One reason William Lane Craig rejects atheistic moral platonism (i.e. robust moral non-naturalism) is that, on his view, obligations require commands from a worthy authority and only God could be sufficiently worthy: “[…] the theist can make sense of moral obligation because God’s commands can be viewed as constitutive of our moral duties” . If all that existed were moral properties, but no divine commands, then we would have no moral duties, or so Craig claims.
I’ve previously considered an argument I call the Cosmological Euthyphro Dilemma (CED), according to which the origin of God’s reasons are mysterious. Either God’s reasons for God’s actions originate within God, in which case God does not have free-will, or God acts, at least sometimes, without reasons, in which case God’s actions are arbitrary and capricious. Theists welcome neither option and both can be used to argue for God’s non-existence. Furthermore, either fork in CED is incompatible with a host of traditional arguments for God’s existence.
Craig’s distinction between the Good, as constituted by God’s nature, and our moral duties, as constituted by God’s commands, reinforces the CED. To review, the atheistic moral platonist asserts morality can exist without God because moral properties, as abstract objects, exist independently of God. Craig responds that this is not enough and that God’s commands are required for moral duties. But, following the CED, we can ask whether God has reasons for God’s commands. Craig would answer that God does have reasons for God’s commands and these reasons originate within God’s nature as the standard of goodness.
We can ask whether God could have commanded otherwise. Recall the traditional Euthyphro dilemma asks whether x is right to do because God commands x or if God commands x because x is right to do. If x is right to do because God commands x, then there can be no further moral reasons as to why x is right to do, so God’s commands would be arbitrary. On the other hand, if God commands x because x is right to do, God’s commands are not arbitrary and are formed in recognition of the Good, but Divine Command Theory is false. Craig responds that these two possibilities are a false dichotomy. Instead, the reasons for God’s commands originate within God’s nature, so God’s commands are neither arbitrary nor God-independent.
Clearly, Craig does not endorse the view that God could have commanded otherwise. In responding to the traditional Euthyphro dilemma, Craig maintains God’s commands are always with reasons, and so are not arbitrary. But then, if God exists, God’s nature, from which God’s commands originate, is necessarily existent. So God cannot choose which commands to issue, as these are determined by the divine nature. Yet if the commands are uniquely determined by the divine nature, we are left wondering why God’s commands were required in the first place. What sort of explanatory role is left for God’s commands to play if, as Craig’s view apparently entails, our moral obligations can be determined from the nature of the Good? Perhaps we are left with a view similar to that endorsed by Ideal Observer Theory  — we have a duty to do whatever God would have commanded had God existed — which does not require God’s actual existence.
Here’s the upshot: if God’s commands are not required for moral obligation after all, and our duties follow merely from the nature of the Good, then Craig’s view is certainly no better than atheistic moral platonism. In fact, Craig’s view may be worse than atheistic moral platonism, because Craig’s equation of God with the Good is, at best, ad hoc. (Assuming one can make sense — which I cannot — of Craig’s grounding a property, e.g. goodness, in a particular, e.g. God. If grounding a property in a particular is incoherent, as I suspect, then Craig’s view is incoherent.)
The Divine Command Theorist may object that God’s commands have an explanatory role left to play, since obligations can only be obligations to someone. For example, suppose I promise Samantha I will walk her dog. By entering into a promise with Samantha, an obligation is created towards her, which can be overridden only by more compelling moral reasons; in other words, my obligation to follow through on my promise to Samantha is a categorical moral duty. Perhaps our moral duties, generally, are duties to someone, so, one might suppose, there needs to be a someone towards whom all moral duties are directed. Formally, we might construct this argument as follows:
- For any moral duty d, the duty to d is directed towards some person S.
- Therefore, there is some person, S, towards which all moral duties are directed.
- The only person towards which all moral duties could be directed is God.
- Therefore, all moral duties are directed towards God.
However, this argument is an example of a quantifier shift fallacy (in proceeding from 1 to 2) and so cannot work. Consider a structurally identical, but clearly false, argument:
1’. For any human x, x has a mother y.
2’. Therefore, there is a mother y for all humans.
3’. Only the Goddess could be a mother to all humans.
4’. Therefore, the Goddess is a mother to all humans.
That all humans have mothers does not imply there is a mother of all humans. (Certainly, Craig would want to deny the Godess’s existence!) Likewise, even if all moral duties were directed towards some person, it does not follow that there is one person towards which all moral duties are directed. Divine Command Theorists may object that only worthy authorities could create moral duties. If so, perhaps there is no quantifier shift fallacy after all, as there may be only one absolutely worthy authority. Nonetheless, recall my previous example. In my obligation to walk Samantha’s dog, I was obligated to Samantha and not to some other person. My obligation to Samantha did not require an authority exalted to any status higher than the two of us.
I have considered, and rejected, several reasons the Divine Command Theorist might propose as to why God’s commands are required for moral obligation, yet I might be missing something. Perhaps there is some further reason, as yet unknown to me, as to why the existence of moral duties requires God’s commands. However, if God’s commands are required for moral obligation, and do issue forth from God’s nature, then — as in the first fork in CED — we have an example of a divine act God cannot choose not to do, for if God’s commands were a necessary precondition for moral duty, God failing to issue commands would be inconsistent with God’s perfectly good nature. Yet God’s inability to choose otherwise is incompatible with God’s free agency; thus, we have further reason to suspect something has either gone awry with theistic ethics or with the classical theist’s conception of God.
 See, for example, Craig, W. (2007) “Theistic Critiques of Atheism”. In Michael Martin (ed), The Cambridge Companion to Atheism. Cambridge: Cambridge University Press, 83.
 For a brief of overview of Ideal Observer Theory, see Shafer-Landau, Russ, 2015. The Fundamentals of Ethics. Oxford: Oxford University Press, 303-306.
 Ethicist David Brink appears to be similarly befuddled by this view. See footnote 9 in his (2007). “The Autonomy of Ethics”. In Michael Martin (ed), The Cambridge Companion to Atheism. Cambridge: Cambridge University Press, 163-164. Erik Wielenberg expresses similar consternation: “I too have trouble grasping the claim that a mindlike Higher Power is identical to the property of goodness.” (2008) God and the Reach of Reason, Cambridge: Cambridge University Press, p 66.