One of the main reasons we can have for believing in God is that, If God exists, we have a good explanation for the existence of an orderly and relatively life-friendly universe such as we find ourselves in. But this is only true if God’s will is not completely free.
I argued that the full argument conflated a distinction important for explanation, and this new post is his response to it. In it he concedes the importance of the distinction, but holds that there is a high price to be paid for it as a response in this particular case. In particular, Jason argues that there's a trilemma here:
Thus the theist, as I see things, has three main options: First, they could accept the externalist account of God’s actions, which would presumably preserve God’s freedom, at the cost of committing them to hold that God could not explain why the world is the way it is in any substantive sense. Second, they could reject Divine Simplicity and hold that God is both free and explanatory, but in doing so they would be giving up a long-held orthodox belief. Thirdly, they could maintain that God is explanatory but unfree in a libertarian sense, because God’s will is the same in every possible world—a position which, as it involves the idea that God could not create a world that is disordered and/or hostile to life, is also unorthodox by traditional standards.
If this were right it would be a pretty serious pinch: either God is not cause (in a robust explanatory sense) or God is not simple, or God is not free. I don't think it's right, and I think the part of the trilemma that crumbles is the middle horn.
Of divine simplicity Jason says:
Traditionally, it has been thought that God’s nature or “essence” is identical to God’s existence—as are all of God’s attributes and/or properties. Thus, God just is his own Omniscience, Omnipotence, Goodness, Wisdom, Mercy, etc. (This doctrine is known as “Divine Simplicity”)
That this is the traditional view of divine simplicity is a common view among philosophers of religion -- I have seen quite a few put it forward. I am very sure, however, that this view does not in any significant way predate the early modern period, and largely only gets called the traditional view because of a misleading translation.
Simplicity is fundamentally not a doctrine about identity; it is not the claim that God "just is" His attributes. Rather, it is the claim that God is not composite, that He is not composed of anything. Aquinas at one point, I forget exactly where, gives us a list of kinds of composition that are not in God, and it is clear that he regards it as exhaustive:
1) part and whole
2) matter and form
3) nature and supposit
4) essence and esse
5) genus and difference
6) subject and accident
All of these compositions in a true and proper sense of the term because on Aquinas's account of composition you have composition only when two things are related in such a way that one is potential to another. Part is potential to whole, matter to form, nature to supposit, essence to actual being, genus to difference, subject to accident. Thus since nothing in God is potential to anything else in God (for various reasons I won't go into here), God is not complex; He is simplex. What is very noticeable about this account is that it doesn't rule out distinctions as long as they don't introduce potential-actual distinctions.
And while not everyone has exactly the same view, it's nonetheless clear in, say, the medieval period that (1) everyone accepts divine simplicity; (2) no one holds that this eliminates every distinction one might make; and (3) there are arguments all over the place about whether, given divine simplicity, one can apply this or that distinction to divine matters. Almost everything everyone says about distinctions in theological matters in the medieval period becomes completely unintelligible if divine simplicity is taken to mean that everything in God is identical to everything else, because this makes distinctions absolutely impossible, by definition.
It's an interesting question why people think the identity thesis is the traditional view of divine simplicity. I think there are two major contributing factors. The first is that the account of divine simplicity which most philosophers have come across in a form where it does clear philosophical work is the Cartesian account of divine simplicity. Now, if there's any account of divine simplicity that comes pretty close to the identity thesis, it's Descartes's version of it. Despite attempts by, say, Dan Kaufmann, to argue that Descartes introduced no serious innovations into the doctrine, Descartes's account of divine simplicity completely baffled almost everyone who came into contact with it at the time and almost every Cartesian after him struggles to rethink divine simplicity in Cartesian terms (and keeps running into new problems). (Shifts in the use and content of the doctrine of divine simplicity in the major Cartesians, by the way, would be a very good dissertation topic; they certainly are there but haven't been looked at very closely.) But since pretty much everyone trained in philosophy today spends a fair amount of time on Descartes, and certainly more time on Descartes than on any other major thinker who takes divine simplicity seriously, philosophers have a bad habit of assuming something like Descartes's view is the traditional view, and that the sort of philosophical ramifications it has will always be those that it has in Descartes's philosophy. This is not so.
The second is that one of Aquinas's key doctrines is often called "the identity of essence and existence". Reading 'identity' here as identity in the sense that essence and existence are identical, so that the one 'just is' the other (without any attenuating qualification like, 'in some way'), is, I think, based on an anachronistic reading of a not-particularly-great translation. As I've talked about this before (here and here) I won't dwell here on the reasons for my view on this point. Suffice it to say that it makes sense as a description of the Thomistic doctrine (if essence and existence are not composed in God, then they are in some way the same, idem), but 'identity' in this context does not mean what we would usually take it to mean. And, of course, there are other traditional accounts of divine simplicity besides Aquinas's.
In any case, the point is that divine simplicity is merely divine noncompositeness; this does not require the identity thesis, and thus one can easily have divine simplicity while rejecting the account of it Jason is assuming here. So God can be free and simple and explanatory. I'm not convinced that even the defender of the identity thesis is in quite such hard straits as Jason suggests; but since this post is long enough, I will simply leave it, and perhaps some day develop this suspicion a bit more fully in a future post.