In certain Hindu circles, one of the divine attributes is lila, which is really hard to translate, but is often associated with playfulness; not that it has to be play in our sense, but that divine actions are purposeful but not bound to a purpose, in the way that (for instance) a young child playing is purposeful in action without having any real restriction to a purpose, able to shift purposes at will. I was reminded of this when reading James Reilly's paper, Two challenges for 'no norms' theism (PDF), which argues that someone who holds "that God is exempt from moral and rational norms" (which he calls 'no norms' theism or NNT) faces two kinds of problems. First, they cannot make use of what Reilly calls inductive arguments for the existence of God, which have a structure in which certain evidences are held to be more likely if God exists than if God does not exist. Second, that if God is not subject to norms, then God is effectively like Descartes's deceiver, and we are left with skepticism about our faculties. Reilly's paper is quite good, but I'm not convinced either challenge is particularly challenging.
I suppose I would be what Reilly calls a 'no norms' theist, although for what might be called ethical rather than theological reasons -- that is, I don't think there's any viable candidate for a serious account of obligations on which it would make much sense to say that God has obligations. If we have a positivistic account of obligations, there is no legislator or obliger who outranks God, so God doesn't have obligations. If we have a sentimentalist account of obligations, obligations are relative to natural sentiments, but God is impassible, and therefore has no obligations; and a sentimentalist account implies that it would be at least extremely difficult to work out what even nonhuman alien obligations might be, much less divine obligations. If we have a rationalist account of obligations, the two best accounts are natural law theory and Kantianism, and God doesn't have the kind of rational nature that is required to generate obligations on either theory. I suppose you could have a rationalist theory, maybe even a generalized natural law theory, on which, by a sort of deontic necessitation, God has an obligation to be God, and that is the entire list of God's obligations, but I don't see any prospect, in any theory of obligation worth taking seriously, of getting beyond that. Malebranche has a view in which God (the Father) loves Order, which is divine Reason and thus God (the Son), and therefore God is obligated by Order, which is, I think, the most serious attempt actually to make sense of what divine obligations would be; but the list of inconveniences and problems attaching to Malebranche's account of how God relates to divine Reason is quite considerable. It's pointless to say that God has obligations if we don't have any good account of obligations under which it makes much sense to say that God has obligations. Of course, there are many norms even in our own case that are not obligations -- norms of aesthetics, norms of etiquette, norms of technique, an entire forest of them. But many of the concerns with obligations do arise for at least many of the other norms.
I have no particular commitment to what Reilly is calling inductive arguments, but I think his account fails to take into consideration the flexibility of such arguments. For instance, if we take an analogous argument, an inductive argument to the existence of a particle, based on the idea of the effects that a particle would likely have, we are not arguing to the bare existence of the particle somehow. If I say, "This effect here in Texas is the kind of thing that would be caused by such-and-such particle," I don't mean that it is the kind of thing that would be caused by that particle if operating outside the Andromeda galaxy; I am saying that this is the kind of effect you get from such a particle doing particular things in particular ways (e.g,. hitting something in Texas). Likewise, inductive arguments for God's existence are not arguments to the bare existence of God acting anyway and anyhow; they are arguments to the existence of God as doing something relevant to the evidence. The inductivist is not committed to saying anything about what other things God might legitimately do. Norms, on the other hand, do cover other things that can be done; there are arguably no norms that cover one and only one action. Thus the inductive arguments don't really require appeal to norms; they just require appeal to something about the intelligible structure of particular actions, which could indeed involve norms, but might not (as there is much more to the intelligibility of any action than its position in a field of possible action governed by a norm).
Likewise, Reilly seems to have a rather expansive view of what involves norms; he takes, for instance, all appeals to goodness to be appeals to norms. This is certainly not true even in our own case; there are lots of goods that we do, not because of any norm, but just because they are good, where we could perfectly well do something else. If you identify something as good in a certain way, you are identifying it as a reason for action; but not all reasons for an action are norms for it. Reilly argues,
If God is not
bound by the norms of goodness, then what reason do we have for supposing that he
would favour a world of beauty and order over a world of chaotic ugliness? Why should
we expect the existence of conscious life, rather than the cold sterility of a dead cosmos?
The answer cannot be that it is better for such things as beauty, order, and conscious life
to exist; once God is exempt from the norms of goodness, words like ‘better’ and ‘worse’
lose all relevant meaning. (p. 2)
But this is simply false; 'better' and 'worse' are not always or even usually defined in reference to norms. They do require a reference point so that you can get a comparison; but nothing requires that this reference point be functioning as a norm. What is more, a world of beauty and order is good, a world of chaotic ugliness is (thus far) not; you don't even need a reference point for that, because one description identifies a good and the other doesn't, so (thus far) there is a reason for one and not the other. Whether there is a norm governing God's favoring one or the other is irrelevant; by simply stating the case, you've given one reason one might favor the one over the other. Even if God were subject to norms, so that normatively he had to favor one over the other, whether a norm were relevant would depend on knowing all the reasons; we don't need to do that to identify that there is a plausible reason. I myself don't think you can argue this way -- it violates the principle of remotion and all good sense to pretend to have the kind of omniscience that can assess the reasons available to omniscience in such a definitive way -- but the inductive arguments only require reasons, not norms. They don't even require the norm that God should always follow reasons that meet certain conditions; they just require that they are reasons that God can consider.
If someone is not bound by norms of goodness, what reason do we have for supposing he would favor a world of beauty and order? The same reason that we would have if he was bound by norms of goodness. If I say, "John might have preferred that painting because of its beauty", I am not saying that John's preference was bound by a norm of beauty, I'm saying the painting meets a standard of beauty that is a reason John might have incorporated into his preference. I'm not necessarily saying that it would have been bad or stupid for John to prefer another painting; I'm not necessarily saying that John had to prefer paintings specifically based on their beauty; I'm not saying there was some rule that John was bound by to prefer beautiful paintings; I'm not even saying that in preferring the painting because of its beauty he was doing so in one way (e.g., because he would like it) rather than another way (e.g., because someone he liked would like it), which I would need to in order to identify what norms, if any, might be relevant. I'm very definitely not saying that if he had preferred another painting, he would have violated any norm, because norms need not be involved at all.
The same goes for the second challenge. Descartes, to block the evil deceiver, didn't need to identify a norm to which God was subject. The principle that blocks the evil genius is not, "God should not be a deceiver", but rather "God is not a deceiver". The Cartesian principle doesn't work -- indeed, it is hard to see how it could work -- as a norm; it does so as a fact. (Since it is a kind of necessary fact, one might hold that it also implies, by deontic necessitation, a norm of not being a deceiver; but, first of all, it's not this that lets it play the role it does, and, second, lots of people don't like deontic systems that have deontic necessitation.)
Reilly seems to have the notion that norms are the only things that make sense of intelligent action at all; it plays a role in his rejection of 'divine love' responses, for instance. But this is simply untrue. Even in our own case, many of our intelligible reasons for doing things are not norms; indeed, arguably, even in many cases in which they are consistent with norms, the norm plays no role in understanding the action itself. If I raise a beer in honor of a comrade, there's no norm that says I have to; I don't do it to fulfill a norm; I'm not being guided by a norm in doing it. There are norms that are maybe relevant to someone else's assessment of what I am doing, but my consistency with those likely has no role to play in my actually doing it. It's just a thing I can do, and there's a reason one might do it, and even though I could perfectly well not do it or do something else instead, I do it for that reason. That makes perfect sense, and I haven't appealed to anything normative at all. Even in our own case, even when we can judge actions by norms, we can often explain them without any appeal to norms at all.
I suspect too that Reilly is confusing 'being guided by a norm or standard in preferring something' and 'preferring something because it has a quality assessed by reference to a norm or standard'. The NNT is not claiming that God's effects are 'no norms', just that God's actions are 'no norms'. God could very well, non-normatively, pick effects to be consistent with certain norms. A painter might not be bound by a standard for the act of painting to produce a certain painting, but he could perfectly will choose to produce a painting in light of a particular standard for the painting that is painted. The professional painter is not in any way bound by the rules of classical style in making a painting, for instance, but he can certainly choose to paint a painting that itself is governed by the rules of classical style. I am not bound by the rules of chess to play chess; in lots of circumstances where other obligations don't come in, I'm not even really required to consider myself bound by the rules of chess in playing chess -- I can make up a game of fairy chess as I go. But this doesn't prevent me from choosing to play a game of chess according to the rules of chess, and I could indeed do so simply because I like a game of chess played strictly by the rules of chess. My choice to play a rule-bound game of chess is not because my choice is rule-bound by the rules of chess. Norms for acts are not norms for intended objects of acts; norms for objects are not norms for acts. NNT is not claiming that God can't choose to make creatures that are bound by norms, but only that God is not bound by any. Indeed, it is entirely consistent with NNT to say that there are norms for any creatures God might create but not for God's creative action itself.
Someone who accepts NNT can perfectly well say that God likes norms for creation without having to say that God's liking is because of a norm on divine likings. Why, for instance, does God create James Reilly? Does he need to be guided by a norm in creating James Reilly? Maybe God creates James Reilly because God likes James Reilly; or maybe God likes something else, and James Reilly makes a sort of nice accompaniment or aperitif or heightening contrast to it, so why not make James Reilly, too; or maybe something of both. There seems nothing wrong, at least from the perspective of what 'no norms' theism requires, with saying that God does some things just because God likes them, just out of divine lila or superabundant divine goodness.
So in short, the basic position, NNT, seems not really to be affected by the challenges; someone who accepts NNT is only really going to have trouble with them if they for some reason also accept some of Reilly's more controvertible views about norms.