This paper develops a challenge to theism. The challenge is to explain why the hypothesis that there exists an omnipotent, omniscient and all-good god should be considered significantly more reasonable than the hypothesis that there exists an omnipotent, omniscient and all-evil god. Theists typically dismiss the evil-god hypothesis out of hand because of the problem of good - there is surely too much good in the world for it to be the creation of such a being. But then why doesn't the problem of evil provide equally good grounds for dismissing belief in a good god? I develop this evil-god challenge in detail, anticipate several replies, and correct errors made in earlier discussions of the problem of good.
So the idea is that there is an "evil God" that parallels the "good God", and that if we don't think arguments for the evil God work, then we shouldn't think the arguments for the good God work either, since the arguments for one can be paralleled in the other. Now Feser's point is that this argument, whether or not it works for a God understood along personalist lines, is not applicable to the God of classical theology, since the God of classical theology is by nature good. Hypothesizing an "evil God" is like hypothesizing a "triangle with four sides"; it is just nonsense.
My purpose here is not to rehash the arguments that followed on Feser's blog, but to explore Law's idea of an "omnipotent, omniscient, and all-evil god" that parallels the good God. Does such a being really make sense? I don't think it does, and I will explain why here.
In Law's paper, he references Charles Daniels, who comes close to making the argument that I will make. According to Law, Daniels argument is that "we always do what we judge to be good. Even when I smoke, despite judging smoking to be bad, I do it because I judge that it would be good to smoke this cigarette here and now. If follows, says Daniels, that no-one does bad knowingly. But then it follows that if a being is omniscient, he will not do bad. There cannot exist an omniscient yet evil being." Law's answer is that "I believe Daniels's argument trades on an ambiguity in his use of the word 'good.' True, whenever I do something deliberately, I judge, in a sense, that what I do is 'good.' But 'good' here need mean no more than, 'that which I aim to achieve.' We have not yet been given any reason to suppose I cannot judge to be 'good', in this sense, what I also deem to be evil, because I desire evil. Yes, an evil god will judge doing evil to be 'good', but only in the trivial sense that evil is what he desires."
The problem with Daniels's argument is that the Platonic understanding of evil is false. It is true insofar as we cannot choose an evil except under the aspect of good; I don't smoke the cigarette because I judge it to be absolutely good for me, but because I desire pleasure, and pleasure is a good, even if I know that cigarettes are bad for me in the long run. So I choose the evil that is cigarettes, knowing they are evil, but under the aspect of a good (in this case, pleasure). Why do I choose the lesser good of pleasure rather than the absolutely better good of health? Because, as Aristotle wrote, our reason does not rule our nature as a tyrant; sometimes our lower nature overpowers reason and leads us to choose a lesser good rather than a greater one. This is why moral education is necessary. Moral education not only trains us to know what is right and good, but disciplines us to develop a nature that chooses the greater good rather than the lesser one. This is the difference between being a virtuous man and a vicious one.
But to really answer Law we must put some meaning to "good" and "evil." In the exchanges over at Feser's blog, Law strenuously resists doing this, and insists he need only use the everyday, "pre-theoretical" understanding of the terms to make his argument work. Yet the words must be defined, pre-theoretically or otherwise, and Law resolutely resists any attempt to define them at any level. I think this is because, as soon as good and evil are defined, even on a pre-theoretical level, it becomes clear that good and evil are not symmetrical, and the argument from symmetrical gods collapses. And he certainly wouldn't create a universe.
Let me show this by providing definitions of good and evil, definitions that are true to our pre-theoretical understanding of the terms and, without engaging in extensive dialects over the meaning of good and evil, show that the parallel between good and evil gods collapses. I think our pre-theoretical understanding of good is that which enhances nature, and evil that which frustrates it. We think smoking is bad, for example, because it damages our health; in other words, it frustrates our body's natural ability to maintain itself. A disease that kills a small child is evil because it, obviously, frustrates the child's natural inclination to survive. Of course we might launch the argument that we have competing natural fulfillments here, since the disease fulfills its nature only by destroying the child's. But since we are staying at the pre-theoretic level and avoiding dialectics, it is sufficient to remark that we commonly understand a child to be more valuable than a disease, and so avoiding the frustration of the child's nature takes precedence.
With this pre-theoretic understanding of good and evil, let us consider the God of classical theology, the all-good God. This God does good everywhere and whenever it can; to the universe and to its creatures. What about itself? Naturally, it does good to itself as well, and so avoids frustrating its own nature. It's in the business of avoiding the frustration of nature. The classical argument from evil arises; how is it then, that so much evil exists in the universe? How is it that creatures so often find their natures frustrated? The classical answer to this question is that "God permits evil only insofar as good may come of it." The key term there is "permits"; God never frustrates the nature of any creature directly, but does permit creatures to frustrate each other's natures, and that only insofar as a further good may come of it. The point is that there is no inconsistency in hypothesizing an all-good God.
Now let us consider the parallel universe evil god, the one who is omniscient and omnipotent like the good god, but tries to maximize evil. In converse to the good god, he will do everything he can to frustrate nature, both the natures of his creatures and himself. We hit an immediate snag: Why would this god ever create anything at all? Since god is the greatest being there is, the greatest evil would be to frustrate his own nature, and so god would always do evil to himself (frustrate himself) before doing evil to anything else. But to create a universe for the purpose of doing evil to creatures (doing good that evil may come of it) is to perform the greater good for a lesser evil, since the evil done to god is always the greater evil compared to an evil done to creatures. So the evil god would always choose to frustrate himself rather than create a universe he could torture.
There is a further problem, for the parallel between good and evil can't hold. The good god permits evil that good may come of it, but he never directly does evil himself; the evil god must directly perform the good of creating a universe if he is ever to engage in the evil of frustrating his creatures. This reveals the asymmetry between good and evil that is latent in even a pre-theoretical understanding of the terms.
There is a more subtle problem with the notion of an evil god creating a universe so that he may commit evil. He commits evil by frustrating the natures of the beings he has created; so when he creates creatures, he does so for the purpose of later frustrating them. When he in fact later frustrates them, he is therefore fulfilling his own purposes; in other words, he is not frustrating his own nature but fulfilling it and, to that extent, he is good rather than evil. But the good god doesn't ever resort to evil; he is purely good. The evil god can't be purely evil; he must in part be good - so there is no real parallel between an all-good and an all-evil god.
In summary, if we use a pre-theoretical understanding of good as what enhances nature, and evil as what frustrates it, then we can see that an all-knowing, all-powerful, and all-evil god doesn't make sense. This god would frustrate himself before he frustrates anything else, since he is the greatest thing that can be frustrated. And if he did attempt to create a world on which he could perform evil, he could only do so by contradicting his own nature. This all-evil God that parallels the good God can't exist.