Then there is NO way of showing a percentage for possibility of God raising Christ from the dead...
Correct.
There are several ways to estimate the probability of a dead person coming alive again. The simplest way is to pick a random selection of the people who have died and count how many of them came alive again. If we take everybody who died in the 20th century (OK not a random sample, but it will do for illustrative purposes), and count how many were resurrected, we get a probability of 0/5.5 billonish. Even accepting all of the reports of resurrection at face value, it's still probably millions to one against.
Of course, if God is involved, he could choose to resurrect just a handful of people in the whole of history in which case the sample size needs to be enormous to get anything other than zero. My methodology above would be equivalent to selecting 100 people who bought lottery tickets yesterday, checking if any of them win the jackpot on Saturday and, when they don't, concluding that it is impossible to win the lottery jackpot.
I could try calculating the probability another way, just as with the lottery we can figure out the probability of winning the jackpot through permutations and combinations. I could look at the probability in quantum mechanical terms of the particles of the dead person spontaneously jumping into a state where they constitute a living person. This is possible but astronomically unlikely. There again, if God exists, he could nudge the particles into the correct configuration.
OK, rather later than planned, here is an attempt at working out a probability.
Quoting from p 271 of Reasonable Faith by William Lane Craig where he is pointing out some problems with stuff David Hume wrote:
Letting M = some miraculous event, E=the specific evidence for that event and B=our background knowledge apart from the specific evidence, the so-called "odds form" of Bayes' Theorem states:
Pr(M|E&B) = Pr(M|B) x Pr(E|M&B)
----------- ------------ -------------
Pr(non-M|E&B) Pr(not-M|B) Pr(E|not-M&B)
On the left-hand side of the equation P(M|E&B) represents the probability of the miracle given the total evidence and Pr(not-M|E&B) represents the probability of the miracle's not occurring given the total evidence. The odds form of Bayes Theorem gives us the ratio of these two probabilities. If the ratio is 1/1 then M and not-M have the same probability; the odds of M's occurring are, as they say, fifty/fifty or 50 percent. If we represent this ratio as A/B, what Hume wants to show is that, in principle, A<B - for example, 2/3 or 4/9 or what have you. So given the odds, one could never rationally believe, no matter what the evidence that a miracle has taken place.
Now whether the miracle is more probable than not will be determined by the ratios on the right hand side of the equation. In the first ratio, the numerator Pr(M|B) represents the intrinsic probability of the miracle and the denominator Pr(not-M|B) represents the probability of the miracle's not occurring. We're asking here which is more probable, M or not-M, relative to our background knowledge along, abstracting from the specific evidence for M. In the second ratio the numerator Pr(E|M&B) represents the explanatory power of the miracle and the denominator Pr(E|not-M&B) represents the explanatory power of the miracle's not occurring. We're asking here which best explains the specific evidence we have, M or not-M.
Now notice that even if the ratio of the intrinsic probabilities weighs heavily against M, that improbability can be offset if the ratio representing the explanatory power of M or not-M weighs equally or greater in favour of M. For example, (1/100) x (100/1) = 100/100 = 1/1 or a 50% probability of M.It would be hard to assign actual numbers to the above equation. I would suggest that the first ratio on the right hand side (let's call it RHS1)
Pr(M|B)
------------
Pr(not-M|B)
is either low or difficult to determine.
The second ratio on the right hand side (RHS2) is high, i.e. the dozen or so alleged appearances written recorded by apparently honest people and the empty tomb after Jesus had been killed on the cross is much more likely if he had been raised than if he had not.
The most important question here then boils down to whether RHS1 x RHS2 > 0.5. Is RHS1 sufficiently small to bring the overall calculation down to less than 0.5, RHS2 being large?
Unfortunately, we can't assume God exists for two reasons. Firstly, the death and resurrection of Jesus is one of Alan's Flakey Five arguments for God. If God id necessary to make the resurrection credible as well, we have a circular argument. Secondly, it's impossible to calculate probability if somebody is loading the dice. If you want to use probability at all, you have to assume God does not exist - or at least is not influencing the experiments.
If we assume God exists, then yes it would be circular. Let's not assume he exists and leave it as a "don't know". That makes RHS1 difficult to calculate though. I think we might find ourselves discussing this at length.
