Originally posted by Cris
Where there is an absence of evidence a null result is perfectly valid and rational. Especially since absence of evidence is not evidence of absence.
Careful with language, Cris. Mathematically, and as commonly used by statisticians, "null result" means a
probability of a hypothesis being true that is less than 0.95
I think you're using "null" in a computer programmer's sense. Correct me if I'm wrong.
Note that the terms ‘tend’ and ‘naturally’ imply a partial set, and hence cannot apply to all weak atheists.
I could've been more assertive. Humans always have a probability judgement, no matter what question you pose to them. Whether the judgement is accurate (and whether they feel their own assessment is well-grounded), is another matter.
However, while your assessment has some truth it is beyond the scope of weak atheism, which is strictly disbelief so an assessment of belief has nothing to do with weak atheism.
Let's say that the strong atheist position is pr(P) = 0 (P being "a god exists"). The weak atheist clearly disbelieves this equality, so he would state pr(P) != 0. If not, he'd be agreeing with the strong atheist and in fact he would be himself a strong atheist. pr(P) != 0 is the same as pr(P) > 0. On the other hand, if you refuse to choose at all, you're agnostic (i.e. you're walking away, whining "I don't know".) But you must have an estimate, even if an extremely ill-founded one.
And without evidence such a conclusion would be irrational and not everyone thinks the way you suggest.
I'm suggesting that the way people think can be expressed without loss of information in terms of their estimating probabilities. In fact, I think beliefs are nothing but probability estimates. They may change in interesting ways and have all sorts of biases, but that's just the mechanics of probability estimation particular to the human brain.
Where there is a total absence of evidence the question of the probability for the existence of a god must be assessed identically to any assessment of any imaginative fantasy.
You sound like a strong atheist there (I agree.)
Without evidence there is no method that can be used to assess the potential truth of a fantasy, and hence the result must be null, as opposed to a zero or positive value.
Of course there is a method. You said it yourself, you must judge that fantasy on the same grounds you judge all fantasies. In probability theory it's very simple. Take 1, divide by size of the set of possible outcomes, and presto! you've got a probability estimate. IMHO, the set of all fantasies is infinite, so the resulting outcome is 0.
This is a so-called apriori probability estimate [pr(P)]. Now, if you're talking about basing your beliefs on some evidence, then you're in the realm of aposteriori probability estimates [pr(P|Q)]. Both are a valid form of estimate in statistics, and both types are used routinely in real life.
