"There is probably no God. Now stop worrying and enjoy your life."
"Probably no God," but how unlikely is it?
Richard Dawkins, pictured above, is famous for his Spectrum of Theistic Probabilities, a scale from one, absolute belief, to seven, absolute unbelief. He remarks that a six on this scale is identifiable as "De facto atheist. Very low probability, but short of zero. 'I
don't know for certain but I think God is very improbable, and I live my
life on the assumption that he is not there.'" Dawkins referred to himself as a 6.9 on this scale. I wanted to explore what "short of zero" means.
In God Doesn't; We Do, I suggest that the existence of God is infinitely unlikely. This is handled by modern mathematical formalism by a term known as "almost surely," which means true off a set of "measure" zero. Measure is a modern mathematical term that generalizes the length of intervals and is the standard accepted basis for analysis (the field in which calculus lives) at present. To avoid a long, detailed, and abstract mathematical discussion, the idea is that places where the claim is false contribute literally no weight, even if they can be said to exist. In other words, the "very low probability" need not actually be "short of zero."
My essential claim in God Doesn't; We Do is that the probability that God exists is zero, almost surely, with the suggestion that Dawkins's position is more accurately 6.999... (which happens to be 7), without any loss of philosophical defensibility on his part to take that stand. Normally, I would say I have to prove such a claim, but the point I want to clarify here is that I don't think I actually do. That, indeed, is the trust of the argument I make in God Doesn't. Technically, I make this argument on a conception of God that does things and don't particularly need to make a case against abstract ideas called "God."
Burden of proof?!
Yes, I know. I accuse theologians of shifting the burden of proof (as an art form that defines their field) all the time, and it looks like that's what I'm doing here. It's not. I'll present an argument here attempting to establish that claim, noting the question-begging fallacy all along. Question begging means assuming the conclusion, for those unfamiliar, and it can be remarkably subtle in occurrence.
So, let's start big and work to small. We should all agree that it is question begging to state a priori that the probability that there is a God (or some specific God) is 100%. If you assume God exists from the outset, you are, by definition, begging the question. So the probability that God exists must be less than 100% to avoid philosophical indefensibility. N.B.: This assumes a position that understands probability as measurement of our state of knowledge. There are others that I am not employing here.
What about 50%? This is what Dawkins calls a four on his spectrum, and it is a state of pure agnosticism, given the nota bene at the end of the previous paragraph. I don't think we can honestly hold this position without begging the question, though, in the same way that I do not think that I can conclude that the fawning eyes of the women in lingerie catalogues have a 50% chance of being indicative of those women being in love with me. It is the case or it isn't, but that need not imply equal likelihood. Unless we start with God, it is very difficult to conclude that anything constitutes evidence for God, and if we look at the hole carved by science in attributional necessity for God, 50% seems a bit steep.
My job here, though, tempting a trap as it might be, is not to make the case that there isn't a 50% chance that God exists. It's to point out that it is the job of the person claiming that there's a 50% chance that God exists to be able to establish that. I don't feel such a number is warranted in any way whatsoever by the evidence of the world, and "it is or it isn't" is a fallacious way to think about the matter. Since "God" is the hypothesis of the theist, though, it is the theist's job to establish that 50% is a reasonable number. What case can be made for this without begging the question?
Fifty percent was the hurdle. The same argument applies going downward, so we can skip quickly to 5%. Is there a compelling argument that the number that describes the likelihood that God exists is at least 5%? My claim is that it is up to the theist to provide such an argument, or that we are not required to accept that claim.
As Dawkins points out with his "very low probability" in his spectrum, we can slide to 1% or 0.1% or 0.00001% on this same construction. Where is the argument saying that God's existence is at least that likely? How low is the "very low probability"? I contend that any positive number that a theist puts out requires defense or begs the question. If I wanted to do this in math-speak, for any small number epsilon greater than zero, assuming that the probability that God exists is epislon begs the question without a proper and solid defense.
But, you can't say that?!
The only defense the theist has at this point is "but you can't say that the probability is zero that God exists without proving it!" Well, two responses are warranted. First, I haven't. I said every positive probability begs the question without an argument to support it. Second, actually, I can say that, so long as I qualify it with "the probability is zero, almost surely, that God exists." Since "almost surely" admits wildly unlikely possibility, it does not run afoul of philosophical defensibility.
So, my claim is that unless a substantial argument can be provided that establishes a nonzero, almost surely, probability for the existence of God, theists beg the question to suggest any positive probability. Some "very low probability" can actually be zero, almost surely, then, and Richard Dawkins can describe himself as a 6.999... on his spectrum. Technically, 6.999... equals 7, but since he defined a seven as "Strong atheist. 'I know there is no God, with the same conviction as Jung knows there is one,'" perhaps this justifies the use of the nonstandard form of that number as a rhetorical device.