I’m sure I’ll get back to Nagel in the future, but, for the time being, I’d like to start another series of responses that is more directly relevant to the question of God’s existence. This time, to J. L. Mackie’s “Miracle of Theism”. This has been called the best philosophical discussion of theism from an atheistic perspective, and is a much more serious and well-reasoned book than any of the popular-level atheist tomes. As such, I find it easy to respect Mackie, even as I disagree with him.
Still, I do disagree, and want to get to the reasons why:
Mackie, after a very reasonable introduction that shows a charitable attitude toward the theist position, begins with a discussion of David Hume’s famous argument against belief in miracles.
For those that don’t already know, the basic thrust of Hume’s argument is this: Since miracles are occurrences that (by definition) are astronomically improbable, our reason for doubting that a miracle happened is always stronger than the weight of any person’s testimony saying that it did happen. This would mean that we should never believe that a miracle actually happened.
Though he isn’t without criticisms, Mackie finds this a good objection to miracles as evidence for God. He basically agrees with Hume and, I think, fails to raise the two most potent responses to the argument.
The first is less a disagreement than a point about what this actually proves. If successful, this argument would block any attempt to prove God’s existence on the grounds of testimonial evidence of a miracle, but I know of no theist philosopher who argues for God’s existence on those grounds.
Some have, after granting theism, looked at evidence for a miracle in order to determine which form of theism is most likely to be correct. But this is a far cry from using claims of the miraculous to argue for God’s existence. Of course, one can try to say that it is still a mistake, but even this is untrue in light of the second response.
That is, Hume’s argument only succeeds if one presumes either atheism or deism. The only reason why Hume can say that a miracle is astronomically improbable is because he’s assuming that the regular patterns of the universe are never interrupted (and that it is astronomically improbable according to those laws). That is, he’s assuming that there is no God which intervenes in history–or, at least that it is astronomically improbable that such a God exists.
Under that assumption, a thing like a resurrection is indeed astronomically improbable, but this assumption is true only if classical theism is false. Thus, Hume is simply begging the question against classical theism.
And I find that this is a common mistake, owing in part to a misconception of the concept of a miracle. At least, many who press this argument seem to think (as Mackie seems to think) that a miracle is “a violation of the laws of nature”. In fact, a miracle is the introduction (or removal) of matter or energy into the universe through divine means. The laws of nature are not changed or “violated”, they are simply acting on altered conditions.
To put it another way, the theist claims that the universe is not a closed system. The atheist is free to reject this view, but is not free to assume a closed system in making an argument about the worthlessness of miracle testimony–which is precisely what Hume’s argument does.
As fair-minded as he strikes me, none of this seems to occur to Mackie, who fails to mention any of it in his discussion of Hume’s argument.
That being the case, I’m left with the conclusion that we need to find a balance between blind trust and absolute incredulity. There is no one-size-fits-all answer to every claim of the miraculous, and no reason to simply dismiss a claim simply because it entails that something more than the physical was present in an event.
That would be to assume the atheism that one is trying to prove, meaning that the argument is simply worthless with respect to miraculous claims–let alone the other arguments for God’s existence.
But, if I completely disagree with Mackie here, I largely agree with him in the next sections. I’ll get to those soon.
Fide Dubitandum 
November 10th, 2013 at 5:22 pm
I think that this entire discussion hinges on one’s epistemology http://lotharlorraine.wordpress.com/category/extraordinary-claims-demand-extraordinary-evidence/ .
Atheists using a Humean approach generelly apply Bayes theorem to all claims and theories and conclude that since the prior probability of a miracle is extremely low, only extraordinary evidence can make it plausible.
As for me I doubt that one can view degrees of belief in something like a probability:
http://lotharlorraine.wordpress.com/category/probabilities/
Cheers from Europe.
November 10th, 2013 at 5:35 pm
Cheers to you as well!
I definitely agree that epistemology is a major factor. Still, Bayes theorem wouldn’t help the atheist (in fact, it would help the theist) specifically because it makes this mistake more obvious.
To put my argument in those terms: the background probability of a miracle occurring depends greatly on whether or not classical theism is true. As such, a claim of “astronomically low” presupposes the falsehood of classical theism–making the argument circular.
But, I’d completely agree with your idea that talk of probabilities here seems misguided.
November 10th, 2013 at 5:40 pm
Atheist Richard Carrier says that even if theism is true, we know that the FREQUENCY of a resurrection is extremely low.
Therefore it’s prior probability is also extremely low and for him it is even exactly the same number.
What would you say about this?
November 13th, 2013 at 10:26 am
Paarsurrey says:
Jesus did not die on the Cross; he survived a cursed death on the Cross though he was put on the Cross. Jesus did not perform a resurrection ; so why insist on resurrection as a miracle of Jesus by the Christians.
The real miracle is in survival of Jesus from the death on the Cross against all odds; this should be focused by the Christians as a miracle and offered as a miracle to the Atheists.
November 13th, 2013 at 10:21 pm
I suppose the debate over which of our respective religions is correct on this point is far too large to address here.
I would say, though, that Christians could never defend that as a miracle, because holding that position would make one (by definition) a non-Christian (whether a Muslim, or a non-Muslim who agreed with the Koran on this point).
Last, but definitely not least, I appreciate that you’re offering a perspective that, in my opinion, is not represented enough. (Keep that coming!)
November 10th, 2013 at 6:03 pm
Well, the first thing I’d say is that it’s clearly a thoughtful response, and that I can understand why someone would say it.
Getting more to answering, I have four basic lines of response to that.
1. I’d underline that there is a clear difference between “extremely low” and “astronomically low”.
Psychologically, I tend to pile both of those into the same “just assume it won’t happen” bin. But the difference is significant in that testimony can often overcome the former problem, but not the latter.
A quick example would be a news report. None of us would seriously dismiss the news as incredible if it reported that, say, Donald Trump was shot–even though that is an extremely unlikely event on any given day.
2. I’d reference other arguments
In a particular case (such as any argument for the divinity of Jesus Christ), other arguments would make it more reasonable to believe that something which God does not typically do would be done in that instance.
A good example of this might be a criminal suspect. Even given the background information that a crime occurred (parallel to God’s existence as background information), it’s pretty unlikely that any one person is the killer. But, if there is other information which points to a particular person, it is more reasonable to think that he is the killer than it would based on the background information alone.
3. The frequency of any event is low
We all know the likelihood of any specific thing happening (a particular lotto pick, a tire going flat on just that spot on the road, etc) is very low. And, not incidentally, it is testimony that teaches us this. This is why Bayes theorem takes account of the probability of testimony being false. That must also be accounted for, and (at least in many cases) further justifies the testimony of an infrequent event.
4. I’d ask him what view he is defending
It’s a bit of a personal frustration that debates often lose track of the fact that we need to select the best option on the table. And many of the atheists I’ve debated often use arguments that do not support atheism in defense of their atheism.
So, while I completely agree that this is something that religious people should think about, it would only defend a position like deism or theistic agnosticism (a theist who is undecided about a particular religion). This is not an argument that atheists, particularly metaphysical naturalists like Richard Carrier, should be using in defense of their views.
In any case, I hope that didn’t ramble on too long. I can’t always help myself.