Hume argues against miracles by means of his balancing principle:
- (HBP) You should believe p on the basis of testimony only if p is at least as probable as the falsity of the testimony.
There are two interpretations of HBP, depending on whether “probable” refers to the prior probabilities (the probabilities before the evidence of the testimony is accounted for) or posterior ones (the probabilities after the evidence has been weighed). On the posterior interpretation, HBP is almost completely obvious (at least if the “should” is that of epistemic normativity). On the prior interpretation, HBP is well-known to be false: the standard counterexample is that it’s reasonable to believe that you won the lottery on the basis of a newspaper report of the winner even if the chance of a newspaper error exceeds your chance of winning the lottery.
I think the prior interpretation fits Hume’s text better, even if it’s bad epistemology.
In this post I want to suggest that there could be reasonable assignments of priors for a theist on which the prior probability of the falsity of the testimony is less than the prior probability of the miracle.
Assume we are theists. Take the resurrection of Jesus. First, let’s say something about the prior probability of the resurrection of a human. Given theism, there is a good God, and it wouldn’t be surprising at all if there were resurrections. In fact, we might expect it from a loving God. But how often would they happen? What is the resurrection rate in human beings? Well, here we need to turn to empirical data. Let’s grant Hume that apart from the case under examination, there are no resurrections. There have been approximately a hundred billion human deaths, so we have an upper bound on the resurrection rate of less than one in 1011. It’s not unreasonable, I think, given the moderate prior probability that someone would be resurrected, and the lack of resurrections in 1011 cases, to suppose the probability of a particular person getting resurrected would be something like (1/2) ⋅ 10−11.
But what is the probability of false testimony? Well, as an initial back of the envelope calculation, suppose we have 11 witnesses, and each has an independent 1/20 chance of lying or being mistaken that Jesus was resurrected. So, the chance that they all lied or were mistaken would be (1/20)11 or (1/2048) ⋅ 10−11.
With these numbers, the prior probability of Jesus getting resurrected is about 100 times bigger than the prior probability of the 11 witnesses lying that he was resurrected. And so even in its prior probability formulation, HBP doesn’t destroy the testimony to the miracle.
Of course the numbers are made up. Probably the main problem has to do with the assumption of the independence of the witnesses. But that problem is to some degree balanced by the fact that 1/20 is way too high for a probability of lying or being mistaken that they witnessed a resurrection. (What percentage of the people you know testified to witnessing a resurrection?)
In any case, I think the above shows that it is far from clear that, assuming theism, a reasonable estimate of the resurrection rate of humans would be lower than a reasonable estimate of the resurrection mendacity rate for groups of 11 people.
Now what if we don’t assume theism, but assume, say, a 1/10 chance of theism? Well, that approximately cuts our estimate of the resurrection rate of humans by a factor of 10. But that’s still not enough to make it clear that the resurrection rate of humans is less than the resurrection mendacity rate for groups of 11.
13 comments:
I know these numbers are just ballpark figures to make a point, but it seems obvious to me that the probability of each false resurrection claim is not independent, right? If one person in a relatively tight-knit religious community makes a particular claim about an experience or encounter, that would dramatically raise the chances that someone else would say, "Hey, me too!" If that isn't the case - if you have to evaluate each probability independently - then how could you handle the dozens of people who have reported miracles about someone like Sathya Sai Baba?
Yeah, but we also have to take into account the fact that the Apostles are persisting in their false claims despite threat of death, a threat that they presumably often face alone, and hence with some degree of psychological independence.
I mean, think about this. You're in this community, and someone in the group "We had breakfast with our Master a couple of days after he died", and you said "Hey, me too!", because it's embarrassing not to go along, even though you know it's not true, because you know that you had a gloomy breakfast of burned fish that day, as John was crying quietly rather than keeping an eye on the pot. The fake feels kind of good. But then years go by, things get tough, people in your group get tortured to death. It's pretty natural to just wonder away, since you know it's all made up, and that if you do get tortured to death, either there will be no life after death as a lot of the intellectuals say or you'll face an Old Testament God who will be really angry at you for making things up and leading people after a false prophet. There is a fair amount of independence here. You might stick to what your friends made up, but will you? I mean, it really hurts *a lot* to get crucified, and you've seen it happen, so it's pretty vivid to you.
I wonder how low the number of witnesses could get while still providing a decently high prior for the resurrection. I think most skeptics will contest the claim that we have good evidence for all of the apostles persevering; scripture and tradition say they did, but that's clearly not going to convince them. But I think most will grant that there's good evidence for Peter, Paul, and James the Just all being martyred. Maybe from there we can come up with different priors based on different scenarios (e.g. what's the evidential effect of having eight witnesses, or nine, and so on)?
Paul was not a witness for the resurrection.
As to Peter and James the Just, there are stories about them being witnesses, but we do not have their personal testimony. Moreover, we do not know why exactly they were martyred.
There is no fact that the Apostles were persisting in their false claims about the resurrection despite threat of death."
Walter:
Technically, none of them were witnesses to the resurrection; they were witnesses to the *resurrected Jesus*. In that sense, Paul was a witness just as much as Peter and James.
Most scholars do think that Peter, Paul, and James were martyred (see Allison's 2021 *Resurrection* book for a useful survey of the evidence). And if this is true, then they were martyred for their religious claims, of which the resurrection was a central aspect.
James
They we're martyred for their religious claims. What exactly those claims were is unknown. Paul claims to have had a vision of Christ. Maybe Peter and James also had visions.or not, who knows? They probably believed in some kind of afterlife and they believed that Christ was some sort of Messiah, and, like many other believers in various cults, wereprepared todie for those beliefs, because they may have thought, just like modern suïcide bombers, that this would guarantee them a place in the afterlife.
But the bottom line is that we do not have firsthand witness accounts if a bidily resurrectef Christ.
Walter:
I don't think anybody doubts that the resurrection was a core part of the apostles' religious claims. Even if they were martyred specifically for claiming that Jesus was the rightful King of Israel, this claim depends upon the resurrection (since it was only their belief in the resurrection which allowed the apostles to continue regarding Jesus as the Messiah). And if they were willing to die for these beliefs, then it is very, very unlikely that they were lying about them (though of course this doesn't prove that their beliefs were *true*).
Also, your last sentence begs the question. If Christianity is true, then what Paul provides us *is* first-hand testimony concerning an encounter with the risen Christ. The same would hold true for the various traditions preserved in the Gospels. Of course, *skeptics* don't regard these stories are relating a real encounter with the risen Jesus; they would argue that they simply relate powerful delusionary religious experiences of one type or another. But that's the whole question at issue; simply asserting that the naturalistic interpretation is correct will not settle things :)
James
Even if Christianity is true, then what paul provides us is not, first hand testimony concerning an encounter with the bodily resurrected Christ.
There are a few reasons for this
The first one is in Paul's own description of what happened. He never mentions that he has met the risen Christ on earth, but that Christ revealed certain things to him. He even says that he (Paul) was taken to heaven, whether bodily or not he can't say.
The second one is that the stories in Acts do not mention a bodily resurrected Jesus, but an experience on the road to Damascus that involved a voice (that nobody but Paul could hear) and a light.
And the most iportant one (to me) is the fcat that when Paul had his experiences, according to Christian tradition, Jesus had already ascended to heaven.
As to the other apostles, all we know about them are second hand accounts. The Gospel of Mark, which is gererally agreed to be the oldest one and which was used by both Luke and Matthew for their own work, does jnot mention any post-mortem appearances by Jesus.
Walter:
Paul pretty straightforwardly claims to have "seen Jesus our Lord" (1 Cor. 9:1). His description of being taken up to heaven (2 Cor. 12:2-4) is *usually* thought to be describing a different experience. But in any case, I'm not sure what the relevance of your argument is. The point of the post is simply that, for theists, the prior probability of a resurrection is probably higher than the prior of the apostles *lying* about the resurrection appearances. Whether those appearances involved Jesus in his body, or whether they were simply veridical visions, seems irrelevant.
James
Alex was talking about 'lying or being mistaken'. The probabilty that Paul or thé other disciples were mistaken', is, considering how many times people think they saw something, actually very high.
Now today, how many of the millions if Christians have actually witnessed thé risen Christ? Most Christians believe in it die to what others have told them.
Now, if what others told them is not true, we have millions of mistaken people. Not just 20
We are not infrequently mistaken about a glimpse of someone. But we are very rarely mistaken about someone's identity when we knew them well and have a meal with them, except in cases like identical twins, or someone we haven't seen in a long time, or prosopagnosia, or they are wearing a mask.
There is no first hand testimony of someone having a meal with Jesus after his death.
From the modeling perspective, I am missing any account of testimony from non-Christians reporting events supporting their religions. Are all their accounts non-trustworthy? Since the blog post suggests specific probabilities in the Christian context, what would be reasonable probabilities for the veracity of non-Christian reports?
Post a Comment