1. An evil God does not exist.
2. An evil God and a good God are the same
3. Therefore, a good God does not exist
4. Craig's God is a good God
5. Therefore, Craig's God does not exist
At this time, I'm not going to focus on the first premise (although it will play a part). I think that Craig handled it adequately in the debate (Randy Everest at Possible Worlds addresses the concerns with the premise and Craig's responses in his analysis of the argument). However, premise 2 is the one that did not receive much attention from Craig and where I think that he could have also shown the argument's weakness. I want to quickly go over a few observations about the second premise.
Defend the Premise
Stephen Law did not state the second premise explicitly. He only implied it. Since he did not clearly articulate it in the debate, he did not really provide a defense for it. Its truth was assumed by Law to be accepted by the audience. He indicated this by challenging William Lane Craig to show that his conclusion did not logically follow. He was demanding an argument from Craig for its falsehood before Law presented an argument for its truth. Even though I don't have a case from Law for the truth of the premise to critique, I can offer a case against its truth.
Same or Indistinguishable
Since Law did not explicitly state the premise, those who discern its presence in the argument are left with a bit of ambiguity about whether Law is speaking ontologically (an evil God and a good God are the same) or if he's merely speaking epistemologically (an evil God and a good God are indistinguishable).
This distinction is quite important. Stephen Law's conclusion that a good god does not exist, is dependent upon the two being the same. If the two are not the same (merely indistinguishable), then if one is shown to not exist, the other may still exist. This distinction is good in theory, but what effect does it have in practice against Law's case?
Good and Evil
Since Stephen Law wishes to conclude that a good God does not exist, let's assume that his second premise is speaking ontologically (that the evil God and the good God are the same). The truth of the premise is dependent upon "good" and "evil" being the same. Ironically, his first premise undermines the second by denying this truth in defense of its own truth.
Law defends the first premise by pointing out the differences between "good" and "evil" in the world. He then asks why an evil God would allow so much "good". For this question to even make sense, "good" and "evil" must be different. If they are different, then a good God and and evil God must also be different. If they are different, then his second premise is not true, the argument is not sound and the conclusions do not follow.
But what if "good" and "evil" are merely indistinguishable from each other? As mentioned above, indistinguishability allows for one to exist while the other does not. Based upon indistinguishability of the two, we could agree that an evil God does not exist, but Law would have to at least be open to the existence of the good God. The closest to atheism this argument has the possibility of taking Stephen Law or any other person is to a type of agnosticism.
But it does not end there. What was quite surprising in the debate is that it seems that Law conceded that he merely means that the two gods are indistinguishable from one another near the end of the debate. If he concedes that, then he has just admitted to the falsehood both of his premises presented above. Law cannot distinguish between "good" and "evil" to know that one God would not permit certain actions and conclude from the existence of those actions that that one God does not exist. Since he has no way to establish that one of the Gods do not exist, he cannot even make the comparison (same or indistinguishable) to another God to say that it doesn't exist either. Which means that neither of his conclusions follow. This argument fails to even get the atheist to agnosticism regarding God's existence. Rather it leaves the person with the existence of SOME god.
Now, one could claim agnosticism about whether or not to worship the god based on not knowing if the god is good or evil. But to get this far, we are dependent upon "good" and "evil" being indistinguishable (if not the same). The question of whether we "should" worship this god, is a moral question that assumes that it would be "evil" to worship an "evil" god or not worship a "good" god and "good" to worship a "good" god or not worship an "evil" god. But if the "good" and "evil" are indistinguishable, whether we worship the god or not is not good or evil. It is a matter of opinion and not an idea that we can really be "agnostic" regarding.
If one wishes to hold that the choice to worship a god is not merely a matter of opinion but rather something that they can be wrong about, then they are also holding that such a decision has an actual truth- and moral- value that is determined by being able to distinguish between "good" and "evil", which defeats the reason one is even at this point.
The "evil god" argument is an interesting twist on the problem of evil. Many times it is characterized as "the problem of good". However, the argument has too many issues to be logically salvaged. Of course, this does not mean that it won't be or is not persuasive to many who are looking to dismiss God's existence. On the surface, it is quite persuasive, but it loses its power when further investigation is completed.
- Full audio from the debate between William Lane Craig and Stephen Law from Apologetics 315: Does God Exist?
- Transcript of debate (registration required)
- Analysis of Stephen Law's first premise that an evil God does not exist from Randy Everest: Craig, Law, and Evil God: Oh My!
- Analysis of the entire debate between William Lane Craig and Stephen Law from J.W. Wartick: William Lane Craig vs. Stephen Law: Thoughts and Links
- How the implications of Law's argument affects his response to the missing ontological argument in Craig's presentation by philosopher Doug Geivett
- Stephen Law, his evil god, and radical skepticism by J.W. Wartick