August 9, 2010

Battleground God is doing the rounds on Facebook. It’s a quiz on God-belief. It doesn’t try to argue for theism or atheism, but rather, checks whether your beliefs about God are consistent. I think I did the quiz a few years ago, but I did it again, and found I’m still a logically consistent atheist (at least as far as a quiz which asks a few True/False questions can tell), which was nice.

You’ll never find a Nessie in a zoo

There was some debate over these two questions:

Q10: If, despite years of trying, no strong evidence or argument has been presented to show that there is a Loch Ness monster, it is rational to believe that such a monster does not exist.

Q14: As long as there are no compelling arguments or evidence that show that God does not exist, atheism is a matter of faith, not rationality.

If you answer True to both, you get “hit” (which is bad, because you’re on a philosophical battlefield, remember). The site says:

Earlier you agreed that it is rational to believe that the Loch Ness monster does not exist if there is an absence of strong evidence or argument that it does. No strong evidence or argument was required to show that the monster does not exist – absence of evidence or argument was enough. But now you claim that the atheist needs to be able to provide strong arguments or evidence if their belief in the non-existence of God is to be rational rather than a matter of faith.

The contradiction is that on the first ocassion (sic) (Loch Ness monster) you agreed that the absence of evidence or argument is enough to rationally justify belief in the non-existence of the Loch Ness monster, but on this occasion (God), you do not.

On Facebook, AH (and perhaps RH) said that the cases were different because of the “years of trying” part: Nessie is supposed to live in Loch Ness, so a thorough search might be enough to convince us that she’s not there. If we cannot find God, however, maybe we just haven’t looked everywhere yet.

I think the “years of trying” thing is a red herring here, so the question is badly phrased. The FAQ for the Nessie question explains that in both cases, you’re presented with a lack of positive evidence that something does exist, rather than a positive evidence that it does not (edit: as I said below, on some definitions of evidence, this is a distinction that makes no difference, but it’s one the site seems to make, so we’ll run with it). If you propose using sonar to search Loch Ness, a true Nessian will not be moved: didn’t they mention that Nessie is undectable to sonar? You might find it suspicious that the Nessian knows in advance what experimental results they’ll need to excuse, but that’s merely a demonstration that the Nessian doesn’t believe in Nessie but rather believes in belief in Nessie. It’s not an argument that there’s no Nessie.

The FAQ adds that the Q14, the God question, doesn’t actually specify our current knowledge level about the universe: if we found explanations for the universe’s existence which did not involve God, this would not show that God does not exist, in the same way that the sonar survey wouldn’t show that Nessie didn’t exist, but the question is whether it’s rational to conclude that Nessie and God do not exist merely because of a lack of evidence that they do. Again, I think the “years of trying” in the Nessie question confuses things here, because it puts us in mind of the current state of affairs.

PH objects that the empiricists who wrote the quiz have assumed that God is the same sort of thing as Nessie. But the FAQ points out that the Nessian could say that, while of course Nessie and God are very different, they still assert that Nessie has the ability to evade detection. The atheist could then legitimately wonder why the theist gets to make the “you can’t disprove God” move while the Nessiant can’t make a similar move.

Edited: simont makes the point that we pretty definitely know what we mean by “Nessie”, and there aren’t in fact Nessians making clever claims that Nessie just isn’t the sort of thing that can be detected, whereas this isn’t the case for “God”. The fact that the FAQ has to go into it that particular “hit” in such detail means they could have done it better. A more obviously supernatural belief which doesn’t entail theism or atheism would prevent some of the arguments about those two questions. So, I’ve now concluded that it is fair to object based on the usual meaning of “Nessie”, and that Q10 should probably be about the existence of ghosts.

Bayesian homily

This talk of absence of evidence and evidence of absence reminded me of the appropriate Less Wrong article. I think the FAQ has confused evidence and proof. Absence of evidence for Nessie is evidence of absence of Nessie if that evidence would even slightly make us believe more strongly in Nessie (assuming the evidence is the sort where you either see it or you don’t, rather than having several possible outcomes: I’ve not worked out what happens in that case). This is just the same as the healing prayer case we mentioned before. Many Christians claim that their God’s existence is so obvious that anyone who doesn’t accept it is culpably deceiving themselves. I think this must mean that they would agree that we should expect to find evidence of the Christian God, in which case, if we don’t, that weighs against his existence.