OrangePhilosophy

As for the epistemology, I suppose I am far more sympathetic to Hume's position. It is bad enough to believe there are miracles. It is far worse to believe of an event that it is miraculous.

I wonder about Hume's remark that only the exceptional deserves the appelation 'miracle'. I can't see any good reason to insist on this. I've heard similar remarks to the effect that a miracle must involve a violation or suspension of the laws of nature. When I hear such things, I tend to nod but of late, I wonder why. Suppose one held a compatibilist view about free or responsible action. Shouldn't one think that God could intervene without violating/suspending the laws? If one holds that miracles are primarily acts of divine intervention, perhaps the bit about laws is a distraction. Perhaps the same should go for the exceptional and unexpected. When we observe some unexceptional event, it would never cross our mind to call it a miracle (Hey, my keys! There is coffee in the pot!) but it isn't as if it is true by definition that all miracles are surprising, unexpected, sexy, etc. When it comes to saying what the essence of the miracle is, I'm not convinced by Hume's remarks.

Well, 'tis a little tricky. Basically they seem to be arguing for constraints on probability distributions and updating: Hume seems to be saying that you ought to distribute conditional probabilities in ways that don't allow you to raise your subjective probs in miracles very high. Chesterton seems to be saying that you should accord testimony, without good evidence against it, a great deal of weight.

For my part, Chesterton seems like a bit of a nut. Of course what the person says may affect your general trust in them. What they say will raise your beliefs about their trustworthiness. Why would you accord the same credence in someone's claims about where their keys are as you would if they claim that they have seen ghosts. You think it is unlikely that there are ghosts so you judge their likelihood of being right against that; big suprise.

If you were certain (not necessarily knowledge but certainty) that there were only 5 diamonds in the world but 5000 things that could fool one into thinking there were diamonds, and someone claimed to see a diamond, you would surely reckon them to be likely to be wrong. Presumably something liek that holds iwth respect to ghosts m=but not with respect to mmore mundane things.

Interesting parallel you've found and worth comparison. Credibility and trustworthiness might not convince us (of miracles) although it provides reasons for us believe in them. Imagine a trustworthy family member telling you of the supernatural thing they've seen. We praise the person and her trustworthiness. We believe that the person has seen something and that they say it was supernatural. This falls short of believing testimony in favor of something being a miracle. Maybe testimony is a dead end sometimes. Many will have us think this is all there is. I'm not sure that's right. As for natural law, I think Hume is right to narrow the definition of "miracle". It appears that for Hume's time breach of natural law just means supernatural/divine intervention. You're right Mark, it's got to be a suprise (otherwise we might predict miracles, which might mean there is a natural basis for the occurrence). I would add that it be beneficial also. The rigidity of Hume's criteria are evened-out by the utter cluelessness we have regarding the "laws" of nature. At subatomic levels, say, can we really know what is perfectly natural? So if Chesterton want us to be democratic about the issue, I say it's in admitting our ignorance of "natural law"-- where quantifiability starts to drop out.

Chuck, predicting accurately may not require a natural basis. For instance, a prophet genuinely interacting with a divine source of knowledge doesn't seem to be natural but would be an accurate predictor.

One of my biggest problems with this whole issue is what counts as a miracle. Leibniz, for instance, wouldn't have any of this breaking laws of nature business. Malebranche wouldn't either. The laws guarantee what will happen, including things link miraculous healings, resurrections, angelic visions, partings of rivers and seas, and prophetic pronouncements. The laws aren't merely the laws of physics, however. For both of them the laws of nature would include ways God controls the universe beyond what we can learn about at the level of physics.

Alternatively, Plantinga's view of miracles is that they involve transmutation of kinds of matter with the laws staying fixed. So if God miraculously saves someone's life by having their car go right through the one in front of them with no damage to either, the laws aren't broken but the matter is transformed so that it can do things like that. That sounds pretty miraculous to me but involves no laws being broken. It doesn't seem to be about mere surprise, however, because the prophecy example seems just as miraculous but not surprising if you know what's going on (that it really is a message from God). Rarity wouldn't do it either, because lots of rare things aren't miracles.

The other problem for Hume is his view of what laws are. By his definition, the laws can't be broken. He defines miracles out of the realm of possibility. The general Humean picture of laws is that they're whatever the most simple expressions of what actually happens across time. So if something happens, the laws account for it. There could be no miracles, then. If the accounts in the Bible are true, then the crossing of the Red Sea and the turning of water to wine are not miracles. No law is broken. We just have the laws wrong if we think the laws don't allow that sort of thing. The laws have to be much more general or have some disjunctive features to allow any such events.

I think the Humean account of laws makes no sense. Laws that don't make things happen or prevent things from happen don't make any sense to me. Humean laws are true only because of what ends up happening, and laws seem to me to be the reason for what happens rather than the other way around. Still, the same problem arises for what I consider to be more traditional accounts. If laws are what guarantee how things will go, then there won't be broken laws. If laws are only tendencies, that would be different, but you just get indeterminism then, and indeterminists don't want to think of everything not guaranteed by the laws as a miracle. They just want to say that the laws leave some things undetermined. People who think of miracles as breaking laws aren't using any of these accounts of what laws are. For that reason, I don't think it makes any sense to see miracles as breakings of laws. Hume's entire argument, if it has any merit, would have to be thoroughly reworked given all this, but you'd need to have some account of miracles to direct that argument against first, and I have none, so I don't know how that would go.

Ultimately, I think theists should hold that every event is a miracle, at least every contingent event dependent on God's creation (and God's continuous creation). I'm not sure that view is subject to Hume's criticisms no matter how he would reframe them. You could distinguish between different kinds of miracles, perhaps, but I think those debates will simply reduce to debates about naturalism or theism rather than about something special that we call miracles.

JP,

Good point about Hume's view on miracles and laws being in tension. I always used to wonder about that too: you can be a priori certain that there are no miracles given what Hume says. I have no idea, however, what it means to say that laws make things happen: could you say a little bit more?

Two things come to mind that lie behind my sense of what laws must be. One is from the standard definition of determinism. Any state of the world and the laws guarantee any later state of the world. Now you can frame determinism in terms friendly to Humean views of laws, but I think the spirit of the definition is that the past and the laws make sure that things will be the way they are later on. It's not as if they're agents. It's more that they govern what happens, and if determinism is true they rule out all options but the one that will indeed happen. On the Humean view, it's as if the laws are an after-the-fact declaration about what could happen that isn't just epistemically arrived at by looking at what in fact happens but is even metaphysically derived from what happens rather than influencing what happens, as laws seem to me to be most basically about.

The other place where I think this view of laws makes more sense is in discussions of the problem of evil. Malebranche responds to the problem of evil, for instance, by saying that a world with simpler laws is aesthetically better than a world with complex laws. He then argues that a good God would appreciate the beauty of the natural order and count that as a good-making feature with respect to allowing evil, enough so that if the laws determine that the world will be worse in other ways it's really ok in the end. You can interpret this in a way that the laws will turn out Humean, but it sounds to me as if laws aren't metaphysically derived. If you read Malebranche more carefully on laws, you see that he does anticipate Hume in one way, but the difference seems to me to be that he declares that God at the outset establishes the laws, which then do the work of organizing how things will develop. This is sometimes misread as God moving all the little particles around (occasionalism), but Jolley argues that Malebranche really thinks of the laws as doing the work. How God controls every atom is by setting up laws to order creation. That seems to me to be against the spirit of seeing laws as metaphysically posterior to what happens, as Hume wants, and it seems to me that Malebranche has a better sense of what laws are supposed to be, but I don't have a good argument for that.

I thought this discussion was going to end up with a chat on testimony. But laws are puzzling things--if even it is right to call them things. Without running back to the Treatise to check, I would imagine the Humean view on laws to be something akin to the positivist view on laws--at least insofar as they are merely statements of regularity (A statements). It is very easy to confuse laws (in the fashion i just described) and the early modern conception of natural law. But you've got to be right about Hume Jeremy, that if something breaks a law, then we just had the law wrong. But isn't the attitude you express quite opposite than "mere statements of regularity"?

I think that Hume is basically right about miracles, and his Bayesian argument about comparing probabilities also works for improbable events that are not miracles (like Bill Bennett's claim that he'd "come out pretty close to even" gambling on slot machines).

One important point is that, as time has gone on, we've been able to investigate these sorts of phenomena in more rigorous ways than talking to peasants, and there have been many particular cases where we've been able to verify that the witness was mistaken, a few (like meteors) where we've found that the phenomenon is real but non-miraculous, and basically none where we've found the phenomenon to be real and miraculous (some investigations may be biased against this finding, but not all are - consider ESP studies). What's more, we are learning more about how people can come to falsely believe that they have witnessed a miracle. There is also a broad class of non-miraculous cases where testimony is considered to have little evidential value (e.g. at Snopes), and multiple sources of testimony may do little to increase the reliability because they are not independent.

I'm seconding blar:

First, Hume doesn't think of laws of nature as forces controlling nature, so powerful only a god could oppose them. Instead, he just says that, in order for us to treat something as a law of nature, it has to be supported by the strongest evidence that experience could provide. Otherwise, we'd call it a mere trend or tendency. I take it Hume's view of a law of nature is something like the limiting case of an observational trend--a consistent pattern of evidence.

In any case, you can drop the laws of nature right out of the argument. You can just say that, for us to treat something as a miracle, it has to run contrary to the strongest evidence that experience could provide. Otherwise, we shouldn't treat it as a bona fide miracle, but merely an outlier to a trend. This gets Hume the intrinsic improbability he wants, and hence it gets him the maxim he wants: for a miracle report to establish a religion, the testimony itself has to be so credible as to outweigh the intrinsic improbability of the event attested to.

Like blar claims, the argument also works with merely improbable events (it's just not as decisive). Outliers to trends (e.g., Bill Bennett's success) are rather improbable, precisely because they are outliers to trends. And this means we must require the testimony to be very credible before we accept it. A fortiori, a miracle (taken as something like the limiting case of a trend-outlier) must be supported by the most credible kind of testimony there is--otherwise we shouldn't accept it. This is the point of Hume's "eight days of darkness" example--it would take the best kind of testimony to outweigh the improbability of the reported event.

And don't forget the second part of Hume's essay, where he gives us reasons to lower our confidence in miraculous testimony--reasons based not on the nature of the reported event, but on the nature of the testimony itself. Passions of wonder, self-deception, piety, and a willingness to check one's judgment at the door, combined with the negative correlation of miracle reports with education; all this should raise our confidence that such testimony is not to be taken too seriously.

I have a couple comments regarding how we are reading Hume's argument. First assuming (with Hume) that testimony is a case of induction, if Hume's argument against induction is a good argument, then no testimony as such could ever justify a belief in miracles. There is nothing special about testimony which makes it an irrational foundation for belief in miracles. Belief in miracles is irrational simply because it is a case of induction. But if this is Hume's point, he doesn't need any of the essay beyond the first two pages. He just has to argue that testimony is a case of induction and then reference his argument in the beginning section of the EHU. So, I tend to think that something else is going on in Hume's argument. He begins by thinking he has established that induction is irrational, then he must explain how people come to have the beliefs that they do. If you read all of his thoughts on probability both in the Treatise and the EHU, you will see that probability (as well as proof) for Hume is a matter of how our experience produces beliefs of differing varieties of assurance. That is, when we are talking about beliefs, there really is no such thing as proportioning our beliefs to evidence (I am assuming that Hume does not believe we have any control over our beliefs). Thus, Hume's whole discussion in the second part of the essay on Miracles, I would argue, is simply a matter of explaining how people who actually believe that miracles occur come to have the beliefs they do. But this is an explanatory account and assumed that no induction is ever justified. And I take it that this is Chesterton's point in short. Those who deny the viability of testimonial justification for miracles do so because of a doctrine (Hume's doctrine is that induction is unjustified).

Hume doesn't think that induction is unjustified. Not at all. He just locates its psychological source, not in reason/the understanding, but in the imagination (i.e., his associationist psychology). He's giving us a psychological story for how we make inductive inferences; he's not recommending that we give up inductive inferences.

Hume's chapter on miracles does take for granted that "experience be our only guide in reasoning concerning matters of fact". But I think that's the only relation to the earlier parts in the First Enquiry, and to Hume's general psychology of belief. (Though there is a connection to the probabilities and proofs discussion found in the Treatise).

Cole,

I think that it is a stretch to say that "Hume doesn't think that induction isn't justified" in one breath and then to say that "Hume locates the justification in psychology _rather than_ reason" in another breath. I would argue that unless we are talking about a basis in reason we are not talking about justification. Of course we all make inductive inferences (and if Hume is right this is exactly what we are doing when we believe on the basis of any testimony) but the worry is whether or not these inferences have some basis in rationality. Simply to assert that we make them does not mean they are rational.
I'll grant you that Hume gives us a psychological story to explain the mechanisms of belief in unobserved matters of fact, but this comes after Hume thinks that he has undermined an argument for why these inferences are rational. Now, my point in the previous post was this: given that Hume thinks that he has shown how induction is not rational and that he thinks _all_ belief based upon testimony is inductive in nature, then _all_ beliefs based upon testimony (including testimonies to every day occurrences) as such are irrational. Like you, I agree that Hume is providing an explanatory account of beliefs. But if this is the case then he is doing it only because he thinks that a justificatory account (one showing the rationality of beliefs based on testimony) is baseless. Thus, Hume accepts a doctrine--that induction is not rationally defensible--and thus must explain away our inferences. Isn't this Chesterton's point?

Cole,

(I apologize if this is a repeat--I posted just a minute ago, but it doesn't seem to be appearing here). I think it is tricky to say "Hume doesn't think that induction is unjustified" and then, in the same breath say "He just locates its psychological source, not in reason/the understanding, but in the imagination (i.e., his associationist psychology)." If we are not talking about rationality when we are talking about justification, I don't know what we are talking about. Secondly, given that Hume thinks he offers his explanatory account only after providing an argument showing that the justificatory account is implausible, it seems that you are using "justification" in a stretched sense.
I'll grant you that Hume offers a solution to the so called problem of induction, but this solution is based upon argument. Given this argument and that Hume identifies beliefs based upon testimony as an example of inductive inferences, we should conclude that Hume thinks that _all beliefs_ based upon testimony (including very ordinary beliefs and miracles) are irrational even though we make them. So, what is the point of the essay "Of Miracles" if it is not supposed to show that belief in miracles on the basis of testimony is irrational?

What defines the miraculous?

Here is one ad hoc definition: a miracle is an event that violates or runs counter to a rule(used in an informal sense) that precludes such an event , such that instead of the event that should have happened according to the rule, a different event occurs.

Given this, what is miracuous and what is not would seem to depend to some degree on how flexible you are about rules.
I remember reading a newspaper story about a baby swept up by a midwestern tornado and deposited unharmed some 100 yards away.
if you believe a rule holds sway which makes it impossible that the baby should not have been dashed to bits then you may choose to call this a miracle.
Alternately, if you hold the baby's survival possible but greatly against the odds --- you may call it a one in a million occurance rather than a miracle.-- unless you believe that a one in a million occurance violates a rule that precludes it--and so is a miracle in itself.

Optionally, you may simply consider the baby's survival a rare manifestation of the usual rule instead the result of the rule's suspension or violation You might say, per example, " I had no idea that that rule included such unusual cases and I will adjust the scope of the rule accordingly."
On the other hand , you might take a different view: " i must have been thinking of the wrong rule-- I should have been thinking of the rule that allows for the baby's survival."
or you might say " this situation must have been different than I thought -- it is a situation in which a rule applies which allows for the baby's survival."
Or even " a force or rule not normally operative was apparently responsible for this anomolous occurance."
Or perhaps even, "this occurance indicates an entirely new rule not only heretofore unknown but heretofore non-existent."
What is suggested by the above is that if you are too flexible about the rules you don't get a miracle out of a baby surviving a tornado -- nor do you get an explanation akin to a contemporary scientific account.
if a follower of Ockam-(as is science) -you will not allow that the situation is different than you thought nor
that there are any more rules than you already have and you will try to fit the occurance into one of those rules by tweaking( i am reminded here of Quine's famous paper and also reminded that in attempting to accomodate new phenomena such tweaking can take on baroque and un-Ockamish proportions )
It seems that the believer in miracles is also a follower of Ockam in that he will also not allow that an unusual occurance signifies that a new rule has popped up but, unlike science, he will not allow that a new or unusual phenomenon must necessarily be accomodated by an old rule. Instead, he believes that, in the miraculous case, the old rule which otherwise holds-- is contravened by some agent or other which acts entirely outside the rules---and this makes the miracle
Where the scientist tweaks the rules
the believer in miracles abandons them for the actions of an agent.
But one may ask, is the agent not simply following a different set of rules coexistent with
the usual set? is the agent then not ultimately a part of the rules governed world?
Surely the believer in miracles must deny this possibility or his agent becomes just another theoretical entity of physical science-- albeit a strange one, and a miracle just another physical phenomenon. The believer must believe then that from the agent the rules arise --( it strikes me that this should make rules themselves rather miraculous)- rather than that the rules manifest the agent.
Here then is a difference between science and faith. For the believer in miracles--usually the religious believer ---
the whole scheme of rules governing the world arises and is sustained by the will
of an agent which at any moment may will
something else to happen which is counter to the rules normally obtaining-- a miracle (without the rules to contrast it, there is no miracle).
By contrast, for the scientist a scheme of rules makes the world and is never contravened-- however, science admits we just don't know all the rules --yet.
To the scientific believer no amount of evidence can overthrow the rules based scientific point of view, just as no evidence will remove the willful agent from the miracle believer's point of view.
To one, the Christ in the tortilla will never be anything but the rules in operaton; for the other, never anything but the agent's will.

Thus fundamentalist creationsim is an account of a will willing as it will. And scientific evolution is permutations of the rules.
Scientific belief and belief in miracles are based, in effect, upon different and irreconcilable fundamental articles of faith and it is no wonder that evidence to one party is hogwash to the other.
Hume's argument will most affect, and was perhaps originally aimed at, those straddling the fence between the two viewpoints. Hume attempts to convert the straddlers to the more scientific point of view.

Interesting post vkaps.
"Laws of nature" based upon the scientific model, are limited to what science is capable of knowing about nature. It's conceivable that nature can be acted upon and made to behave in a manner which isn't "natural" (i.e. rule based and predictable) by means of some type of ritual inducement. For example, there are well documented stories of rain makers and dowsers who discover/generate water by means of requisite hokus pokus.