said "now that's what I call faith!"

I was reminded of that when I read the passage in The God Delusion lovingly
typed out by Pim for our education, about the seven levels of belief or
disbelief in god (given probabilistically). The 10^(-100) guy was in the
same camp as Dawkins puts himself - at level 6, leaning towards level 7.
The 0 guy is interesting, because even Dawkins says in TGD that he does not
expect to find many of those people around. Mathematically put P( Person
thinks P(God)=0) < epsilon.


To unsubscribe, send a message to with
"unsubscribe asa" (no quotes) as the body of the message.
Received on Wed May 2 17:37:05 2007

This archive was generated by hypermail 2.1.8 : Wed May 02 2007 - 17:37:05 EDT