The Euthyphro dilemma goes like this: God commands us to do what is good. But is something good simply because God commands it, or does He command it because it is already good? If we take the first option, then it seems we are committed to the possibility that God could make it good for us to torture babies just for fun, simply by commanding it. If we take the second option, then it seems we are committed to saying that there is a standard of goodness independent of God, to which He refers us when He commands. Neither option seems a good one from the point of view of theism. The first makes morality arbitrary, and the claim that God is good completely trivial. The second conflicts with the core theistic claims that God is the ultimate cause of all things, and in particular the source of all goodness. So, we have a problem, right?So far so good, unless you don't like the last sentence, but it's what you might expect a theist to say. He goes on to say:
Actually, we don’t, because the dilemma is a false one
The actual situation, ..., is this. What is good or bad for us is determined by the ends set for us by our nature, and ..., that means that there are certain things that are good or bad for us absolutely, which even God could not change (since God’s power does not extend to doing what is self-contradictory). Now God, given the perfection of His intellect, can in principle only ever command in accordance with reason, and thus God could never command us to do what is bad for us. Hence the first horn of the Euthyphro dilemma is ruled out: God can never command us to torture babies for fun, because torturing babies for fun is the sort of thing that, given our nature, can never in principle be good for us. But the essences that determine the ends of things – our ends, and for that matter the end of reason too as inherently directed toward the true and the good – do not exist independently of God. Rather, ..., they pre-exist in the divine intellect as the ideas or archetypes by reference to which God creates. Hence the second horn of the Euthyphro dilemma is also ruled out.I wonder what "good" means here. In what sense could torturing babies for fun never be good for us? I don't think Feser means it could never be fun, although that might be true. I think he means that people who take pleasure in such things thereby stray from the proper path for human beings. But what makes this path proper? The answer to that seems to have to do with the essences that pre-exist in the divine intellect. But why should anyone care about them? This includes God. Why does he create by reference to them? Does he have any choice about doing so? Or is the idea that he has a choice but realizes that doing so is the best choice to make and therefore makes that choice, i.e. creates by reference to these archetypes? Or does he choose (or create) the archetypes themselves? In which case we can ask whether he chooses/creates them because they are good, or vice versa. The problem looks to have been pushed inside the divine intellect, not solved.
Feser goes on:
Keep in mind also that, ..., the metaphysics underlying the arguments for classical theism lead to the conclusion that God is not one good thing among others but rather Goodness Itself. Given divine simplicity, that means that what we think of as the distinctive goodness of a human being, the distinctive goodness of a tree, the distinctive goodness of a fish, and so on – each associated with a distinct essence – all exist in an undifferentiated way in the Goodness that is God.If God is goodness (I assume this is what "Goodness Itself" means) then, as I think Leibniz worried, it doesn't seem to make sense to praise or thank him for being good. Doing so would be like praising chocolate for being chocolatey. That does make a kind of sense, I think: it is an expression of love. But I don't think that's what Feser wants. And if God is goodness, then 'good' means 'godly,' and God's commands are good by definition (because they are commands of or pertaining to God, i.e. goodness). This might be the best way to go (other than dropping God-talk completely). To suggest that God(-as-goodness) might have commanded something bad makes no sense. It would be like saying that red might have been blue. Of course we might have called torturing babies for fun 'good,' but then 'good' would simply have meant what we mean by 'bad.'
So I think the best options are either atheism, in which case we are stuck with the question of what we ought to do, and defining God as goodness, in which case we should probably drop all talk of "the divine intellect" (because how could goodness have an intellect?) and, while God's commands can never be bad, we still have the question of what we ought to do. We should do what is good, or godly, but what is that? God knows.