The Challenge
Stephen Law has issued the ‘evil God’ challenge. This challenge argues that it is at least as plausible to argue for an omnipotent, omniscient, and totally evil God as it is to argue for an omnipotent, omniscient, and totally good God. Associated with that challenge is the ‘symmetry thesis’, which is that for every argument for a good God there is an equally strong argument for an evil God, which is exactly similar in form. I will attempt to refute both these suggestions.
The concepts of omnipotence and omniscience are notoriously difficult to define in non-paradoxical ways. I am sure such non-paradoxical definitions can be found, though they may be rather complex. For the evil God challenge, however, ‘Omnipotence’ seems to mean that God can do absolutely anything which can be described without self-contradiction. I shall take it in this sense. ‘Omniscience’ seems to mean that God knows everything, past, present, and future, and in the fullest possible way, and I shall use it in this sense. There are many different Gods, or ideas of God, and some are evil, vindictive, or malevolent. But the idea of God I will consider is the idea of a being who is always and necessarily omnipotent and omniscient. An evil being is a conscious being who knowingly and intentionally causes suffering to others, either by indifference or positive cruelty.
The argument from knowledge of suffering.
Armed with these definitions, we may ask, could an omnipotent omniscient being intentionally cause suffering to others? It obviously could, being omnipotent. But would it? You can imagine a very powerful and knowledgeable being who does so – films like ‘Predator’, ‘Prometheus’, and ‘Alien’ depict such beings. This is because such beings have evolved by competition with other species, and so have learned to enjoy hunting and fighting, and are indifferent to the sufferings of less evolved species. God, however, has not evolved. Being omnipotent, God does not have to compete with any other powers, fighting would be futile, since God would effortlessly win any conflict, and hunting would be unnecessary, since God could find and destroy anything at will. An omnipotent being has no reason to cause harm to others, since it is never threatened or hurt in any way.
But might God not just enjoy causing harm to others, as a matter of fact? Being omniscient, I assume that God will know what it is like for another being to feel suffering. Suppose God has no feelings. Then God will take no pleasure in causing harm, since nothing can cause God either pleasure or pain. It cannot be a reason for causing something to exist that it is rationally undesirable. There is then no reason to cause harm. But it is a reason for causing something to exist that it is objectively good or rationally desirable. There is then some reason to cause good. Therefore an omniscient and omnipotent being will cause good, not harm.
If, on the other hand, as I think more likely, God has feelings, then God knows what suffering is like, and knows, whenever others suffer, exactly what their suffering feels like. This entails that God will feel that suffering with perfect empathy. That is what total omniscience involves. If God feels pleasure or pain, and God is omnipotent, God will be able to choose divine mental states that are pleasant. An omnipotent God will not choose mental states of pain. God will choose to be happy. An omniscient God will therefore not cause pain to others, since that entails some corresponding lessening of pleasure in God. God has good reason not to cause harm to others, but to cause pleasure and happiness in others. Fully empathetic knowledge of the happiness of others will increase God's own happiness. Therefore an omnipotent omniscient God will not intentionally be evil.
This argument also undermines the symmetry thesis, since there is no corresponding argument from omnipotence and omniscience to absolute evil. Very powerful and knowledgeable wicked beings could have a reason for causing harm – they still compete with others, and may ignore the feelings of others completely, and may even take pleasure in seeing them squirm. An omnipotent being has no competition, and feels the feelings of others, so it has no reason to cause pain and good reason to cause happiness in others. As Stephen Law says, ‘no-one does bad knowingly’, and an omnipotent being cannot desire anything that causes pain or discomfort to itself, or to others if it must share in that discomfort.
The Platonic argument.
In discussing Plato's ‘Euthyphro’ argument, Stephen Law argues that goodness is independent of the arbitrary will of a God. Most Christian theologians agree with this. Yet goodness is not independent of God. The usual Christian view is that good states – states or processes of intrinsic value, worth-while for their own sakes – are necessarily what they are, and are parts of the necessary divine nature. Indeed, God is usually seen as an instantiation – an actual instance, and the only instance – of supreme intrinsic value. God is the Good itself. As instantiating to a supreme degree every intrinsic value, God will be maximally powerful, knowing, and blissful. If positive co-operation, friendship, and self-giving love are necessarily good, independently of what God may arbitrarily will, then those values will also be instantiated in the Supreme Good. Therefore if goodness really is objective, and if objective and necessary truths are part of God's nature, then God will necessarily not be evil.
The symmetry thesis fails here too. There is no argument from evil to the existence of Supreme Evil. For evil is not of intrinsic value. It is of intrinsic disvalue – no rational being could desire it to exist for its own sake. There could not be a supremely evil being, for that would be a being which could not will itself to exist. In fact the idea of a wholly evil being is self-refuting. It would have to realise all disvalues. So it would be maximally weak, maximally ignorant, in maximal pain and misery, and maximally depressed. It would know that it would be better if it did not exist, so it would not be true that there is any reason for it to exist (whereas there is the best possible reason for a maximally valuable being to exist). If it was omnipotent, it would commit suicide. But it is too weak to do so. A very miserable and weak and ignorant God would not be maximally evil anyway, since it would be unable to cause maximal evil successfully; so the idea of a supremely worthless being is incoherent.
The Kantian argument
An omnipotent being would choose to act rationally, not irrationally, if only in order that it could discern efficient means to its goals, and understand how to build a universe. Any rational being knows that a principle that applies to a being X in situation S, will apply to any being of type X in situations relevantly like S. So if pain is undesirable for an intelligent being like God it is undesirable for any intelligent being, just as such. A perfectly rational being, such as God must be, will therefore find it undesirable to cause pain in others. It might be argued that God's pleasure in causing pain vastly outweighs all the pains that are caused. But, as well as the implausibility of this view, it would clearly be more rational to avoid the balancing problem by not causing pain in others. If God's pleasure was increased by the pleasures of others – and an omnipotent being could arrange this – any rational pleasure/pain calculus would see that the greatest pleasure would be caused by not causing suffering. As a Kantian would put it, any fully rational being would will that all rational beings, including God, act on a principle of doing good to others, as long as all rational beings would have their happiness thereby increased.
The ‘problem of good’ is not just that there are lots of beautiful and pleasing things around. It is that there are objective moral obligations (like the obligation to act only on a principle that each rational being could will all rational beings to adopt), as well as objective and intrinsically valuable states (states that any rational being would have a good reason to choose). If these obligations and values are grounded in God, then God will necessarily act in accordance with them. An evil God would have to act in contradiction to its own essential nature, and that would be irrational. An omnipotent God will be a rational God, and a rational God will be a good God.
The mental pathology argument
Any being that takes pleasure in cruelty is pathological or diseased. A being can act cruelly in a competitive world in order to gain power or wealth – but that could not apply to God, who already has unlimited power and wealth. How then could one explain taking pleasure in another's pain? Only if the other is treated as not a person of the same sort as oneself, as sub-personal. Such a belief is based on false or inadequate knowledge, which an omniscient being would not have. Moreover, one of the greatest intrinsic goods, producing one of the highest forms of happiness, is sharing experiences with others and co-operating with them in achieving one's goals. An evil God would lack these goods. Nothing would be worth creating, and nothing would be an object of appreciation and delight, for everything would exist only to be damaged and destroyed. Lacking love for anything, evil God would not even love itself. If God loved Godself, then as a matter of logic God would love anything relevantly like itself. God would not cause pain to what it loves. Conversely, hatred of other persons entails hatred of persons as such, and that, when thought through, entails hatred of one's own personal being. Evil God's being would ultimately consist in self-hatred. No omnipotent and omniscient God could choose such a negative existence.
The argument from worship
People do not believe in God primarily as a theoretical explanation of the universe. They seek an ideal, greater than themselves, to which to give their loyalty. They seek deliverance from moral failure and despair. They seek forgiveness and moral renewal. The question for them is: what is worthy of worship? What can bring hope out of despair? What can deliver us from evil? This search entails that their God must be powerful and knowledgeable enough to deliver from evil. But there is no analogous search for an evil God, who could only bring increased despair and oppression. This is not an argument that there is a God. It is an argument that the most powerful reasons for believing in a good God cannot be paralleled by arguments for an evil God. Thus it is an argument that the symmetry thesis is importantly misleading.
This is my response to the evil God challenge. It does not solve the problem of evil – show why there is so much evil in the world. It does not establish the existence of God. But it does, in my opinion, show that there could not be an omnipotent, omniscient, and totally evil God. While Stephen Law admits that the symmetry thesis does not apply in all cases, the asymmetries outlined in this response are so important that they put the thesis itself in question.
I must conclude, however, with the thought that I find the definitions of ‘omnipotence’ and ‘omniscience’ used here misleading. For instance, God is necessarily what God is, and so God cannot be otherwise, and cannot change the divine nature, do evil, or commit suicide. Consequently, it is better to say that God is the most powerful being there could possibly be. But God might still be subject to internal necessities of the divine nature. With regard to ‘omniscience’, a better definition might be that God knows more than any other being could possibly know. But even God might not know, for instance, things that God has not decided yet.
If God is maximally powerful and knowing, and eternally and necessarily so, then God is already constrained by features that are necessary to the divine nature. Therefore not everything happens by the free intention of God (for instance, God cannot choose to alter the divine nature). Then the problem of evil is to show why intrinsically undesirable states are necessary to the values that a necessarily good God creates. I believe this can be shown in outline by using four arguments: the entangled cosmos argument, the emergent purpose argument, the libertarian freedom argument, and the afterlife argument. That is another story.
