Thursday, May 24, 2012

Be Positive, Atheist

Tris Stock over at My Godless Life has started a multi-part refutation of an article by Christian apologist Wayne Jackson. In it he claims that negative atheism is the default position (I agree). Negative atheism stands on the very firm ground that supernaturalists all utterly fail to provide any support for their positions and not accepting their hypotheses is the correct position to take. He also claims to have met very few positive atheists, those who go on to make the assertion that there are no gods. I don't know how many he's met, but I'm one, and I want to argue (again) that I'm right.

I'm a positive atheist, and I know I'm not alone in that. Leaving aside philosophical arguments about the incoherence of the god concept and other such, I rely on a Bayesian analysis.

Simply put, in a Bayesian analysis you way the evidence in the probability that it supports a hypothesis and the probability that it does not support your hypothesis. For example: is there any mayonnaise in my fridge? Without looking in my fridge you can guess that there probably is because your extensive experience with American culture has provided you with evidence for the ubiquity of eggy emulsions of oil and vinegar. So you weigh a lifetime of observational evidence against the two hypotheses: there is mayonnaise or there is not. Sight unseen, the evidence weighs in favor of the presence of mayonnaise.

But then you open the door of the fridge and you see no mayonnaise. This new evidence is added to both hypotheses. It doesn't add strength to the "there is mayo" hypothesis, but it does add strength to the "there is no mayo" hypothesis. You can contend that it weakly adds to "the mayo is in the cupboard" hypothesis, but it's simply not as powerful an argument as the "no mayo" hypothesis.

You can take it further by looking in the cupboards and under the stairs and in the dog, and each place you fail to find mayo takes strength away from the positive hypothesis and piles it onto the negative hypothesis.

Bayesian analyses can be very rigorous, using actual numbers and probabilities, provided you're a good statistician and can actually do that math. Even if you can't, however, you'll see that we're typically intuitive Bayesians, even if the numerous logical fallacies to which we're prone interfere.

Let's apply it to unicorns. A few centuries ago it was possible to be agnostic about unicorns to the extent that "Well, we haven't been all around the world yet". But then biologists started scouring the world and recording every little thing they saw. Everywhere they went that they failed to unicorns piled up evidence in favor of the "no unicorns hypothesis".

The alternative hypothesis is, and has always been, "There are unicorns, but we haven't seen them yet", which became fatuous two centuries ago. Once upon a time the lack of evidence for unicorns applied to both hypotheses equally because the pool of places we'd been and things we'd seen (and communicated to the world about) was small. Now the only way to sustain the "haven't seen them" hypothesis is to claim that unicorns are like squids, living somewhere remote and inaccessible (which no one does) or to claim that they're like rare jungle flowers, living high up trees motionless and even if you do see them it's easy to mistake them for another species of unicorn that's already been recorded.

What about god? They god hypothesis only gets away from this kind of analysis by special pleading. Somehow it's perfectly reasonable to dismiss all other supernatural phenomena because absence of evidence actually is evidence of absence when you would expect evidence, but not the god phenomenon, no sir. That gets my dander up because the only kind of god you could make that claim for would be the deist god and, really, what's the difference between a god that doesn't exist and a god that has all the characteristics of non-existence except the final one1? All other deities I've heard posited are active and interventionist, and the believers all claim to have evidence. Why, once their evidence is proven spurious, would we suddenly accept their claim that their god is beyond evidence? I don't see a need to bend over backwards for them in that way.

The three hypotheses under consideration are T) There is a god and we have evidence, D) There is a god, we just haven't seen him, and A) There is no god. Obviously all atheists have considered and rejected T, because the purported evidence really isn't. Miracles, scriptures, revelations, they all fail as evidence because there are other hypotheses for each of them that better explain the evidence. Also, eve when you consider them all together, they don't count as evidence because each class of purported evidence as a whole has better explanations.

Thus the argument comes down to whether we live in a time where we simply haven't explored enough of the world to know whether our unicorn exists, or if that unicorn doesn't exist.

SETI is a good analogy here. There may be intelligent life elsewhere in the universe that, despite our ongoing exploration, we haven't yet found evidence for. Does that count as evidence against the proposition? Yes, actually. However, the amount of the galaxy we've examined is so small compared to the amount we haven't gotten to yet that agnosticism is still the default position. For the three hypotheses of SETI, we've got nothing going for "other life exists", a very tiny amount for "we're alone" and a whole heaping massive galaxy in the corner of "we haven't explored here yet". The longer we explore, the longer we continue to fail to find evidence, the more we scrape away from the agnostic position and add to the negative position. Of course, all it will take is one piece of evidence to add to the "yep" pile to render the other two moot.

The difference between SETI and the god search is that absolutely no one claims that god is far away. The god hypothesis always, always posits that we don't need to cross the unfathomable depths of space to find their god. It's supposed to be right here. All of a sudden the "we haven't explored here" option loses all its weight, because we have explored there. The whole of human existence has been the exploration of that position, and the signal has always come back null. We're constantly told that we should expect evidence, and we have absolutely none in favor of the position. Therefore, the only conclusion is not merely to reject the position as unproven, but to affirm its opposite: the position is wrong.

1 - Hint: There's no difference. The deist position should be treated as identical to the atheist position. So, I think, should the Spinoza position.
Post a Comment