I’m currently taking a week off work, having submitted a draft of my book manuscript (‘The Purpose of Existence: Between God and Atheism’). Sometimes when I take time off, I get lost in a rabbit hole or two. On Easter Sunday, I listened to a debate on my favourite Christian vs. Atheist debate podcast on whether you could demonstrate historically that Jesus rose from the dead. I Tweeted brief thoughts about this, and someone tweeted back at me about a recent six-hour debate on this topic by atheist New Testament scholar Bart Ehrman and Christian New Testament scholar Mike Lacona. I ended up paying $50 to watch all six hours of it, and would like to share some thoughts about it.
I found the debate incredibly frustrating. There was very little disagreement on the historical facts, apart from whether a miracle occurred. Both accepted that after Jesus died several of his followers had experiences which persuaded them that Jesus had physically risen from the dead (Ehrman thinks there’s good evidence that at least Peter, Paul and Mary Magdalen had such experiences). Rather than disputing the history, a lot of time was taken up arguing about whether it is possible for a historian, as a ‘historian’, to argue for a miracle. This seems to me a very silly thing to argue about. We could define the word ‘historian’ however we wish. Surely the interesting question is what we have reason to believe.
Moreover, there was also no mention of our mathematically precise was of understanding how evidential support works, namely Bayes theorem. According to Bayes theorem, the probability of a hypothesis is determined by two things: evidence and prior probability. The ‘prior probability’ is simply how likely the hypothesis is before we take the evidence into account. It seems clear that whether a case can be made for the resurrection is going to hang on the prior probability of the resurrection. Suppose you’re an atheist who thinks the odds of God existing are one in a billion. You’re obviously going to attach an even lower probability than that to God having raised Jesus from the dead. From that starting point, you’re going to need quite extraordinary evidence to get the probability of the resurrection up anywhere reasonable. And even if there is evidence for the resurrection that is fairly strong by the standards of ancient history, it’s clearly not that impressive. For this reason, I don’t think atheists should be worried about conceding to Christian apologists that there is non-negligible evidence for the resurrection. There is strong evidence for all kinds of things that we nonetheless have absolutely no reason to believe, precisely because the prior probability is so low.
Bayes theorem tells us that if a hypothesis renders the evidence less improbable than it would otherwise be, then the evidence supports that hypothesis. Suppose, for example, that Joan’s DNA was found on the body, and that that’s really unlikely unless Joan is the murderer. It follows that we have evidence that Joan is the murderer, as that hypothesis renders the DNA evidence less improbable than it would otherwise be. On the face of it, it is fairly improbable that several people would hallucinate Jesus being back from the dead, especially when one of them (Paul) was a zealous and violent opponent of the Christian movement. Whereas if Jesus really did rise from the dead and appear to people, then it’s not so surprising that those people would have experiences that persuaded them that he was risen from the dead. It’s pretty plausible, then, that the resurrection hypothesis renders the evidence less improbable that it would otherwise be, and that’s all that’s required, according to Bayes theorem, for there to be evidence for the resurrection hypothesis.
But that doesn’t mean, by any stretch of the imagination, that we should believe in the resurrection. Perhaps the evidence for the resurrection is strong enough to get the probability up from one in a trillion to one in a billion; this still ends up not being a hypothesis we should take seriously. Both Ehrman and Licona agreed that a skeptic about the resurrection needs to have some alternative hypothesis to explain the evidence. That’s simply not true. If we’re dealing with an event that ends up, even after the evidence has been taken into account, being very improbable, then it’s quite rational to say ‘I don’t know what happened, but that didn’t happen.’
To be fair, Erhman did press the old Carl Sagan line ‘Extraordinary events require extraordinary evidence’, arguing that the resurrection is very improbable because it would require violating a law of nature.* But without bringing in Bayesian notions, this is just a rhetorical slogan, and it ended sounding like Ehrman took naturalism to be unfalsifiable. The only mention of Bayes in the whole six hours was Licona saying that evidence can overcome low prior probabilities. That’s right, in principle. But given that, what this debate should have been about is:
(A) What is the prior probability?
(B) How strong is the evidence?
(C) What probability do we end up with as a function of (A) and (B)?
Because they didn’t engage with the Bayesian framework, this discussion spent six hours getting nowhere.
Improbable things happen. And when they do, we often end up with evidential support for whacky hypotheses. If Christian apologists want to make a case for the resurrection of Jesus, then they need to argue for a worldview in which the resurrection of Jesus should be assigned a non-negligible prior probability.
*Actually, I was more on Lacona’s side on the specific point of whether a miracle would violate the laws of nature. That’s because I follow my colleague Nancy Cartwright in conceiving of the the laws of physics as ‘ceteris paribus’ laws, which tell us what would happen in the absence of other causal factors. I’m to be arguing with Sean Carroll about this (again!) on next month’s Mind Chat.