God, Probability, and Noncontingent Propositions

Matt Hart comments:

. . . most of what we conceive is possible. So if we say that

1) In 80% of the cases, if 'Conceivably, p' then 'Possibly, p'
2) Conceivably, God exists
Ergo,
3) Pr(Possibly, God exists) = 80%
4) If 'Possibly, God exists' then 'necessarily, God exists'
Ergo,
5) Pr(Necessarily, God exists) = 80%,

we seem to get by.

I had made the point that conceivability does not entail possibility.  Hart agrees with that, but seems to think that conceivability is nondemonstrative evidence of possibility.  Accordingly, our ability to conceive (without contradiction) that p gives us good reason to believe that p is possible.

What is puzzling to me is how a noncontingent proposition can be assigned a probability less than 1.  A noncontingent proposition is one that is either necessary or impossible.  Now all of the following are noncontingent: 

God exists
Necessarily, God exists
Possibly, God exists
God does not exist
Necessarily, God does not exist
Possibly, God does not exist.

I am making the Anselmian assumption that God (the ens perfectissimum, that than which no greater can be conceived, etc.) is a noncontingent being.  I am also assuming that our modal logic is S5.  The characteristic S5 axiom states that Poss p --> Nec Poss p.  S5 includes S4, the characteristic axiom of which is Nec p --> Nec Nec p.  What these axioms say, taken together, is that what's possible and necessary does not vary from possible world to possible world. 

Now Possibly, God exists, if true, is necessarily true, and if false, necessarily false.  (By the characteristic S5 axiom.)  So what could it mean that the probability of Possibly, God exists is .8?  I would have thought that the probability is either 1 or 0.  the same goes for Necessarily, God exists. How  can this proposition have a probability of .8?  Must it not be either 1 or 0?

Now I am a fair and balanced guy, as everyone knows.  So I will deploy the same reasoning against the atheist who cites the evils of our world as nondemonstrative evidence of the nonexistence of God.  I don't know what it means to say that it is unlikely that God exists given the kinds and quantities of evil in our world.  Either God exists necessarily or he is impossible (necessarily nonexistent).  How can you raise the probability  of a necessary truth?  Suppose some hitherto unknown genocide comes to light, thereby adding to the catalog of known evils.  Would that strengthen the case against the existence of God?  How could it?

To see my point consider the noncontingent propositions of mathematics.  They are all of them necessarily true if true.  So *7 + 5 = 12* is necessarily true and *7 + 5 = 11* is necessarily false.  Empirical evidence is irrelevant here.  I cannot raise the probability of the first proposition by adding 7 knives and 5 forks to come up with 12 utensils.  I do not come to know the truth of the first proposition by induction from empirical cases of adding.  It would also be folly to attempt to disconfirm the second proposition by empirical means.

If I can't know that 7 + 5 = 12 by induction from empirical cases, how can I know that possibly, God exists by induction from empirical cases of conceiving?  The problem concerns not only induction, but how one can know by induction a necessary proposition.  Similarly, how can I know that God does not exist by induction from empirical cases of evil?

Of course, *God exists* is not a mathematical proposition.  But it is a noncontingent proposition, which is all I need for my argument.

Finally, consider this.  I can conceive the existence of God but I can also conceive the nonexistence of God.  So plug 'God does not exist' into Matt's argument above.  The result is that probability of the necessary nonexistence of God is .8!

My conclusion:  (a) Conceivability does not entail possibility; (b) in the case of noncontingent propositions, conceivability does not count as nondemonstrative evidence of possibility.

Comments

[John]

I am certain that the following suggestion will prove mind-blowingly naive. Nevertheless...

You characterize the noncontingent propositions of mathematics as "necessarily true if true". Thus, for any noncontingent proposition of math, if that proposition is true, then it is necessarily true.

Likewise, regarding the noncontingent proposition that God exists, it is necessarily true if true. Thus, for the noncontingent proposition that God exists, if that proposition is true, then it is necessarily true.

Thus, we can imagine a little argument of the following sort:

(1) If the noncontingent proposition 'God exists' is true, it is necessarily true.
(2) The noncontingent proposition 'God exists' is true.
(3) Therefore, it is necessarily true.

Here is where empirical considerations enter in: the assessment of (2). Presumably, when atheists take evil to be evidence against the existence of God, they do not question the noncontingent status of the proposition 'God exists'. What they question is its truth.

Of course, atheists of this sort should therefore regard God's existence as impossible, since the proposition that God exists is noncontingent. But in fact I think many atheists, and certainly atheists of this sort, DO regard God's existence as impossible. That explains why the standard response on their part to Plantinga's modal ontological argument is to deny the premise that it is possible that God exists.

[Matt Hart]

Bill,

I see the puzzle, but I'm pretty sure it trades on some confusion or other. For one thing, if what you say here is true then most of metaphysics goes out the window.

A: 'Lewisian modal realism is absurd!'
B: 'Well, sometimes absurd things happen. So I guess you want me to infer that it is unlikely?'
A: 'I suppose so.'
B: 'But LR, if true, is nec. true. Probabilistic considerations are irrelevant and your increduluous stare counts for nothing.'

I think that gives us good reason to think that something is wrong with your argument, although it doesn't tell us what.

One suggestion is this: forget probabilities, let's talk about credences instead. Since credences are degrees of rational belief (or something like that, I'm not too sure), they can take values between 0 and 1 even when the the putatively necessarily existent is in view.

But in fact I think I can show where you err.

Probability in the way it functions in these debates is conditional on our shared background knowledge or beliefs (K), but it would be illegitimate to place either the possibility of God or the impossibility of God in K since that is the question at issue. But what is in K is this:

1) Either God is necessary or God is impossible.

And you cannot move from (1) to Pr(G|K) = 1 or to Pr(G|K) = 0. Significantly, and this is where I think you go wrong, it follows from this that we have no reason to grant the following proposition, which you require:

2) Either (Pr(G|K) = 1) or (Pr(G|K) = 0).

For it is only if you have (2) that you can dismiss probablity raising arguments, since you know that the probability must be either 1 or 0. (Though even that move is questionable, I think.) We have already ruled that both disjuncts cannot be straightforwardly derived from K - so on what grounds do you make this claim?

You say, either God's existence has a probability of 1 or 0. But upon what set of propositions are these probabilities conditional upon? You face a dilemma: if this set includes 'God is necessary' or 'God is impossible' then it won't get you (2) because neither of those sets is K. But if it merely includes 'God is either necessary or impossible' then you have no way (that I can see) of deriving (2).

Finally, as to your parting consideration. The conceivability of the non-existence of an x isn't of as much force as conceiving of the existence of an x, for you can never be sure that you aren't just conceiving of a patch of empty space, leaving it open that an x is still somewhere around (a fortiori with non-spatial entities such as God).

[Joseph A.]

This is an amateur and possibly silly question, so pardon in advance for asking it: Is it really possible to conceive of a universe where theism is not true?

I ask this because it seems to me that anything I conceive is exactly that: A product of my mind, sustained entirely by the actions of my mind. And anything which depends upon the action (particularly the continual, at-that-moment action) of a mind for its very existence is a world where some form of theism is true. So to 'conceive of the non-existence of God' would require me conceiving of a world that itself was not the result of a being's conceiving.

It seems similar to the problem of 'conceiving of something coming from nothing without cause.' Can someone really do that? Or does a person conceive of something that 'comes into existence' due to their mind, and they slap a "This was uncaused!" label on it?

[Bill Vallicella]

Joseph asks, "Is it really possible to conceive of a universe where theism is not true?" Yes. Surely one can conceive of the nonexistence of God. Atheists do it all the time. And theists, who understand what atheists are claiming, do it as well.

The rest of what you say I simply don't understand.

[Bill Vallicella]

Matt,

The underlying issue is what your talk of probability (in the argument of yours that I quoted) amounts to. I note that you shift your ground in your comment. >>forget probabilities, let's talk about credences instead. Since credences are degrees of rational belief (or something like that, I'm not too sure), they can take values between 0 and 1 even when the the putatively necessarily existent is in view.<<

What are you measuring? Degrees of conviction?

[Bill Vallicella]

John,

Are you the guy who mad ethose very good comments about mereology some months ago? Welcome back.

Yes indeed, the atheist questions (2). But if *God exists* is noncontingent, then how can any empirical evidence be relevant to its evaluation? That's what I don't understand.

The answer to my puzzle must be along the lines of Matt's comment. What we are talking about here is epistemic probability -- whatever that is.

[Account Deleted]

Bill asks: "how can I know that God does not exist by induction from empirical cases of evil?"

The problem Bill raises is a general one regarding all non-contingent propositions (including mathematical ones).

A partial Answer: Non-contingent propositions have logical consequences that are themselves contingent. These later can be directly falsified or falsified to a certain degree of probability.

E.g., the proposition that 5+7=12 is necessary. Yet it together with a few other contingent premises logically entail that there are twelve utensils on my table. However, the proposition that there are twelve utensils on my table is not necessary; there could have been eleven, fifteen, or none, etc.

Now, if something goes wrong and the conclusion that there are twelve utensils on my table turns out to be false, then we must conclude that one or more of the premises is false. Of course, given our strong epistemic credentials regarding these simple mathematical truths, we would conclude that one of the contingent premises is most likely false.

Obviously, no confirmation of the consequences entails logically that the mathematical proposition is *true*. Hence a mathematical proof is needed.

Same goes regarding God's existence and evil. God's existence, together with certain premises regarding the nature of God, entails that there are no cases of gratuitous evil in the world. This later proposition is contingent. Thus, even one case of gratuitous evil refutes one or more of the premises.

Now, since

(i) In this case all the premises are necessary (God is omnipotent, omniscient, morally perfect, etc., are all necessary if true); and

(ii) Our epistemic credentials regarding the existence of a divine being featuring all of these perfections are pretty weak;

it follows that the truth of a contingent proposition can falsify the contingent consequences of a set of necessary truths and, thereby, logically entail the falsity of at least one of the necessary truths themselves.

[Matt Hart]

Skip credences. That was one suggestion, but I don't think it is necessary. Just start at 'Probability in the way it functions...'

Here's another way of making my point which might seem clearer. God only has a probability of 1 or 0 if you are taking the probability of a thing to be a function of its occurrence across all possible worlds. But the problem is we don't know which possible worlds are the real ones (the theist thinks God exists at all of them, and the atheist says he exists at none). So we estimate that prob. that God exists according to contents that both atheists and theists think possible worlds plausibly have. It is upon these incomplete descriptions of worlds that the probability of the existence of God can have values between 1 or 0.

[Bill Vallicella]

Peter writes, >>God's existence, together with certain premises regarding the nature of God, entails that there are no cases of gratuitous evil in the world. This latter proposition is contingent. Thus, even one case of gratuitous evil refutes one or more of the premises.<<

I agree that just one case of gratuitous evil refutes the proposition that God exists. But how do you know that a given case of evil is a case of GRATUITOUS evil? Is gratuitousness an empirically detectable feature?

To avoid begging the question you need an independent way of establishing that your favorite example of evil is gratuitous, a way independent of just assuming that there is no God.

If God exits,then there are no gratuitous evils. Contrapositively, if there are some gratuitous evils, then God does not exist. Now how willyou establish that there are some gratuitous evils?

If x is apparently F it doesn't follow that x is F. So you can't appeal to the appearance of gratuitousness. Such an appeal would be like an appeal to ignorance: we don't know of any justifications for certain evils, so there aren't any.

[TaiChi]

"Yes indeed, the atheist questions (2). But if *God exists* is noncontingent, then how can any empirical evidence be relevant to its evaluation? That's what I don't understand."

Equally, if Goldbach's conjecture is necessarily true or false, how can any empirical evidence be relevant to its evaluation? But empirical evidence is relevant to its evaluation: if the newspaper headlines tomorrow say "GOLDBACH'S CONJECTURE PROVEN", then I should take this as empirical evidence of the truth of Goldbach's conjecture. Not indefeasible evidence, mind you - newspapers and even mathematicians can be wrong about these things - but as making Goldbach's conjecture very probable indeed.

[Account Deleted]

TaiChi,

"if the newspaper headlines tomorrow say "GOLDBACH'S CONJECTURE PROVEN", then I should take this as empirical evidence of the truth of Goldbach's conjecture."

Newspaper reports have no bearing on the truth/falsity of GC and they certainly have nothing to do with proving the conjecture. They merely disseminate information about whether such a proof has been found. So while relying on a newspaper report may provide you with an empirical means of knowing that a proof was found, in itself it is not the sort of evidence that could be relevant to the truth/falsity of GC.

[TaiChi]

A couple of relevant links, though I don't think they settle the issue:

http://www.askphilosophers.org/?q=&cat=Probability&panelist=Questions

http://www.infidels.org/library/modern/paul_draper/serious.html (Search for "I defended my premise" to see Draper's response to Plantinga, who makes a similar move to the one mooted by Bill here).

[Account Deleted]

Bill,

(A) "To avoid begging the question you need an independent way of establishing that your favorite example of evil is gratuitous, a way independent of just assuming that there is no God."

But I have given such an argument above. I have said that

(i) The strongest defeater against the move from apparent to genuine gratuitousness is the existence of God. But since we cannot assume that God exists, this defeater cannot be introduced to block the move;

(ii) Radical skepticism about our moral judgments is unreasonable and moreover neither side is willing to adopt it. Moreover, it seems to me that all sides would agree that at least sometimes we are right about our moral judgments,

Therefore,

(iii) There is no systematic reason that rules out that at least sometimes we are correct when we judge that some evil that appears to be gratuitous is so.

(B) You further ask: "Is gratuitousness an empirically detectable feature?"

First, I have claimed that whether gratuitous evil exists or not is a contingent matter. I think we agree about this. Second, I take it that we agree also that a case of gratuitous evil is one whereby the evil in question is not necessary to bring about a higher good nor is it necessary to prevent a worst evil.

Assuming these two things, the question is whether we possess the resources to identify a case of gratuitous evil, if such exists. Surely we possess the resources to identify evil (i.e., suffering or death, etc.,). We also have the resources to identify the moral properties of good/bad; o/w we could not make moral judgments at all. And, finally, we should be able to understand what it is for some evil to be (causally or logically) necessary to bring about a future good or prevent a worst evil. We make such judgments all the time. It follows that there are no conceptual barriers that hinder us from identifying a case of gratuitous evil, if such exists.

(A) and (B) together entail that there is a high likelihood that some evil that appears to us as gratuitous is indeed so. I have also pointed out that the conceivability/possibility case is not the same as the gratuitous evil case because in the former the defeater cannot be ruled out even if we rule out radical skepticism.

[TaiChi]

Peter,

If A->B, then evidence for A is also evidence for B. That there is a proof of GC entails that GC is true. So evidence for there being a proof of GC is also evidence that GC is true. Hence the newspaper report is evidence that GC is true. I can't see what's wrong with that.

Of course, it's true that the newspaper report is not relevant to the truth/falsity of GC in the sense of making it true. But I never suggested it did. I only claimed that this would be evidence of GC, and, as I hope you'll agree, what counts as evidence for a proposition often ranges beyond what would make that proposition true.

[Account Deleted]

TaiChi,

"If A->B, then evidence for A is also evidence for B. That there is a proof of GC entails that GC is true. So evidence for there being a proof of GC is also evidence that GC is true."

While the above may or may not be true, your newspaper report example does not have the above form. Instead, your newspaper/GC example requires that evidence is transitive. So your example requires one or another of the following two principles, where 'EV(Q,S)' is short for the proposition that "Q is evidence for S":

(P)EV(R,(EV(Q,S)) --> EV(R(S):

or the transitive version would go like this:

(P*) If EV((R,Q) & EV(Q,S))-->EV(R,S).

Both (P) and (P*) are false. In other words, evidence is not transitive. You basically assume that evidence is transitive.

[T. Hanson]

TaiChi, you said,

"Hence the newspaper report is evidence that GC is true. I can't see what's wrong with that."

Say the newspaper reports scientists have concluded there is very likely no life on Mars. Should the scientists now clip out the newspaper article and include it in their research as more evidence their theory is true?

[Account Deleted]

TH,

Excellent counterexample to TaiChi's thesis.

[TaiChi]

"Say the newspaper reports scientists have concluded there is very likely no life on Mars. Should the scientists now clip out the newspaper article and include it in their research as more evidence their theory is true?"

No, because the newspaper reports are not independent of their research - they are double counting. Am I double counting when I take the newspaper headline "GOLDBACH'S CONJECTURE PROVEN" as evidence of the truth of GC? No, since I am not acquainted with the proofs on which the headline is based.
So perhaps we should say: what counts as evidence depends upon what one's epistemic situation is. But then: empirical evidence, such as testimony, can be relevant to one's evaluation of a proposition's truth.

"You basically assume that evidence is transitive."

I assume evidence can be transitive. I don't think that's controversial, as testimony is always considered a form of evidence, and is almost always transitive.

[Bill Vallicella]

Pter,

Thanks for the detailed and patient comments. You may be right, and I may just be confused. But I'm still not clear on this. Your view, I take it, is that the likelihood that some evils are gratuitous (and that therefore God does not exist) is greater than the likelihood that God exists (and therefore that there are no gratuitous evils).

One question is: How do you evaluate these likelihoods? How do you know that they are not equally likely? Consider the 20 or so arguments that have been given for the existence of God. They constitute a cumulative case for the existence of God. How do you evaluate this body of evidence relative to the counterevidence?

After all, you are committed to saying that the evidence against theism is weightier than the evidence for it. But to be justified in claiming this, you need some method of weighing this evidence, and what would that be?

[Account Deleted]

TaiChi,

I am not sure what you mean by "evidence *can* be transitive"? Either the relation '...is evidence for ___' is transitive or it is not. If some instances of a relation are transitive and others are not, then the said relation is not a transitive relation.

If the evidential relation is not transitive, then inferences of the form:

(EV(R,Q) & EV(Q,S))-->EV(R,S).

are invalid. Therefore, one cannot infer from the fact that a newspaper report provides evidence that a proof for GC was found and that a proof of GC provides evidence for the truth of GC, the conclusion that the newspaper report provides evidence that GC is true. Yet you need this sort of inference for the case you are making.

"No, because the newspaper reports are not independent of their research - they are double counting."

What do you mean by "independent of their research"? Do you mean that the newspaper report is not an independent proof of GC, but rather a report of such a proof? But that is the point. Newspaper/TV/Radio reports cannot be independent of what they report, o/w they would not be reports. And precisely for this reason they are not evidence for the truth of GC, but at most evidence that a proof was found for GC.

Testimony is considered evidence in courts only when it is a report of a direct experience; i.e., of what you saw or heard. Courts do not allow reports of what you heard someone else say about what they saw or heard.

[Bill Vallicella]

Peter argues,

(i) The strongest defeater against the move from apparent to genuine gratuitousness is the existence of God. But since we cannot assume that God exists, this defeater cannot be introduced to block the move;

(ii) Radical skepticism about our moral judgments is unreasonable and moreover neither side is willing to adopt it. Moreover, it seems to me that all sides would agree that at least sometimes we are right about our moral judgments,

Therefore,

(iii) There is no systematic reason that rules out that at least sometimes we are correct when we judge that some evil that appears to be gratuitous is so.

To block a move, in plain English, is to invalidate an inference. But surely the inference from 'x is apparently gratuitous' to 'x is gratuitous' is invalid and shown to be invalid by the mere possibility of God's existence. Right? What you want to say is that apparent gratuitousness is defeasible, nondemonstrative evidence of genuine gratuitousness.

For now we assume that skepticism in ethics is unwarranted. So let's assume we know that some acts or events are evil and some good. Your argument confuses this with the gratuiousness question. If I know that E is evil, it doesn't follow that I know that E is gratuitously evil. Your argument conflates these two issues. Your talk of moral skepticism is a red herring. Since your second premise is confused, that fact invalidates your argument as it stands. But repairs are always possible!

Granted, I may be correct in judging of a certain evil that it is gratuitous. But I may also be correct in judging that it is not. Further, I may be correct in judging that God is possible, hence actual, and that therefore there cannot be any gratuitous evils.

I suspect that considerations of epistemic probability are just too murky to make it rational to say things like 'The existence of God is highly unlikely given the evils in our world' or 'It is more likely that there are gratuitous evils than that God exists.'

[T. Hanson]

Hi Bill,

I don't want to distract from the main issue, but I have heard you say this before:

"Consider the 20 or so arguments that have been given for the existence of God. They constitute a cumulative case for the existence of God. How do you evaluate this body of evidence relative to the counterevidence?"

Since there are reasonable doubts and objections to all the arguments, and none can be considered conclusively sound or strong, how do they accumulate to constitute any evidence? Is there a post where you make this argument or can you explain your reasoning briefly? Why doesn't the sum of 20 zeros add up to zero? And does Peter grant you this premise?

[T. Hanson]

Correction: I meant "conclusively sound or cogent," above.

[Account Deleted]

Bill,

(I) Let me first make clear that my primary target is the unknown-purpose-defense (UPD) and not whether God exists. So the following three propositions are inconsistent:

(A) God exists;
(B) Gratuitous evil exists;
(C) UPD is true.

My ultimate goal is to show that (A) and (B) are consistent; therefore, (C) is false.

(II)) “If I know that E is evil, it doesn't follow that I know that E is gratuitously evil. Your argument conflates these two issues. Your talk of moral skepticism is a red herring.”

(a) I grant the following two propositions:

(i) It does not *logically follow* that E is gratuitous evil from knowledge that E is evil;
(ii) It does not *logically follow* that E is gratuitous evil from the mere appearance that E is a case of gratuitous evil.

(b) But the question in front of us is not a matter of logical inference. The question is whether we are justified to believe that there is at least *one* case of evil E such that:

(b1) E appears to us to be a case of gratuitous evil;
(b2) Based on the fact that E appears to be a case of gratuitous evil we judge E to be a case of gratuitous evil;
(b3) We are correct in judging E to be a case of gratuitous evil.

(c) What could falsify (b3)?

Well, one consideration that could falsify (b3) is the conjunction of (A) and (C). For if God exists and UPD is true, then it follows that God has a purpose for every evil and therefore is justified to permit it. Hence, there cannot be even one case of gratuitous evil. Hence, (b3) must be false. So the conjunction of (A) and (C) is a potential defeater of (b3).

Another defeater of (b3) is moral skepticism. Suppose that we are wrong about every moral judgment we make. If so, then (b1) and (b2) may be true, while (b3) is false. Hence, the issue of moral skepticism is not a “red herring”. It could be one of the defeaters of (b3), given (b1) and (b2).

(d) I cannot think of any other defeater for (b3), given (b1) and (b2). Now, since none of the disputants believe that moral skepticism is true, this defeater is out. Furthermore, the conjunction of (A) &(C) entails that God exists (i.e., (A)). But we are not entitled to presuppose this fact. Moreover, as I have argued, our epistemic credentials regarding this matter are pretty weak; certainly much weaker than our epistemic credentials regarding moral judgments. Hence, the only other defeater of (b3), given (b1) and (b2), is out.

(e) In the absence of any other defeater, what possible ground could there be to insist that (b3) is false, given (b1) and (b2)? Insisting that under these conditions we are nonetheless wrong about every case such as (b1)-(b3) would be analogous to the following. Assuming that radical skepticism about our senses is out, one nevertheless insists that we are wrong about *every* one of our perceptual judgments; i.e., that in *every* case that we perceive that so-and-so is in front of us, we are wrong and there is no so-and-so in front of us. Surely no one would maintain this view, assuming they do not believe radical skepticism. Similarly, since the only primary defeaters for (b3), given (b1) and (b2), are ruled out, we are entitled to conclude that at least sometime we are correct in judging that some evil is gratuitous based on what appears to us as gratuitous evil.

(III) Does my argument entail that (A) is false; i.e., that God does not exists? It does not, unless one links God’s existence with UPD (i.e., the conjunction of (A) and (C)). So my argument shows only that UPD is false, which is what I intended to do at the outset. The following question arises: How can God and gratuitous evil co-exist? Or alternatively: Are (A) and (B) consistent? I maintain that they are. Showing this, of course, is an altogether different story.

[Bill Vallicella]

Hi Tony,

Excellent and pertinent questions! On the one hand, 20 inconclusive arguments do not add up to one conclusive argument. But if it makes sense to raise the probability of theism -- an issue on which I am not clear -- then it would seem that a cumulative case will be stronger than a case resting on one argument.

Suppose I give 20 arguments for a thesis, all valid, but I don't know whether any of them are sound. It seems more likely that one of them will be sound if I give 20 as opposed to only one.

Why do philosophers, when they write their papers, often give more than one argument for their thesis? If one argument is probative, why give more?

Why did Aquinas give Five Ways and not just one? You can imagine his amanuensis saying to him, "Look Tom, if you have a knock-down argument, just give it, and shut up. My hand is getting tired."

This is a huge separate topic obviously. I don't have a worked-out view about it. But I do wonder about the divide-and-conquer approach of many atheists when it comes to the theistic arguments.

[Joseph A.]

Peter,

How does a given evil "appear to us to be a case of gratuitous evil" as you claim in b1?

Is any evil I hear about and for which I don't have a ready explanation of how said evil could be instrumental-as-is in securing a good, an evil which "appears to me to be a case of gratuitous evil"?

[Account Deleted]

Joseph,

A case of suffering or death, etc., that appears to us to be without any valuable purpose (either insofar as it brings about some greater good or prevents a worst evil). There are many such cases of apparent gratuitousness. The question is whether they are genuine or merely apparent.

[Joseph A.]

Peter,

Is there a distinction between my not knowing what such a valuable purpose of the evil could be, and of it "appearing to us be without any valuable purpose"?

To give what I hope is a comparable example, let's say I come across an equation. I find myself unable to solve the equation. Does my inability to solve it mean that equation appears to me to be unsolvable?

[TaiChi]

"I am not sure what you mean by "evidence *can* be transitive"? "

I mean that sometimes when A->B, we can take evidence for A as being evidence for B.

"Yet you need this sort of inference for the case you are making."

You appear to have confused my offering of a counterexample to the suggestion that 'empirical evidence cannot be relevant to the evaluation of a non-contingent proposition's truth' as the assertion of some general rule of inference. It is not. I'm not making any claims to how and why the example works, interesting though that may be, and I'm not interested in endorsing any of your proposals for what principle of inference may be inferred from the case. The fact that the counterexample is genuine is all I require.

"What do you mean by "independent of their research"? Do you mean that the newspaper report is not an independent proof of GC, but rather a report of such a proof"

Yes.

"But that is the point. Newspaper/TV/Radio reports cannot be independent of what they report, o/w they would not be reports. And precisely for this reason they are not evidence for the truth of GC, but at most evidence that a proof was found for GC."

I don't think that last sentence follows: you must be using an entirely different concept of evidence than I do. I suggest that evidence is what provides ground for a belief or theory. My reading of the newspaper headline provides a ground for my belief that 'GC is true'. So the newspaper headline is evidence for my belief that 'GC is true'.

"Testimony is considered evidence in courts only when it is a report of a direct experience; i.e., of what you saw or heard. Courts do not allow reports of what you heard someone else say about what they saw or heard. "

What courts do and do not allow, what counts in legalese as 'evidence', isn't much to the point if we're arguing over what counts as evidence in the common sense of the term. The legal profession has its own motives for regimenting language, and do not merely provide a conceptual analysis of terms as they are widely employed.

[TaiChi]

"For now we assume that skepticism in ethics is unwarranted. So let's assume we know that some acts or events are evil and some good. Your argument confuses this with the gratuiousness question. If I know that E is evil, it doesn't follow that I know that E is gratuitously evil. Your argument conflates these two issues. Your talk of moral skepticism is a red herring. Since your second premise is confused, that fact invalidates your argument as it stands. But repairs are always possible!" ~ Bill

"Another defeater of (b3) is moral skepticism. Suppose that we are wrong about every moral judgment we make. If so, then (b1) and (b2) may be true, while (b3) is false. Hence, the issue of moral skepticism is not a “red herring”." ~ Peter

This is interesting. I think Bill's point is that it does not matter to our judgments of good or evil whether or not their instances are also gratuitous, but if moral skepticism be only the denial of that our moral judgments of good and evil are correct, then skepticism about gratuity does not imply moral skepticism.
It's a good point, but I think Peter is right to suppose the gratuity of evil is a matter of moral judgment. The reason for this is that our moral judgments include, not only evaluations of morally good or bad states of affairs, but also judgments of what one ought to do. Since a gratuitous evil is an evil which ought be prevented if one can do so, whereas a non-gratuitous evil is an evil which one may be permitted or even required to occur, skepticism about the gratuity of evil does amount to moral skepticism.

[Bill Vallicella]

Gentlemen,

Let's avoid italics. There appears to be a glitch in the system, or else you are failing to turn them off. The forward slash is the switch to turn off an HTML command.

[Bill Vallicella]

TaiChi,

A red herring is something that distracts from the point at issue. The point at issue is whether or not some evils are gratuitous. The issue is not whether we can know whether or not some acts or events are evil. I don't think anyone here is questioning whether we have moral knowledge. So that's a red herring. What I am asking is: how do we know, of an act or event judged to be evil, that it is gratuitously evil?

If Peter can prove that one evil is gratuitously evil, then he has proven that God does not exist.

>> Since a gratuitous evil is an evil which ought be prevented if one can do so, whereas a non-gratuitous evil is an evil which one may be permitted or even required to occur, skepticism about the gratuity of evil does amount to moral skepticism.<<

That's interesting. Perhaps we need to distinguish between general moral skepticism (we cannot have moral knowledge at all) from skepticism about judgments of gratuitousness.

But note that I am not advocating either brand of skepticism. I am simply asking HOW one knows of a given evil that it is gratuitous.

No doubt it seems to Peter that the world is filled with gratuitous evils. But if theism is true, those are mere seemings, and indeed in every possible world in which there are such seemings.

[Bill Vallicella]

Italics off

[Account Deleted]

(I) Moral Skepticism.

1) Bill keep insisting that moral skepticism in the present context is a “red herring”; yet he admits that the central question is

“how do we know, of an act or event judged to be evil, that it is gratuitously evil?” (Bill Vallicella | Saturday, March 19, 2011 at 06:31 PM)

2) So by Bill’s own admission the problem is whether or not we can have a certain kind of *knowledge* and under what conditions do we have such knowledge. Denying that a certain kind of knowledge is possible is asserting that there are no conditions under which such knowledge is possible. But this is clearly a form of skepticism. Two questions arise:

(a) What kind of skepticism is involved?
(b) Is the skepticism in question moral skepticism?

I will address the second question first.

3) Are judgments about gratuitous evil moral? The term ‘evil’ is misleading. It may simply mean a case of suffering or death, both of which are in and of themselves undesirable. So judgments of evil in this sense are not inherently moral, for not everything that is undesirable is immoral. But in the present context we mean much more. We also mean that the suffering or death is at least prima facie morally wrong; i.e., E appears to be a case of a moral wrong. The next question is whether E is a case of gratuitous evil; i.e., not necessary for any good nor to prevent a worse evil. Since these questions are completely laced with moral considerations, judgments about whether a case of evil is gratuitous are moral questions. I can’t see how this conclusion can be avoided.

4) Putting Humpty-Dumpty (i.e., (2) and (3)) together yields that Bill’s question involves moral skepticism. For his question is how, and whether, we can know that a case of evil is gratuitous. Since judgments (correct or incorrect) about gratuitousness are moral judgments, the question is smack in the middle of the moral sphere. Moreover, since one answer (Bill’s!) might be that we can never know whether a case of evil is a genuine case of gratuitous evil, it certainly involves skepticism.

5)Therefore, moral skepticism is not a “red herring”.

6)So it seems that Bill is wrong about the “red herring” claim. But, Bill might have something else in mind. And now I turn to the firsts question raised in (2) above. We distinguish between two forms of skepticism: epistemological vs. ontological skepticism. Epistemological skepticism is the claim that knowledge about a given subject matter is not possible because the requisite conditions on *justification* cannot be satisfied regarding beliefs about the subject matter in question, although propositions about the subject matter are true or false.

Ontological skepticism, by contrast, denies that the *truth-conditions* of putative propositions about a given subject matter can be satisfied because suitable *truth makers* do not exist (e.g., fictional entities). Therefore, ontological skepticism denies knowledge claims not because certain suitable conditions on justification cannot be satisfied, but because there is nothing to know: there are no truths about the subject matter because the world fails to include suitable truth makers.

Notice that ontological skepticism trumps epistemological skepticism in the sense that if ontological skepticism is true regarding a given subject matter, then epistemological skepticism is irrelevant regarding that subject matter. This is so because if ontological skepticism holds, then there are no truths to know about the given subject matter (except this very truth) and, hence, questions of justifications cannot arise.

7) I suspect that Bill thinks that moral skepticism of the epistemic variety is a “red herring” because ontological skepticism holds regarding gratuitous evil. Therefore, epistemological skepticism does not even enter the picture. It is in this sense that Bill thinks that the sort of moral skepticism I have been talking about is a “red herring”.

8) Why does Bill think that ontological skepticism holds regarding cases of gratuitous evil? Bill maintains (wrongly, I shall argue) that God’s existence entails that cases of gratuitous evil cannot exist. Hence, the challenge for the proponents of the evil-argument is to prove that cases of gratuitous evil can exist without begging the question against the theist: i.e., without presupposing that God does not exist. And Bill thinks that it is simply impossible to show that cases of gratuitous evil can exist without begging the question about God’s existence. Since the possibility of God’s existence has not been ruled out and since the possibility of God’s existence entails the impossibility of gratuitous evil, Bill concludes that the proponents of the evil-argument cannot make their case.

(II). God, UPD, and Gratuitous Evil.

1) Bill thinks that the following proposition is true:

(P) Necessarily, if God exists, then it is not possible for gratuitous evil to exist.

2) I believe that (P) is false. For instance, Deism maintains that a divine being such as God exists, it created the world including moral agents such as ourselves, yet it does not need to or simply does not maintain it by keep interfering with the world’s goings on. Now, if Deism is true, then the question of whether or not God prevents certain evils or could do so does not arise. Hence, under these conditions, it is very likely that some evils are gratuitous. There may be other models which admit the existence of a God, but allow for gratuitous evil to exist.

3) Unlike (P), the following proposition is true:

(P*) Necessarily, if God exists & UPD is true, then it is not possible that gratuitous evil exists.

Reminder: ‘UPD’ stands for ‘unknown-purpose-defense’. UPD is the view that every case of evil allowed by God is necessary in order to bring about a desirable good or in orders to prevent a worst evil.

4) Assume that the following is true:

(i) Some cases of gratuitous evil exist.

(P*) together with (i) entail:

(ii) Either God does not exist or UPD is false.

5) Now suppose also that God exists. Then this assumption together with (ii) entails that UPD is false. Therefore, God’s existence is compatible with the existence of gratuitous evil, provided that UPD is false.

6) I maintain that there are many other independent reasons to reject UPD (which I will not currently discuss). However, suffice it to say that if I am right, then Bill’s view that God’s existence alone entails ontological skepticism about gratuitous evil is mistaken. God’s existence together with UPD entails ontological skepticism regarding gratuitous evil. In the absence of UPD, God’s existence is compatible with gratuitous evil. Therefore, under these conditions, ontological skepticism about gratuitous evil is unwarranted. And if ontological skepticism about gratuitous evil is unwarranted, then the only question that remains is epistemological skepticism: i.e., Can we ever be justified in our belief that some apparently gratuitous evil is indeed so? And this question, I believe, we can answer in the affirmative, unless of course one insists upon radical moral skepticism of an epistemic variety. Such skepticism, however, is endorsed by neither theists nor atheists.

[TaiChi]

No comment on Peter's response, Bill?

[Bill Vallicella]

I'm working on a separate post that will attempt to focus the discussion as time permits. Thanks for your interest.

[Bill Vallicella]

Peter,
Surely the deistic concept of God is not under discussion. The argument from evil is not an argument against that God.