The
argument from evil, on the other hand, is an attempt to show the nonexistence of God from the fact of evil, where 'fact of evil' is elliptical for 'the existence of natural and moral evils of the kinds and in the amounts encountered in our world.'
THE DIFFERENCE BETWEEN THE PROBLEM OF EVIL AND THE ARGUMENT FROM EVIL
The main difference between the problem of evil and the argument from evil is that the former is an ad hominem argument whereas the second is not. I am using ad hominem in the way Peter Geach uses it on pp. 26-27 of his Reason and Argument (Basil Blackwell 1976):
This Latin term indicates that these are arguments addressed to a particular man — in fact, the other fellow you are disputing with. You start from something he believes as a premise, and infer a conclusion he won't admit to be true. If you have not been cheating in your reasoning, you will have shown that your opponent's present body of beliefs is inconsistent and it's up to him to modify it somewhere.
As Geach points out, there is nothing fallacious about such a dialectical procedure. If A succeeds in showing B that his doxastic system harbors a contradiction, then not everything that B believes can be true. Now can an atheist prove the nonexistence of God in this way? No he cannot: at the very most he can prove (with the aid of various auxiliary premises that he and his interlocutor both accept) that God exists and Evil exists cannot both be true. But it does not follow therefrom that God exists is not true. For the atheist to transform the ad hominem problem of evil into a non-ad hominem argument from evil, he would have to establish, or at least assert, that evil exists, and not merely that the theist believes that evil exists. To see my point consider the following conditional, where P is the conjunction of auxiliary premises:
C. If evil exists & P, then God does not exist.
The atheist who raises the problem of evil for the theist asserts (C), or rather a proposition of that form. But to assert a conditional is not to assert its antecedent, or its consequent for that matter; it is to assert a entailment connection between the two. Now although it is the case that for each argument there is a corresponding conditional, and vice versa, arguments must not be confused with conditionals. In Arguments and Conditionals I explain why they are different. Transforming (C) into an argument from evil yields:
Evil Exists
P
-----
God does not exist.
Clearly, an atheist who gives this argument, or rather an argument of this form, must assert both premises. Doing so, he ceases his ad hominem examination of the consistency of another person's beliefs, beliefs he either rejects or takes no stand on, and 'comes clean' with his own beliefs.
THE ARGUER FROM EVIL NEEDS TO AFFIRM OBJECTIVE EVIL
If the atheist's aim is merely to poke holes in the logical consistency of the theist's belief-set, then it doesn't matter whether he thinks of evil as objective or subjective. Indeed, he needn't believe in evil in any sense. He could hold that it is an illusion. But if the atheist's goal is to support his own belief that God does not exist with an argument from evil, then he needs to maintain that evil is objective or objectively real.
Consider all the enslavement of humans by humans that has taken place in the history of the world. Suppose it is agreed that slavery is morally wrong. What makes this true? Define a moral subjectivist as one who agrees that the claim in question is true, but holds that the truth-maker of this moral truth, and of others like it, is an individual's being in a psychological state, say, the state of being repulsed by slavery. For the moral subjectivist, then, sentences like 'Slavery is wrong' are elliptical for sentences like 'Slavery is wrong-for-X,' where X is a person or any being capable of being in psychological states. Furthermore, the moral subjectivist grants that moral claims have truth-makers, indeed objective truth-makers; it is just that these truth-makers involve psychological states that vary from person to person.
Now if our atheist subscribes to a theory of evil along those lines, then, although there will be objective facts of the matter regarding what various individuals feel about the practice or the institution of slavery, there will be no objective fact of the matter regarding the wrongness or moral evil of slavery.
If so, the fact of evil subjectively construed will have no bearing on the existence of God, a fact, if it a fact, that is objective. Suppose a torturer tortures his victim to death solely for the satisfaction it gives him. And suppose that moral subjectivism is true. Then the torturing, though evil for the tortured, is good for the torturer, with the upshot that the torturing is neither good nor evil objectively. Now if I were on the scene and had the power to stop the torturing, but did not, would my noninterference detract from my moral goodness? Not at all. (The same goes a fortiori for God.) For nothing evil is transpiring: all that is going on is that one person is securing his pleasure at the expense of another's pain. If you insist that something evil is going on, then that shows that you reject moral subjectivism. But if you accept moral subjectivism, then nothing evil is going on; the torturing is evil only in the mind of the victim and in the minds of any others who symptathize with him. If you accept moral subjectivism and continue to insist that the torturing is evil, then you would also have to insist that it is good, since it is good from the persepctive of the torturer. But if it is both good and evil, then it is (objectively) neither.
What I am claiming, then, is that the atheist arguer from evil must construe evil objectively. This will result in trouble for the atheist if it can be shown that objective evil cannot exist unless God exists. For then the atheist arguer from evil will end up presupposing the very being whose existence he is out to deny. No doubt this is a big 'if.' But it is worth exploring.
Bill,
It took me some time to comment on this post. Sorry!
Let me focus in this reply on the so-called logical argument from evil (LAFE, for short).
You distinguish two cases.
(I) First, the purpose of LAFE is to unveil a logical inconsistency within a theistic system of beliefs. Let us call this the Inconsistency-Argument-from-Evil.
(II) Second, the purpose of LAFE is to prove the proposition 'God does not exist' from the proposition 'Evil exists' together with some other premises representing the theist's core beliefs. Let us call this the Non-Existence-Argument-from-Evil.
Now, you point out quite correctly that while the former does not commit the proponent of LAFE to the existence of objective-evil, the later does. And if it turns out that (and this requires a separate argument) that objective-evil requires the existence of the theist's God, then the proponent of LAFE of the second type is in deep trouble: his or her argument collapses. I think you are right.
My question is this:
Why should a proponent of LAFE opt for the Non-Existence-Argument-from-Evil directly, when s/he might be able to achieve (almost) the same result indirectly?
Here how:
(i) The atheist proposes first the Inconsistency-Argument-from-Evil;
(ii) The atheist then points out that the theist cannot rationally accept all the propositions represented by the Inconsistency-Argument; something has to be given up.
(iii) What are the theist's options?
(a) Reject the premise that God Exists;
(b) Reject the premise that objective-evil exists;
(c) Reject and/or modify the premises involving God's divine perfections; e.g., God is not morally-perfect, or omniscient or omnipotent.
(d) Argue that one of the auxiliary premises used in the argument is not a necessary truth and, therefore, the soundness-conditions of the Inconsistency-Argument are not satisfied: i.e., while the argument is valid, it is not sound; hence, the atheist cannot legitimately conclude that the primary premises are inconsistent.
(iv) The theist cannot opt for (a) and should not at this stage;
The theist also should not pursue (b), because it will lead to other very unfavorable outcomes;
The theist will certainly resist (c) because it involves accepting the idea that God's perfections turn out not to be unlimited and as "perfect" as we thought initially (this involves some interesting questions about the coherence of unrestricted concepts);
So the theist should pursue (d) and hope it offers an escape route from LAFE (it is the one that Pike, Plantinga, and others take). Suppose it does not work!
Now, the proponent of LAFE could proceed to the next stage:; namely, prove that all the options the theist might pursue, (a)-(d), are logically equivalent to accepting the proposition 'God does not exist'. However, such a proof might indeed commit the atheist to the existence of objective-evil.
Now, IF it can be shown that objective evil presupposes the existence of a deity featuring certain divine properties, then the atheist is in serious trouble. But the atheist is in serious trouble only when pursuing (II) above; i.e., the Non-Existence-Argument-from-Evil. A proponent of LAFE might just be content with (I), and watch some baseball while the theist sweats out his or her options.
peter
Excellent comments. I hope I can get to them tomorrow.
Take your time.
peter
Yes, the LAFEr might take this tack. But isn't the theist within his epistemic rights in saying that, although he cannot figure out how to rebut the LAFEr's inconsistency argument, he does have reason to believe that his central tenets are consistent as I tried to explain in More on the Compatibility of God and Evil?
Briefly, if I have reason to believe that God exists and reason to believe that evil exists, then I have reason to believe that these beliefs are consistent. And this despite my inability to explain exactly how they are consistent. Wouldn't it then be a stand-off?
Bill,
I think your proposal has merit. There are two ways of approaching the consequences of LAFE.
(a) LAFE shows that the relevant theistic beliefs are logically inconsistent just like Russell's famous example against Frege's comprehension axiom shows that the later is untenable. (There are other such examples that show that the concept of an unrestricted collection in set theory is unacceptable because it leads to contradictions). Now Frege had fairly good reasons to believe in his comprehension axiom and, therefore, he had good reasons to think that his comprehension axiom is true and compatible with the rest of his axioms. Russell's example shows that he was wrong in these beliefs.
(b) LAFE shows that there is a *paradox* with maintaining the relevant theistic beliefs just like Zeno's example shows that there is a paradox in maintaining that motion exists (together with a few other premises). Since we have very strong rational reasons to maintain the belief that motion exists, we are not inclined to conclude from the paradox that motion does not exist; instead we attempt to show that there has to be something wrong with Zeno's argument or some of the other premises.
Now, it is very difficult to decide on any given occasion which option (a) or (b) is most suitable. You think that there is room for the theist to opt for (b) and I certainly agree. For I do not see LAFE, and never have seen it, as sufficient to refute the proposition that God exists. But I do see it as forcing the theist into a position where it is incumbent upon him or her to review their traditional *conception* of a deity featuring the well-known perfections.
Personally, I think there is a very serious difficulty with the conception of any being that is a rational agent (i.e., their actions are intentional and guided by rational considerations), active in the world, and at the same time it features such unrestricted properties as omnipotence, omniscience, and moral-perfection. I think LAFE-type arguments indicate that indeed there is a problem here. I myself have been working on a way out of this predicament on behalf of the theist. It is not a way that retains everything reflective theists wish they could, but then again this is the price one pays for taking reflective thinking seriously.
peter
That is a very intelligent and balanced response. You write,
. . . I do not see LAFE, and never have seen it, as sufficient to refute the proposition that God exists. But I do see it as forcing the theist into a position where it is incumbent upon him or her to review their traditional *conception* of a deity featuring the well-known perfections.
I agree with that. You appreciate that the theist is not tied down to a particular conception of deity. He is free to modify his conception -- within limits of course -- just as the physicist is free to modify his conception of nature.
You sepak of the omni-attributes as "unrestricted." But surely they are restricted? Perhaps a post on this topic is in order . . .
Bill,
It has been absolutely my view and there are several interesting option for the theist in this direction. Some very interesting things might emerge.
I do think that a post on the nature and limitations of the perfections would be both useful and interesting.
peter