A further reply to Mullins on divine simplicity (Updated)

UPDATE 8/25: David Mahfood replies to Mullins at Eclectic Orthodoxy.  I've got a couple of followup posts, here and here.

UPDATE 8/24: Brandon Watson and John DeRosa also respond to Mulllins.

UPDATE 8/21: Look out!  The Scotist Meme Squad has entered the fray.

At Theopolis, Ryan Mullins has now replied to those of us who had commented on his essay criticizing the doctrine of divine simplicity.  (The other commenters were Peter Leithart and Joe Lenow.)  What follows is a response to what he has to say in reply to my comments on the essay, specifically.

Mullins’ main argument

In his original essay, Mullins had in a discursive way explained what he takes the doctrine of divine simplicity to be and how it relates to the notions of divine freedom, grace, and God’s necessary existence.  Along the way he set out a sixteen step argument intended to show that divine simplicity cannot be reconciled with divine freedom.  In his latest essay, Mullins complains that “it is not clear that my dialogue partners have adequately attempted to engage with the argument,” so he repeats the argument again – this time by setting the sixteen steps off by themselves without the surrounding material, for clarity’s sake.  As he did in his original essay, Mullins challenges his critics to identify which premise, specifically, they reject.

Now, Mullins writes:

My dialogue partners have not been entirely forthcoming about which premises they reject.

For example, consider Edward Feser’s reply.  Feser comments on my argument from divine simplicity to the necessary existence of the world.  This is curious because none of the premises in my argument even mention the necessary existence of the world.  This leaves me to wonder if Feser read my original essay closely. 

End quote.  But in fact, as anyone can easily verify by going back and reading my original reply to Mullins, I did explicitly identify the premise I reject – namely premise (9), which says “All of God’s actions are identical to each other such that there is only one divine act.” 

I also fail to see why it is “curious” that I “comment[ed] on [Mullins’] argument from divine simplicity to the necessary existence of the world.”  Was I supposed to ignore what Mullins said about that topic in his original essay simply because the phrase “necessary existence of the world” doesn’t appear in any of the sixteen steps of his more formal argument?

I did, in any event, make it clear in my original essay why I presented my objection to Mullins in the way that I did.  In his sixteen-step argument, Mullins emphasizes God’s freedom to refrain from giving grace, specifically, but in fact there is nothing special about that example vis-à-vis the controversy over divine simplicity.  God’s freedom with respect to giving grace is just a special case of his freedom with respect to any of his acts, such as creating the world, causing miracles, sending prophets, etc.  What all these effects of divine action have in common is that they are contingent.  Hence if the doctrine of divine simplicity implies that these effects are necessary, then it seems the doctrine also implies that divine action is not really free after all.

The way to block this outcome is, again, to reject Mullins’ premise (9).  Now, what is wrong with that premise, as I explained in my original reply to Mullins, is that it ignores what Barry Miller calls the distinction between real properties and Cambridge properties (where, as I there hinted in a footnote, this echoes Aquinas’s distinction between real relations and logical relations).  Mullins writes:

To be fair to Feser, he does at least attempt to identify which premise in the argument that he rejects. Feser says,

Now, what the doctrine of divine simplicity claims—contrary to what Mullins supposes (in what he labels premise (9) of his argument)—is not that all of God’s properties are identical and thus are necessary as he is, but rather that all of his real properties are.

Feser then goes on to say that he rejects premise (9) because of a distinction between real properties and Cambridge properties.  According to Feser, God is identical to His real properties, but God is not identical to His Cambridge properties.

I find this reply from Feser curious for several reasons. 

End quote.  Mullins sure finds a lot of things “curious.”  Well, here’s something I find curious: that Mullins at first accuses me of having “not been entirely forthcoming” about which of his premises I reject, and then in the next breath conceding that in fact I identified premise (9) as the problem.  Go figure.

Anyway, Curious Ryan’s first complaint is that:

[I]t is curious because my premise (9) does not even mention the word property.  What my premise (9) actually says is, “All of God’s actions are identical to each other such that there is only one divine act.”  My actual premise (9) is something that proponents of divine simplicity explicitly endorse.  By this one act, God is said to will Himself and everything else that He has made.  Moreover, I intentionally avoided any mention of properties in the premises of my argument.  I avoided this because, as I explained in my original essay, proponents of divine simplicity explicitly deny that God has any properties, forms, immanent universals, or tropes.  As the proponent of divine simplicity, Katherin Rogers, makes clear, the simple God does not have any properties.  Instead, God is simply act.

End quote.  There are several things wrong with this.  The first is that it simply isn’t true that all proponents of divine simplicity would say that God does not have properties.  That way of talking makes it sound as if God is some sort of bare particular, and that is certainly not what the doctrine of divine simplicity claims.  What proponents of divine simplicity would all say is that God does not have any real properties that are distinct from the divine substance.

In any event, the word “property” is used by different philosophers in different ways, so before one makes sweeping claims about what “proponents of divine simplicity explicitly deny” one had better get clear on exactly how this or that proponent uses the term.  Now, as Mullins surely knows, the word “property” is commonly used in a very loose way by contemporary analytic philosophers, to mean just any old thing you might predicate of something.  To be sure, this is definitely not my own preferred usage, since the term has a much narrower technical sense in Aristotelian-Thomistic metaphysics.  But it is, again, a common usage among analytic philosophers.  Since analytic philosophers of religion are Mullins’ own main audience, since Miller (who I was citing) uses the term in this way, and since I had a word limit and didn’t have space in my essay to get into an explanation of the various uses of the term “property,” I decided for expository purposes to acquiesce to the common analytic philosopher’s usage. 

Now, Mullins surely realizes that that is how I intended to use the word “property” in my essay.  But given that broad sense of the term, it is quite obviously false to say that defenders of divine simplicity deny that God has “properties.”  On the contrary, defenders of divine simplicity typically hold that God has lots of “properties” in that broad sense of the term – omniscience, omnipotence, perfect goodness, and so forth.  To be sure, defenders of the doctrine also say that God’s omnipotence is his omniscience, which is his perfect goodness, and so on.  So they would say that God’s “properties,” in this broad sense of the term, are identical.  But to say that is very different from saying that he doesn’t have “properties” in the broad sense.

Now, in this broad sense of the term “properties,” God’s actions would also be “properties” of God.  Hence the distinction between real properties and Cambridge properties is obviously relevant to evaluating Mullins’ premise (9), because we need to know whether the divine actions Mullins focuses on (such as giving grace and creating the world) are real properties or Cambridge properties – again, in the broad sense of “properties” used by analytic philosophers.  And again, Mullins surely has to realize that this is what I meant.  Mullins’ book on the divine nature is in a series devoted to analytic theology, and it cites Miller in its bibliography, so the usage cannot be unfamiliar to him.  So, Mullins’ heavy going about my reference to “properties” seems to me to amount to mere quibbling and kicking up dust.  All very curious, as Mullins would say.

Anyway, here’s the payoff.  When we get clear on the terminological issues, it is manifestly false to claim, as Mullins does, that his premise (9) “is something that proponents of divine simplicity explicitly endorse.”  For, to use Miller’s language, defenders of divine simplicity claim only that God’s real properties are identical to each other, not that his Cambridge properties are.  Hence any divine action that amounts to a Cambridge property is not one that the defender of divine simplicity will identify with the divine nature.  And that is exactly what is going on with properties like “being creator of the world” and “being the source of grace.”  They are Cambridge properties.  So, the problem with Mullins’ premise (9) is that it is too sweeping.  Once one restricts the premise to God’s real properties, the revised premise will be true, but then the rest of Mullins’ argument will no longer go through.  In particular, it will no longer follow that God’s act of giving grace is identical to God and thus as necessary as he is.  And thus it will not follow that that act is not free.

Mullins tells us that a second reason he finds my reply “curious” is that I seem, in his view, to be confusing his position with that of Thomas Morris, against whom I deployed Miller’s distinction between real properties and Cambridge properties in my book Five Proofs of the Existence of God.  (More precisely, Miller himself deploys this distinction against Morris – I was just citing Miller’s response to Morris in my book, rather than presenting it as my own.)  Mullins says his argument is very different from Morris’s, since Morris’s focus is on the claim that divine simplicity would make God’s creation of the world necessary.  But I have already explained why this is not a difference that makes a difference.  What is doing the work in Mullins’ argument is the claim that divine simplicity makes God’s actions, in general, necessary – the same thing that is doing the work in Morris’s argument.  The fact that Mullins focuses on the specific example of God acting to impart grace, whereas Morris focuses on the example of God acting to create the world, is irrelevant.  Hence my objection applies to Mullins’ argument no less than to Morris’s.

But Mullins’ third reason for finding my reply “curious,” he tells us, is that he does not think it really applies even to Morris’s position.  The reason, he says, is that “divine actions are intrinsic to God,” and thus cannot be Cambridge properties.  Hence my appeal to the distinction between real properties and Cambridge properties is a “category mistake” and amounts to “nothing but hand-waving.”

Well, if Mullins isn’t here waving his own hands, that’s only because he’s too busy missing the point.  To see what is wrong with his response, go back to the examples I used in my original essay to introduce the distinction between real properties and Cambridge properties.  When Socrates grows a beard, that involves the acquisition by him of a real property.  But when he becomes shorter than Plato, not because of any change he undergoes, but rather because Plato has grown taller, that involves the acquisition by Socrates of a Cambridge property.  It is, in this case, Plato rather than Socrates who acquires a real property.

Now, it is possible for something to have a Cambridge property by virtue of also having a certain real property.  To modify my example, suppose that Socrates had been Plato’s father, so that it is by virtue of his having been begotten by Socrates and inheriting Socrates’ genes that Plato eventually grew to be taller than Socrates.  Then there is a sense in which you could say that Socrates’ action, the action of begetting Plato, caused Socrates (later on) to become shorter than Plato.  Now, there is obviously a sense in which Socrates’ action of begetting is intrinsic to Socrates.  Something has to happen in him in order for the begetting to occur.  And so, something had to happen in him in order for Plato later on to grow taller, which resulted in Socrates being shorter than Plato.  But does it follow that Socrates’ becoming shorter than Plato is not a Cambridge property after all but a real one?  Of course not. 

Now in the same way, it is certainly true to say that there is something intrinsic to God himself that makes it the case that the world is created.  No defender of divine simplicity denies that.  But it simply doesn’t follow that God’s acting to create the world is not after all a Cambridge property, any more than it follows that Socrates’ growing shorter than Plato is not after all a Cambridge property.

You might characterize the situation in my revised example as Socrates acting in a way that (eventually) resulted in his becoming shorter than Plato.  But though the acting is something intrinsic to Socrates, that the acting resulted in his becoming shorter than Plato is not intrinsic to him.  Similarly, you might characterize creation as God acting in a way that resulted in the world’s existing.  But though the acting is something intrinsic to God, that the acting resulted in the world’s existing is not intrinsic to him.

You may or may not find this plausible.  But here’s the thing.  The unwary reader of Mullins’ response is likely to get the impression that my reply to him was some eccentric concoction of my own, cobbled together to rebut his novel and clever objection.  But my response is neither my own concoction nor eccentric, and Mullins’ objection is neither novel nor particularly clever.  In fact it is pretty old hat.  As I have already said, I am simply making a point made by Barry Miller, but Miller was in turn only putting a contemporary gloss on a line of argument that goes back at least to Aquinas, and that has been repeated by many Thomists since.  And all of them were responding to variations on the sort of objection Mullins gives.

The way Aquinas puts the point is, again, in terms of the distinction between a real relation and a logical relation (sometimes called a relation of reason).  A stock example is the relationship between perceiver and perceived.  If I perceive you, I bear a real relation to you.  But if you are perceived by me, you bear a logical rather than real relation to me.  Miller’s way of putting this is to say that by virtue of perceiving you, I have a real property, whereas by virtue of being perceived by me, you have a merely Cambridge property. 

Now, Aquinas’s view is that by virtue of being created by God, the world bears a real relation to him, whereas by virtue of creating the world, God bears a merely logical rather than real relation to it.  Miller’s way of putting this is to say that the world’s being created by God is a real property of the world, whereas God’s creating the world is a Cambridge property of God.  (When replying to Mullins, I decided to use Miller’s formulation because I supposed it would be less likely to be misunderstood by readers more familiar with analytic than with Thomistic jargon.  Silly me.)  One wonders whether Mullins is even familiar with this traditional Thomistic position.  If he isn’t, that would explain the, er, curious circumstance that Mullins at one point actually tries to employ Aquinas against my objection – an objection I borrow, by way of Miller, from Aquinas.

Anyway, the point is this.  The Aquinas-Miller position is not some obscure or incidental aspect of the debate over divine simplicity.  It is a standard part of it, certainly among Thomists, and has been for centuries.  Mullins does nothing to refute this position.  All he does is make assertions of precisely the kind the Aquinas-Miller position claims already to have answered.  So, while Mullins seems to think he has made some devastating new objection against divine simplicity, in fact he hasn’t advanced the discussion one inch beyond where Aquinas left it in the thirteenth century.  Maybe the Aquinas-Miller position is wrong (though obviously I don’t think it is), but if so, Mullins certainly has done nothing to show that it is.

Analogy and dire consequences

Mullins has some equally anticlimactic things to say in response to my remarks about analogy.  He writes:

I am not entirely sure what the doctrine of analogy does for divine simplicity with regards to my argument.  Can analogical predication somehow remove a contradiction like “God is free and God is not free.”  No.  Classical theists are quite clear that God cannot be free if His actions are performed of absolute necessity.  No amount of analogical predication can make God’s free actions consistent with those actions being performed of absolute necessity.  Can analogical predication somehow make God’s contingent and necessary actions identical?  Of course not.  No amount of analogical predication can make necessity and contingency mean something completely different so as to remove the contradictions that I have pointed out.  I think what Feser is really working with is equivocation.

End quote.  Well, this is – one more time, everyone, all together – all very curious.  First, there can be no mystery about why I brought up analogy in my earlier essay, because I explicitly said why.  The reason was that in his own earlier essay, Mullins had made some glib and sweeping remarks to the effect that proponents of divine simplicity resort to “cheap” appeals to mystery, and in particular to “mysterious language.”  I was pointing out that this is an unjust characterization, because what they actually appeal to is the notion of the analogical use of terms.  This notion is neither particularly mysterious nor cheap.  It is a familiar part of both ordinary and scientific discourse, is motivated independently of any concerns about divine simplicity, and has been worked out in theoretical detail.

Second, and as should go without saying, Mullins is attacking some pretty crude straw men in this paragraph.  No defender of divine simplicity holds that God is both free and unfree, or that something can be both contingent and necessary in the same respect.  So, naturally, no defender of divine simplicity appeals to analogy in order to defend such absurd claims.

Mullins, once again, has nothing new or of interest to say here, and the other remarks he makes about analogy are just bare assertions that defenders of divine simplicity have heard and answered many times.

Finally, Mullins comments on the classical theist view, which I endorsed in my earlier essay, that to deny divine simplicity entails denying the ultimacy and uniqueness of God, and thus entails atheism.  He characterizes this as a “slippery slope” argument that “assume[s] an elaborate set of metaphysical theories that a Christian does not have to affirm in order to articulate a biblical doctrine of God.”  He then goes on to mention several of the relevant metaphysical claims, and to tell us that he finds each of them “implausible.”  He suggests that classical theism and atheism are not the only options, and he is dismissive of the classical theist’s claim that the further options reduce God to something merely superhuman but still creaturely.

Of course, these are all just more bare assertions, and question-begging ones at that.  For example, the classical theist would not agree that the relevant metaphysical considerations need not be affirmed in order to defend the biblical conception of God.  But what is really curious (whoops, said it again) is that Mullins ignores the main reasons I brought up the dire consequences in question in the first place.  In his initial essay, Mullins had given the impression that the doctrine of divine simplicity is merely the bizarre obsession of theologians too enamored of some obscure philosophical ideas, and irrelevant to anything of specifically Christian or biblical concern.  This is a common rhetorical ploy among critics of the doctrine.  And I was pointing out that in fact, what motivates defenders of the doctrine is a concern to uphold the uniqueness and ultimacy of God – both of which are, needless to say, pretty central to the Christian and biblical conception of God.  I was pointing out that in fact, defenders of divine simplicity would appeal to the ultimacy and uniqueness of God precisely as the basis of positive arguments for divine simplicity.

So, my complaint was that Mullins’ original essay failed to convey what is actually at stake in the debate over divine simplicity, and thus failed to grapple with the main arguments for the doctrine.  And that failure is recapitulated in his latest essay, insofar as all he has to tell us is that he personally finds the arguments in question “implausible” and prefers other views to classical theism.  Well, we already knew that.  What we need is an actual response to the arguments.  Once again, Mullins has failed to advance the debate over divine simplicity even an inch.