jt and Caveman,
- I haven't figured it out yet.
- As far as I can tell, you guys haven't either.
- But then, I'm known to miss a lot...
- Do you have more to say about this? Is my claimed lottery analogy not analogous?
To be honest, I'm not really following the specific debate on your claims all that closely, I'm just here for the probability stuff. But since you asked, I took a look through the past couple of pages, and here's my 2 cents:
Mojo's quest for finding fallacies is counter-productive. When playing that game, at some point you just see what you want to see. The one that I can see to be well-supported is that you are assigning a lower probability to a compound hypothesis including a simple hypothesis to which you, on its own, assign a higher probability (ie body vs body + soul). The other ones seem to be pushing it a little, increasingly so as increasingly more are "discovered".
I find that Texas Sharpshooter/special case/HARKing argument unconvincing. One can certainly make deductions based on one's own existence, even probabilistic, and even after the fact.
For example, suppose I have an electrical wire in front of me. If I touch it and the wire is live then I die, if it isn't live then I survive. I touch it, and I survive. I conclude that the wire wasn't live. Perfectly fine deduction. Making it probabilistic doesn't change this. Suppose that if it's live that I have a 99% chance of dying and 1% chance of surviving and vice versa for it not being live. I conclude that the wire likely wasn't live. Also a perfectly fine deduction. Even coming up with this after the fact doesn't change this. Suppose I remember now that, indeed, 10 years ago I touched such a wire and survived, then I can still reach the same conclusion, based on my current existence, regarding that wire from 10 years ago.
I think that the problem rather is that your conditional likelihoods are in fact equal, contrary to what you claim. Yes, the odds of you existing under the - as you call it - scientific explanation are pretty damn small. Out of the insanely large number of potential people which could have been created under the scientific explanation, it's a very small chance that you would have been among the actual people who got created. But then, the same argument applies for the immortal soul. Out of the insane number of potential immortal souls which could have been created, it's a very small chance (the same small chance) that you would have been among the actual immortal souls which got created.
Hence, I think the error is simply your claim that P(E | H) =/= P(E | ~H) whereas they are, in fact, equal. Given that the relative likelihood is 1, all you're getting out of this exercise is your own prior beliefs getting reflected back at you. As I've said much earlier here:
As far as I know there is no evidence either way for mortality or immortality, so that discussion is pointless - you just leave with whatever preference you started with.
You can believe in immortality as much as you want, it is not
a priori better or worse than to believe in mortality. But you're not doing what you seem to think you're doing, ie providing an argument as to why people should adopt your belief in immortality - and you sure ain't proving it mathematically.
Then on the other hand, people who self-identify as skeptics in this thread have a tendency to claim that science supports their case for mortality, forgetting that just because evidence is consistent with one hypothesis doesn't mean it is necessarily any less consistent with another. If I
really had 5$ every time someone made that error I'd be rich now.
All in all, this thread is a huge exercise in futility, but the tangents on probability theory every once in a while are interesting though.