Brother Daniel

What significance does the supernatural hold for history?

I ask this question because of the common Christian complaint about presuppositions by skeptics. In particular, it is frequently alleged by Christian apologists [1] that those who deny the historicity of biblical miracles (in particular, the bodily resurrection of Jesus) do so because they have presupposed that such things simply do not happen, and are therefore closed-minded against the evidence that such things in fact did happen.

Is that a fair charge? This month's (April 2004) featured article [2] on the Secular Web points out, quite convincingly, that complaints of presuppositionalism made by Christians against skeptics are usually hypocritical, and, perhaps slightly less convincingly, that non-traditionalist scholars usually do not actually presuppose that there are no miracles. Despite my admitted doubts about the latter claim, I will not try to address what historians do or do not presuppose, for I am no historian, and thus I do not have my finger on the pulse of the matter.

Instead, the question I will try to address is: What ought we to presuppose about the supernatural in history? To begin to answer the question, let's take a look at how presuppositions in general ought to affect our consideration of evidence.

Calling Reverend Bayes

If we consider the relative probabilities of claims A and B in light of evidence E, we have the following formula [3]:

p(A|E) / p(B|E) = (p(E|A) p(A) ) / (p(E|B) p(B) )

Here, p(A|E) should be read as "the probability of claim A, given evidence E", and p(A) should be read as "the probability of A, without considering E".

To ground the discussion, one might take A to be the claim that Jesus rose bodily from the dead, and B to be the claim that Jesus had not died on the cross but recovered in the tomb – or any other alternative theory about Jesus that you prefer.

An important consequence of the formula is that a better fit with the evidence can be balanced by a worse prior probability. By "fit with the evidence" I refer to the terms p(E|A) and p(E|B), that is, the probability of the evidence, given the claim. If we assume claim A to be true (for the sake of argument), then how likely should it have been for the evidence E to turn up?

I do not propose to insert actual numbers here. In history, such matters seem usually to be handled intuitively, not quantitatively [4]. Nevertheless, the pattern is important. If one feels that evidence is a "slightly" better fit for A than for B, but A is a priori "much" more unlikely than B, then it is more rational to go with B. Alternatively, if one feels that the evidence is a "much" better fit for A than for B, but A is a priori "slightly" more unlikely than B, then it is more rational to go with A.

But what prior probabilities are appropriate? The question is especially vexing when dealing with the supernatural. If we do not assign extremely low prior probabilities to the supernatural, then why are we calling it "supernatural"? What does supernatural really mean, anyway?

To study the non-miraculous

When I was a Christian, I used to describe science as "the study of the non-miraculous". Science deals with that which is repeatably observable, especially in controlled conditions. No matter how many times a consistent result is obtained for a given experimental setup, there is no logical justification for the claim that it can never go otherwise. When dealing with an omnipotent God, there is no reason to expect that God will co-operate with your experiments, so He is clearly outside the realm of what science can possibly study. And so are all His works, should He choose to interfere in human history.

My cosmos, then, was modelled thusly: The physical universe has rules according to which it normally functions – Natural Laws – but God can (and occasionally does) interfere and make Grand Exceptions. Thus anything is possible (barring logical nonsense), but not anything is naturally possible. An event that breaks Natural Law can be termed "supernatural".

This definition of supernatural, in a Christian framework, leads to serious problems when trying to sort out claims of the supernatural in history. To see this, suppose that historians are arguing about two or more competing non-miraculous claims. In this case, no one is offended by an appeal to science to sort out what kinds of things are likely or unlikely to happen, and thus what the relative prior probabilities are. By contrast, the supernatural, as presented here, depends on the whim of God [5]. One needs to have some understanding of God, surely, in order to make any kind of assessment of prior probabilities of natural versus supernatural claims of past events (or of multiple competing supernatural claims, for that matter). Yet the unfathomable nature of God is an essential part of the traditional Christian (or Jewish) worldview – we are hardly permitted even to imagine what God is likely to do. How, then, can we make a guess at the prior probabilities of supernatural events? How is it even possible for a human to analyze the possibility of supernatural events? A skeptic could be forgiven for suspecting that the primary purpose of supernatural claims is to throw people off-balance!

The Parable of Newton

A better way of dividing the natural and the supernatural is to define natural phenomena as whatever is possible within the current consensus of science, and supernatural phenomena as whatever is not. (The open questions in science form a grey area between the two.) This division may seem slightly counterintuitive, since proponents of astrology (for example) consider the influence of the planets on our lives as a natural matter, rather than a supernatural one. But for now, let us consider such things as supernatural.

This view of the natural/supernatural divide should be acceptable to Christians, for it is supported by the response of characters in the biblical narratives to reports of miracles. Miracles that are "predicted" in the Bible (according to the sequence of the narrative as written) are often met with disbelief by the recipient of the message. Sarah "knows" that she cannot conceive at age 90. Mary "knows" that she cannot conceive without any sexual contact with a man. But the messenger typically points out that whatever is normally "impossible" is nevertheless possible with God. The underlying message is that nothing (barring logical contradictions) is to be ruled out absolutely, yet the writer implicitly gives the nod to the notion of "natural impossibility" – based on the ordinary experiences of ordinary people (which is the closest thing to "science" that would have been available to the biblical writers).

More importantly, this way of drawing the line between the natural and the supernatural has an important precedent in the history of science [6]. Newton's proposed law of universal gravitation was met with criticism (by his rival Leibniz) on the grounds that Newton was introducing the supernatural into his theory. The criticism carried weight, for already by Newton's time, the workings of God or of angels had retreated into the realm of exceptions to natural law, rather than as explanations for natural law. Today, we would not dream of describing Newton's theories as supernatural; their precise mathematical nature accords well with what we would term "natural" and even "mechanical". But in Leibniz's view, all the motion-inducing forces of nature had to involve objects in direct contact with each other, and Newton's "action at a distance" suggested the supernatural. (Perhaps the difference between Leibniz and us is simply that we have gotten used to Newton!)

For all these reasons, I will adopt this view of what "natural" and "supernatural" mean. But what are the implications for our assessment of prior probabilities?

The necessity of Natural Theology for historical apologetics

The central question to be asked, for purposes of sorting out the problem, is this: Are supernatural events essentially comprehensible by humans?

If we answer "no", we have embraced the idea of Divine Inscrutibility. We allow supernatural events, as whims of an unfathomable God. By rendering impossible any rational assessment of prior probabilities, this option undermines the very possibility of rational examination of the implications of any historical evidence. The Christian Story is therefore unsupportable (even if plausible), for literally anything goes. Christian historical apologetics are excluded, and thus Christian claims that the evidence demands acceptance of the resurrection are immediately falsified.

If we answer "yes", then we are adopting a Natural Theology. We treat theology (or atheology) as a starting point, and allow ourselves to judge alleged supernatural events in light of our theology in the same way as we judge alleged natural events in light of science. In a sense, the supernatural world is treated as an extension of the natural. We judge, according to our theological preconceptions, what kinds of miracles, in what circumstances, are likely or unlikely to happen. This is uncomfortable for the Christian, for it involves us humans in the process of judging what does or does not make sense for God to do. Nevertheless, any rational Christian historical apologetic must adopt this approach.

Hardcore Skepticism – in which one presupposes against anything that appears "supernatural" according to science, unless and until the progress of science itself absorbs the kinds of phenomena in question (as in the parable of Newton) – is simply a special case, deistic or atheistic, of Natural Theology. The traditional Christian position is immediately excluded, for there is no serious scientific theory according to which dead people come back to life after more than a day (then walk through walls, appear and disappear at will, and ascend to heaven).

Naturally, the Christian will cry foul when confronted by Hardcore Skepticism. But if these presuppositions are bad, then one ought to ask what presuppositions lie behind the too-easy Christian dismissal of the alternative explanations of the resurrection story presented by skeptics – such as the swoon theory [7] or the Mythical Jesus theories [8]. Does a Christian use naturalistic arguments – such as "people wouldn't do that" or "that kind of thing never happens" – in order to dismiss a skeptic's alternative explanations? If so, it is not even consistent to complain about the presuppositions of Hardcore Skepticism – for the skeptic is asked to suspend her view of what is possible (or not), while the Christian does not submit to any such obligation. Nevertheless, this approach is clearly not useful for purposes of having a discussion with a Christian.

But some further general comments about (any) Natural Theology are in order. Here, our thoughts about the possible nature of God take centre stage. It is not rational, under Natural Theology, for us to do what a typical Christian apologist wants us to do: first, to recognize strong evidence for Jesus' resurrection; second, to respond to the evidence by believing that it actually happened; and third, to adopt some form of Christian theology in order to give the event its relevance. Consideration of the historical evidence cannot be treated as independent of the theological (or "philosophical") questions raised by it, because those theological questions strongly impact one's view of the prior probabilities of Christianity's historical claims. Thus, for example, an unbeliever's objections about the moral bankruptcy of the idea of "substitutional atonement" must be admitted into court, so to speak, in any rational deliberation of whether Jesus really was a Son of God who died for our sins and rose again. In short, if Christian theology makes no sense to you, then it is not rational for you to accept the Christian view of history with its supernatural claims – unless the historical evidence is utterly overwhelming, as historical evidence can never be. If a Christian tries to play the "unfathomable God" card, to say that we should not expect to understand God's ways, to suggest that the fault is with you rather than with their theology, then they have slipped back into the Divine Inscrutibility option, thus undermining their entire enterprise.

In conclusion, if your Internal Nonsense Meter reads "high nonsense" in response to Christian theology, then there is no rational way to use historical evidence to override it.


Notes:

1. I hardly need to dig up a reference - it is easy to find this accusation in the works of C S Lewis and J W Montgomery; Price (see note 2 below) responds to this charge as leveled by W L Craig; and it turns up in the online work of the insufferably rude J P Holding.

2. Robert M Price: "By This Time He Stinketh", http://www.secweb.org/asset.asp?AssetID=350

3. This follows directly from Bayes' Theorem.

Currently, there appears to be a cease-fire in the ongoing battles between Bayesians and non-Bayesians. To the former, "probability" is the end-all and be-all of all discussions of uncertainty, and the formula I have given is generally applicable. For a good argument for Bayesianism, see Scientific Reasoning – The Bayesian Approach by Colin Howson and Peter Urbach.

To their opponents, there is some question about whether "probability" is an appropriate way of dealing with all kinds of uncertainty – perhaps it should only be used for "chance", in the context of repeatable random experiments, and used sparingly (or not at all) for the more general concept of "partial belief". For an intriguing alternative to Bayesian probability, see A Mathematical Theory of Evidence by Glenn Shafer. This (Dempster-Shafer) approach has intuitive appeal because it treats symmetrically the prior information (or prior assumptions) and the new evidence, and also because it allows a more natural expression of admitted ignorance.

Since the present argument is merely qualitative, the differences between Bayesian and non-Bayesian approaches are not important here. Either way, prior information cannot be ignored when assessing the importance of added evidence.

4. On the other hand, see Richard Carrier's article, "Why I Don't Buy the Resurrection Story" (http://www.infidels.org/library/mode...on/index.shtml), in which the author calculates (conservatively) a probability for a man who has been crucified, presumed dead, and entombed in 1st century Judaea, to recover in the tomb and escape.

5. Or of other non-corporeal intelligences, ignored here for the sake of simplicity.

6. I wish I could remember where I came across this story. It was in a book about the philosophy of science. Unfortunately, I've moved thousands of miles away from the chap from whom I had borrowed the book.

7. See note 4 above for a good variant of this idea.

8. For example, The Jesus Mysteries by Timothy Freke and Peter Gandy presents the Jesus story as a Jewish adaptation of the then-widespread Mystery Religions with their mythical dying-and-rising godmen.

GakuseiDon

(I'd made an earlier attempt at replying, but unfortunately the intenet ate my post!)

BD, I'm a liberal Christian, and a big fan of CS Lewis. I'm fairly agnostic to claims of miracles.

Yes, CS Lewis did make that complaint about people making presuppositions. But isn't it valid?

What you've raised here is a strawman, at least as stated. BD, if someone evaluates evidence based on their presuppositions, isn't it valid to complain about that?

You're confusing two things here: historical evidence, and scientific evidence.

Science can be used to, say, evaluate the age of a manuscript. It can't necessarily be used to, say, evaluate the manuscript's claim that A did B.

A lot of claims are not examinable by science. Miracles, by their very nature, usually fall into this class.

A nitpick: I think a better term may be "metaphysical" rather than "supernatural". Plato's idealised forms are not supernatural.

I'm sorry, this makes no sense. "Comprehensible"? Is a miraculous healing essentially incomprehensible? No. The method may be, but not the result. Is Christian apologetics involved in using historical evidence to prove the method?

It doesn't make me uncomfortable. If miracle healings today are proved to occur, this probably increases the likelihood that they occurred earlier. And vice versa.

"Too-easy Christian dismissal of alternative explanations" and "the Christian does not submit to any such obligation"? This is a strawman, and a mildly offensive one at that. I went to Note 7 expecting to see an example of such a dismissal, but nothing on that there.

Sure, some Christians dismiss alternative explanations without much thought, but this can apply to any group. I'm not sure what kind of debate you are having with them, but if it is not one based on evidence of claims, then I can't imagine it being a very satisfying one.

What if the historical evidence proves the claim of the Christian? Do you still reject it?

Hopefully you see the problem. "No amount of historical evidence will prove a supernatural claim" is probably true, but that is due to the limitations of historical investigation, not to some inherent impossibility in the supernatural. A time machine may very well clear up a few claims of quite a few reported historical supernatural events.

Vorkosigan

Don, does that mean that when my kooky neighbor claims that the parking space she found right in front of the store was the result of her being guided by God, she has sealed off that event forever from scientific explanation?

A MUCH better way would be to regard the supernatural as any claim that an entity can operate on matter by direct action of the will. Most entities with supernatural accreditation are simply disembodied minds (God, ghosts, pixies, tree-spirits). The supernatural is actually just the old mind-body duality blown up to the cosmic scale. Belief in the supernatural is essentially belief in other minds....and is the result of our innate cogntive machinery for thinking about how other minds act.

You cannot go by the current consensus of science because you have not specified how that consensus is constructed. In other words, scientists have some way of demarcating the natural from the supernatural, and it isn't by vote.

Now that is dead on. Your essay was very nicely written, BTW.


Vorkosigan

GakuseiDon

Yes, in that case. At least, I can't see how that particular case could be investigated scientifically.

Vorkosigan

Well, for starters, a scientific inquiry might ask how the woman acquired this knowledge, and what evidence she was for believing it.

I mean, taken to its limit, if I claim that everything in the universe is sustained by a miracle, does that mean that we have to stop science?

Vorkosigan

Brother Daniel

I'm a little confused by your calling this a strawman. You appear to have agreed that Christians (at least sometimes) make the complaint in question. So how is it a strawman?

I raised this issue in order to present my motivations for thinking about the question: What ought we to presume about the supernatural? The question is important, for I don't see how it is possible to proceed without any presuppositions.
On the contrary: The distinction between historical evidence and scientific evidence was kept in mind throughout.
"Can't necessarily": correct. But in some cases it can. For (partial) example, if we have a manuscript by a known author who is mostly reliable, and one historian takes the manuscript's claim (that X happened) at face value while another historian suggests an alternate set of events (Y) that could easily have led the author to think X happened, then our experiences (scientific or otherwise) about whether X-like or Y-like events are the more likely will impact our assessment of the prior probabilities of those competing claims.

We cannot proceed (in evaluating conflicting historical claims) without some kind of idea (at least intuitively; not necessarily quantitatively) of the prior probabilities of those claims. What science can do for the natural, theology (or metaphysics) must do for the supernatural.

Agreed. But I don't know what point you're making here.
But neither are they "phenomena". The whole discussion of the relationship of the natural to the supernatural is a metaphysical one, but I still think "supernatural" is a better word to describe (alleged) phenomena that go beyond the "natural".
Perhaps I was unclear at this point (for which I apologize). No, I was not referring to the "method" of a miracle. If I make a claim of a miracle, then my claim is "comprehensible" (in the sense I meant it) if you can evaluate whether it "makes sense" for that miracle to happen in that circumstance.

In a Christian context, this (almost) reduces to the question of whether it "makes sense for God to do that". For example, a liberal Christian might argue that it does not make sense for Jesus to (miraculously) kill a fig tree in anger at its barrenness, when it isn't even fig season in the first place! If a conservative Christian responds with "who are you to question God?", then the latter is embracing the Divine Inscrutibility option.
If miracle healings today are proved scientifically to occur, then they can be treated as "natural", according to the way I drew the demarcation between natural and supernatural. Otherwise, you are stuck with evaluating what does or does not make sense for God to do, according to your theology. That may not make you uncomfortable, but it does so for some Christians. "Who are we to know the mind of God? For His thoughts are not our thoughts." (I'm pretty sure that's from the Bible, but I can't remember where exactly.)
For a "too-easy Christian dismissal of alternative explanations", check out John Warwick Montgomery's booklet, History and Christianity. The swoon theory is simply waved off as "infinitely more improbable than the resurrection itself" with no indication of why it is improbable. And that dismissive attitude is, in my experience, very common in Christian discussion. But clearly not all Christians are so bad, and I probably should have been more careful in how I worded that paragraph.

You may be right that I shouldn't have said "the Christian does not submit to any such obligation", and I will retract it. But there is still a potential problem of consistency in any defence of miracles. Anyone who asserts a miracle has already pushed nature aside, and it is inconsistent for such a person to turn around and appeal to nature in order to refute an alternative explanation offered by a skeptic.

And that is all I was claiming.
An interesting thought. But I'm not convinced. To put this into perspective, suppose that you personally experience a supernatural event. It might make sense for you to believe it, since your experience is first-hand. But would it make sense for me to believe it if you told me about it? I would somehow have to weigh the probability of the miracle against the probability of your being mistaken or of your telling me a lie. After all, human mistakes and lies happen all the time, and there is nothing particularly amazing about them.

C S Lewis liked to claim that if a person is generally reliable, then their word should stand no matter how amazing their claims. Thus, in The Lion, the Witch, and the Wardrobe, Professor Kirke tells Peter and Susan that they should believe Lucy (who is generally reliable) over Edmond (who is much less reliable) with respect to their respective claims about what they did or did not find through the wardrobe. Peter objects, "but if something is real, it's there all the time." (Recall that Peter and Susan have already tried to check out the portal to Narnia but have found only the wooden back of the wardrobe.) Kirke engages in Proof by Intimidation in order to get Peter to back down on that claim. So Kirke considers human sanity and honesty to be far more reliable than Object Permanence, even though the latter is firmly entrenched in everyone's experience – and he pushes this whopper of a fallacy on the children even while blustering about the lack of instruction in logic in current schools. The reader, of course, is distracted by the fact that Kirke's conclusion happens to be correct within the framework of the story. Lewis was certainly a skillful writer, no doubt about that!

Brother Daniel

Yes, that could work. I was holding open the possibility that some such phenomena could, in the future, be proved valid. If science absorbed such things, they would then become "natural" (see the parable of Newton), and would no longer need special treatment as historical claims. But there was probably no need for me to do so.
As long as one can identify what the consensus is, it needn't matter (for my present purposes) how that consensus was reached. In fact, scientists could collectively make a blunder here and there without it impacting my argument.
Thank you.

GakuseiDon

You're right - it wasn't really a "strawman". Still, I can't what the problem is with complaining about presuppositions when investigating something.

In a scientific investigation, there is nothing wrong with establishing premises, if that's what you mean. "Presuppositions" presuppose the answer.

Our experiences form part of emperical evidence. They help to establish premises. I don't disagree with you, though.

OK, that makes sense.

No problem, it is hard not to generalise sometimes in these debates.

I honestly don't see why not. "When all other possible explanations have been disproven, whatever remains, no matter how improbable, must be true".

The only way to prove a supernatural event is to dispel all natural explanations. But if you can't appeal to the natural world, or if natural explanations (no matter how improbable) will always take precedence over supernatural claims, how do you prove a supernatural claim? I suppose you could say "with overwhelming evidence", but wouldn't part of that evidence be ruling out natural explanations?

But would you believe it even if you experienced it first-hand? Hallucinations aren't particularly amazing either... CS Lewis once wrote that some people could appear before God and be thrown into the lake of fire, and all the time be convinced that it was just a hallucination.

Brother Daniel

I'm not convinced that there's such a clear distinction between premises and presuppositions. They're practically synonymous – each has shades of meaning that include the other.

But (as I argued in the OP) there is no way of dealing with the supernatural in history without having some kind of premises that other people will find questionable. A rational argument about the supernatural in history, between people who hold different premises, will have to veer off into non-historical questions in order to sort out the relative merits of the participants' respective sets of premises. The historical evidence alone will never be enough, for the different premises about the supernatural will strongly impact one's interpretation of the evidence.

There are some Christians who say that all historical claims are to be treated equally: that is, that supernatural historical claims are to be treated the same way as any other historical claims. But this approach will not work in support of the traditional Christian claims. As I pointed out in the OP, if we are comparing ordinary (non-supernatural) historical theories, our assessment of which theory is better will be based partly on our experience of how the universe (including the people in it) normally works. But the supernatural, by its very nature, transcends our experience of how the universe works*. The whole point of a supernatural claim is to overrule whatever our usual experience of the world would have us believe. So the only possible ways to treat all historical claims equally are to presuppose against the supernatural, or to throw up our hands and give up on history entirely.

The only alternative to these, as I argued previously, is to take as premise a theology that allows you (in principle) to assess prior probabilities of supernatural events. But if we take this approach, we would not be treating all historical claims equally, and we will certainly not all get the same answers. (The hardcore-skeptical view can be regarded as the simplest possible theology.) It remains for the Christian to show how his/her theological premises are better than those of the skeptic. Whatever premises we bring to bear on the question of the supernatural in history, they will (at best) depend ultimately on non-historical arguments. History fails as the foundation-stone of Christian apologetics.
The Sherlock Holmes principle doesn't work for the Christian, for history never renders anything impossible, only improbable. The Christian needs to invoke Dirk Gently, another fictional detective: "The impossible sometimes has a certain integrity to it that the merely improbable lacks."

There are many possible natural explanations for the existence of the NT as we have it, and a Christian might (at best) be able to argue that these explanations are improbable. The problem is in rendering them sufficiently improbable to make a supernatural explanation appealing. We all have our "mystery thresholds" set at different levels. I would suggest that anyone whose threshold is both consistent and set low enough to be swayed by Christian apologetiics will see the supernatural everywhere, even in everyday natural phenomena; conversely, one who is swayed by Christian apologetics while otherwise living as an apparently sane modern person will be treating Christian claims in a fundamentally different way than they are treating everything else. But that's just my opinion.
You simply can't prove a supernatural claim. Allowing that supernatural phenomena may exist in general, it does not follow that it is ever rational to believe in any specific case of them. So it makes no sense (in my opinion) for God to set things up in such a way that people "must" believe in any given supernatural historical claim, even if the claim happens to be true. If there's a God who cares about what we believe about Her, then the prescribed beliefs must be accessible through our individual experiences. Rational theism (if there is such a thing) must be Gnostic, I think. (But that's getting into GRD territory.)
It's true: Hallucinations ought to be considered as possible explanations for weird experiences. But I think this comment from Lewis is a straw man. Generally, it's the skeptics who trust their own experiences, and the traditionalists who tell people that they must override their own experiences and believe instead what the Authorities tell them to believe. By making it look as if unbelief is a goal that skeptics will pursue at any intellectual cost, Lewis is simply engaging in projection. (His essay on "Obstinacy in Belief" is revealing in this regard, if my memory of it is accurate.) But of course, if unbelievers go to hell, the traditional Christian cannot allow himself to recognize the fact that unbelief can be reached honestly, and skeptics must therefore be portrayed in this unfavourable light. So Lewis's comment is not surprising.


* footnote: This statement is a consequence of the way I defined "supernatural" in the OP (i.e., that which falls outside of whatever the current consensus of science allows) for I take science to be the measure par excellence of "experience". If I took Vorkosigan's suggestion and defined "supernatural" in terms of specific kinds of phenomena, then my present argument would have to be much more complicated in order to work.