First, let me summarize Law’s position. Law claims in his article that “even if most of the popular arguments for the existence of God do provide grounds for supposing that there is some sort of supernatural intelligence behind the universe, they fail to provide much clue as to its moral character.” In particular, Law says, even if a design argument could show that such an intelligence exists, it could no more show that the intelligence in question is supremely benevolent than that it is supremely malevolent. In fact, he suggests, the overall evidence such arguments appeal to should lead us away from belief in a supremely benevolent supernatural intelligence. Law allows that what is often labeled the “logical problem” of evil – which supposes that the existence of evil is strictly incompatible with the existence of a good God – may not pose a serious challenge to theism. But he thinks the “evidential problem” of evil – which assumes only that the existence of evil is strong evidence against the existence of a good God – does pose a serious challenge, at least given that there are no strong arguments for the existence of such a God. And the standard theodicies – such as appeals to free will, to soul-making, or to the way in which certain goods presuppose evils – succeed in explaining at most only some of the evil that exists, not all of it, so that the overall evidential situation still fails to point in the direction of a supremely benevolent God.
So far all of that is just standard atheist argumentation, and Law’s overall position takes it for granted. In particular, Law presupposes that there are no strong arguments for God’s existence, that even if there were they wouldn’t lead us to a supremely good God, and that the evidence we do have points away from the existence of such a God. Law’s innovation is to suggest, first, that the hypothesis of an “evil god” – an omnipotent, omniscient, but supremely malevolent intelligence – is at least as well supported as the hypothesis of a supremely good God. And if a skeptic were to pose against such a hypothesis the challenge of an evidential “problem of good” – that is, if a skeptic were to ask why a supremely malevolent intelligence would allow the good that exists in the world – the defender of an “evil god” hypothesis could offer “reverse theodicies” which parallel the theodicies put forward by theists. He could say, for example, that free will makes possible certain evils that an evil god couldn’t realize without it; that certain evils presuppose the existence of good; that the evil god intends the world to be a vale of soul-destruction, which requires that there be some good in it so that we can be tormented by its loss; and so forth.
Now, Law is happy to acknowledge that such defenses of the evil god hypothesis would not be very strong. But he thinks they are no weaker than the parallel attempts to defend the existence of a good God. There is, he says, a conceptual and evidential “symmetry” between the two views. But everyone, including theists, acknowledges that there is no good reason to believe in the existence of an omnipotent, omniscient, and supremely malevolent intelligence. So, shouldn’t they also acknowledge that there is no good reason to believe in a supremely good God? Isn’t the one view as unreasonable as the other? That is Law’s “evil-god challenge.”
So, what should we think of all this? Well, having now read Law’s paper, I must say that I find that my original comments, based on Eric’s summary alone, were exactly on the mark. Law’s argument may be an interesting challenge to a theistic personalist conception of God – I’ll leave it to theistic personalists themselves to figure out how they might respond to it – but it is completely irrelevant to classical theism. And that is no small lacuna. It means that Law’s argument is completely irrelevant to evaluating the truth of theism as it is understood by writers like Aristotle, Plotinus, Augustine, Anselm, Maimonides, Avicenna, and Aquinas (to name just a few), and as it has been defined within Roman Catholic theology and traditional Christian theology more generally.
The reasons why should be evident from my recent post on the differences between classical theism and theistic personalism (which readers who are unfamiliar with those differences might want to read before continuing on). Consider, first, that central to classical philosophy and to the classical theist tradition that it informed is the thesis that evil is a privation, the absence of a good that would otherwise obtain rather than a positive reality in its own right. Accordingly, for classical theism, there simply is no symmetry between good and evil of the sort that Law’s argument requires. Astonishingly, though, Law’s article does not even consider, much less respond to, this core element of the classical theist position, despite the fact that he evidently regards his argument as a challenge to all forms of theism, and not just to non-classical forms. To borrow an analogy from another recent post, this is like evaluating The Godfather Trilogy without mentioning that the lead characters are Mafiosi, and focusing instead only on one of the romantic subplots in The Godfather Part III.
But Law did have something to say about the subject in his combox remarks. Here are those remarks, quoted in full:
Fesser’s [sic] “refutation” of my evil god argument is awful:
(i) it depends on the privation view of evil, which is wrong. (Why not flip this and say good is a privation of evil?!) Actually, *some* evils, like blindness, are best seen as privations of goods. But many appear not merely to be merely privations. And in fact in some cases it is more natural to see the good as a privation of evil (look up “peace” in the dictionary). That evil is in every case nothing more than a privation of some good is a myth that even many theists reject (philosopher Tim Mawson, for example). Fester [sic] is one of those theists who, when asked to justify the privation view, waffle and refer us to Aquinas, Aristotle, etc. Ask him him [sic] to explain, clearly, *exactly* what the argument is.
(ii) in any case, the privation view is not obviously incompatible with the existence of an evil God (we are at least owed some explanation for why it is – this is particularly clear if we see good as an abstract Platonic Form, say. (Fesser [sic] at this point just seems to *define* God as good – well, that doesn’t establish the impossibility of an evil God!)
(iii) even if the privation view were incompatible with an evil God, and it could thus be shown that an evil God was impossible, the evil God challenge can *still* be successfully run, as I point out in the paper. Perhaps Fesser [sic] should read it.
Let me respond to these points in order. Regarding (i), I cannot resist noting first of all that it is rather silly of Law to complain that I “waffle and refer us to Aquinas, Aristotle, etc.” rather than “explain[ing], clearly, *exactly* what the argument is,” when what he is replying to is something I said in a brief combox response to a reader’s off-topic question, not a formal argument presented in a book, a paper, or even a blog post. This sort of thing is depressingly common on the Internet: “You didn’t prove the truth of [Thomistic metaphysics, Darwinian evolution, quantum mechanics, etc.] to my satisfaction in your latest combox remark; therefore you’re an idiot!” One would hope a professional philosopher like Law would be above it (and perhaps he is – as I have acknowledged, he was probably just blowing off steam, which we all do from time to time). Suffice it to say that I have in fact addressed these issues at length elsewhere, such as in my book Aquinas. To paraphrase someone, perhaps Law should read it.
Second, since Law is the one claiming that his “evil-god challenge” is a threat to theism generally, including classical theism, the burden of proof is on him to show that the “evil as privation” view is false, not on the classical theist to show that it is true. It would be perverse for a critic of The Godfather Trilogy of the sort alluded to above to insist that those who disagree with him have the burden of showing that the organized crime theme really is, contrary to his analysis, a significant part of the story. It is similarly perverse for Law to insinuate – and in a dashed off combox afterthought at that, rather than in his original article! – that classical theists are the ones who need to show that the privation analysis that is central to their position is something he needs to trouble himself with.
Third, that means that Law has a lot of homework to do before he can pretend to have shown that his “evil-god challenge” really threatens theism generally, because it is evident from his remarks that he doesn’t understand the privation view, much less the classical theist tradition of which it is a part. To understand that view, one must first understand classical essentialism, whether of the Platonic or Aristotelian variety. That is a big topic – again, see Aquinas for the details – but it is clear enough how a privation view follows from the thesis that things have essences. For example, if it is of the essence of the visual apparatus – eyes, optic nerves, relevant areas of the brain, and so forth – that it serves the function of enabling an organism to see, then obviously blindness is a defect and it would be silly to suggest that perhaps it is sight that is the defect insofar as it involves the absence of blindness. Law himself acknowledges that the privation view is the most plausible way to understand blindness. He nevertheless insists that such an analysis wouldn’t work in all cases, yet he doesn’t offer any examples, and if some form of classical essentialism is true, the privation analysis would apply across the board.
The only purported counterexample to the privation view Law does suggest is, not a case of an evil which is not a privation, but rather an example of a good – peace – which, appealing to the dictionary, he evidently would define (quite plausibly) as the absence of war. Now, the privation view is certainly not the sort of thing one could refute by appealing to dictionaries, because it is not a theory about how we use words like “good” and “evil,” but rather a theory about the metaphysical status of good and evil themselves. But that is beside the point in the present case, because the privation view doesn’t entail that there are no goods that can be defined in terms of the absence of evil. It holds only that not all goods are so definable, while all evils must ultimately be understood in terms of the absence of some good. That is to say, the bottom level of the analysis of good and evil will include only goods, even if there will also be some goods appearing at higher levels of the analysis. In the case of war, the analysis will involve an appeal to the idea that moral goods are to be understood in terms of the ends set for us by nature. Since among those ends is giving to others what is due to them, war can be analyzed as a certain kind of failure to give others their due, namely by using force to take from them what they have a right to (their lives, property, security of their borders, etc.). Even if peace is the absence of war, then, war itself is the absence of a certain kind of good, a good which cannot in turn be analyzed in terms of the absence of some evil.
Regarding Law’s point (ii), for Law to claim that I “just seem to *define* God as good” – as if what is in question here is some eccentric ad hoc stipulation on my part – and to assert that “the privation view is not obviously incompatible with the existence of an evil God,” is just to manifest his unfamiliarity with, or at least to ignore, the central arguments of the classical theistic tradition and the metaphysical ideas underlying it. For when one takes account of those ideas – the act/potency, essence/existence, and simple/composite distinctions; the doctrine of the convertibility of the transcendentals; the principle of proportionate causality; the doctrine of privation; and so on – there is no mystery at all as to why the classical theist regards a demonstration of God’s existence as ipso facto a demonstration of that which is necessarily devoid of evil. Given the underlying metaphysics, to assert that God cannot possibly be evil is no more a matter of arbitrary stipulation than saying that the Pythagorean Theorem must hold of right triangles is a matter of arbitrary stipulation.
Consider that the classical (Platonic, Aristotelian, and Thomistic) arguments for God’s existence are arguments to the effect that the existence of compounds of act and potency necessarily presupposes the existence of that which is Pure Actuality; that the existence of compounds of essence and existence necessarily presupposes the existence of that which is Being Itself; that the existence of that which is in any way metaphysically composite presupposes that which is absolutely simple; and so forth. Given the doctrine of the convertibility of the transcendentals, on which being is convertible with goodness, that which is Pure Actuality or Being Itself must ipso facto be Goodness Itself. Given the conception of evil as a privation – that is, as a failure to realize some potentiality – that which is Pure Actuality and therefore in no way potential cannot intelligibly be said to be in any way evil. Given the principle of proportionate causality, whatever good is in the world in a limited way must be in its cause in an eminent way, shorn of any of the imperfections that follow upon being a composite of act and potency. Since God is Pure Actuality, he cannot intelligibly be said either to have or to lack moral virtues or vices of the sort we exhibit when we succeed or fail to realize our various potentials. And so on. All of this is claimed to be a matter of metaphysical demonstration rather than probabilistic empirical theorizing, and the underlying metaphysical ideas form a complex interlocking network that is (as anyone familiar with Platonism or Aristotelianism realizes) motivated independently of the problem of evil or the question of God’s existence. That is to say, the concepts are not introduced in an ad hoc way so as to get around objections of the sort Law raises. They are already there in the underlying metaphysics, and rule out from the get-go objections of the sort Law raises, at least insofar as they are directed at classical theism.
Law’s point (iii) – which he develops on p. 20 of his paper – is equally misdirected, because it too simply assumes that good and evil are on a metaphysical par. Law suggests (if I understand him correctly) that any reasons a theist could have for denying that an “evil god” is in principle possible could be mirrored by reasons suggesting that a good God is in principle impossible. But that just begs the question against the classical theist, who holds that evil is metaphysically parasitic on good, and thus (given the convertibility of the transcendentals) on being, in such a way that whatever is Being Itself would have to be Goodness Itself and therefore in no way evil. Hence, since God is Being Itself, the claim “If God exists, then He is good” is metaphysically necessary, while the claim “If God exists, He might be evil” is necessarily false. In any event, since Law is the one raising the “evil-god challenge,” the burden is on him to show that the idea of an “evil God” is even intelligible given the metaphysical presuppositions that classical theism rests on, and not on the classical theist to show that it is not intelligible.
Now, I am not here attempting to convince the uninitiated or hostile reader that this complex metaphysical picture I have been describing is correct or even plausible. That would take at least a book, and since Aquinas is just such a book, I direct the interested reader to that. I am also not saying that no reasonable person who familiarizes himself with it could disagree with that picture. I am merely saying that before one disagrees with it, one ought at least to try to understand it. And the things Law says seem to me to show that he does not understand it. An atheist could intelligibly say “I don’t believe that the God of classical theism exists.” He could intelligibly reject the whole metaphysical picture – the privation view, the convertibility of the transcendentals, God as Pure Actuality, the whole ball of wax. What he cannot intelligibly say is “The God of classical theism might in principle have been evil.” Again, the metaphysical system underlying classical theism simply rules out the very idea of an “evil God” on entirely principled and independently motivated grounds – not as a matter of mere ad hoc stipulation – and thus rules out Law’s “evil-god challenge” on entirely principled grounds. Hence, if you want to reject classical theism and not just theistic personalism, you had better look for grounds other than Law’s “evil-god challenge.” To insist on pressing that challenge against it is just to demonstrate one’s fundamental misunderstanding of the position one is criticizing, like the creationist who rejects Darwinism on the grounds that he just can't see how a monkey could have given birth to a human infant.
The reason theistic personalism doesn’t rule Law’s challenge out from the get-go is that theistic personalism typically rests on a very different sort of metaphysics, and conceives of God in far more anthropomorphic terms. In particular, the theistic personalist tends not to think of God as Pure Actuality, Being Itself, Goodness Itself, or the like, but rather as “a person without a body,” like us but without our limitations, who might intelligibly be said to be morally virtuous and to have duties he lives up to. (Again, see the earlier post of mine linked to above.) Theistic personalism is also often associated with a conception of God’s relationship to the world on which it is at least in principle possible that the world might have existed apart from God, so that the question of whether God is the cause of the world becomes an “evidential” or “probabilistic” matter, rather than a matter of strict metaphysical demonstrations of the sort classical theists typically attempt to provide. Hence it becomes a real question for the theistic personalist whether the balance of probabilities really supports belief in a supremely powerful disembodied person who lives up to all his moral obligations, etc. – a way of framing the issue that is, from a classical theistic point of view, totally wrongheaded from the start. In any event, as all of this indicates, the way Law sets up his challenge to the theist clearly presupposes an essentially theistic personalist construal of theism. He does not seem to be aware that there is any difference between this construal and that of classical theism, or that it is the latter view that has, historically, characterized mainstream Christian theology and philosophical theism.
For those who are interested in exploring in greater depth the classical theist approach to the problem of evil, I recommend, as I have before, Brian Davies’ The Reality of God and the Problem of Evil. There is also the late Herbert McCabe’s God and Evil in the Theology of St. Thomas Aquinas, which was recently published and which I have only just started to read, but which promises to be a useful exposition of the Thomistic approach to the subject.