This doctrine is absolutely central to the classical theistic tradition, and has been defended by thinkers as diverse as St. Athanasius, St. Augustine, St. Anselm, St. Thomas Aquinas, Maimonides, Avicenna, and Averroes, to name just a few. It is affirmed in such councils of the Roman Catholic Church as the Fourth Lateran Council (1215) and Vatican I (1869-70) – which means that it is de fide, an absolutely binding, infallible, irreformable teaching of the Church, denial of which amounts to heresy. Divine simplicity is generally understood to follow from the Aristotelian-Thomistic doctrine of God as pure actuality. For something composed of parts presupposes the combination of those parts and thus a reduction of potentiality to actuality; and a purely actual being has no potentiality to actualize.
Nevertheless, contemporary philosophers and theologians are often critical of the doctrine of divine simplicity, and a reader has asked me to comment on this critique of the doctrine by William Lane Craig. (Craig is a Protestant, and thus is not troubled by the centrality divine simplicity has in Catholic doctrine.)
Before commenting, let me say that I have the greatest respect and admiration for Craig, who is, needless to say, one of the great Christian apologists of the age, a brilliant philosopher, and a fine scholar. His work on the history of the cosmological argument played a role in my own conversion, since it helped lead me to see how very badly most critics of the argument misunderstand it. (Craig and I have met only once, over a decade ago when he was visiting the UC Santa Barbara campus and kindly presented a guest lecture on the kalam cosmological argument to the Introduction to Philosophy class I was then teaching. I was still an atheist in those days, though the intellectual barriers to theism were just starting to crumble thanks in no small part to him.)
In the short piece linked to above, Craig offers three criticisms of the doctrine of divine simplicity. First, in response to the notion that the divine attributes are not distinct from one another, Craig says:
Existence is part of God's nature. But existence is not the same property as, say, omnipotence, for plenty of things have existence but not omnipotence. It remains very obscure, therefore, how God's nature or essence can be simple and all His properties identical.
Second, in response to the claim that God’s nature is not distinct from His existence, Craig says:
In a sense, God has no essence on this view, rather He just is the pure act of being unconstrained by any essence. He is, as Thomas says, the pure act of being subsisting. The problem is, this doctrine is just unintelligible.
Third, Craig says that the doctrine of divine simplicity entails that “God has no properties distinct from His nature,” and objects that:
[This claim] runs into the severe problem that God does seem to have accidental properties in addition to His essential ones. For example, in the actual world, He knows, loves, and wills certain things which He would not know, will, or love had He decided to create a different universe or no universe at all. On the doctrine of divine simplicity God is absolutely similar in all possible worlds; but then it becomes inexplicable why those worlds vary if in every one God knows, loves, and wills the same things.
Let me take the first two objections first, and begin by making two observations. First, note that both objections more or less amount to little more than the assertion that we can’t make sense of the doctrine of divine simplicity – that it is “very obscure” or “unintelligible.” Little or no actual argument is given for this claim, at least not in Craig’s brief piece. (The bit about how existence and omnipotence are different seems, without additional argumentation, merely to beg the question.) But the fact that a great many major philosophers and theologians have regarded divine simplicity as intelligible should at least give us pause; surely we need more than the mere assertion of unintelligibility, or an expression of one’s personal difficulty in making sense of the doctrine, if we are to be justified in rejecting it.
A second, and by no means unrelated preliminary point is that Craig makes no reference here to the famous Thomistic doctrine of analogy, which from a Thomistic point of view is crucial to properly understanding divine simplicity. To illustrate the idea of analogy, consider the word “see.” When I say that I see a tree outside my window and that I see the details of an insect’s eye through a microscope, I am using “see” in a univocal way, in the same sense in both cases. When I say that Rome is the Holy See, I am now using “see” in an equivocal way, that is, in an entirely different and unrelated sense. But when I say that I can see the truth of the Pythagorean Theorem, I am now using the term in neither a univocal nor an equivocal sense, but rather in an analogical way. That is to say, what one does when he “sees” the truth of the theorem is not the same as what he does when he sees a tree, but it is not completely different either. There is an analogy between the sort of thing we do with our eyes and the sort of thing we do with our intellects that makes it appropriate to describe both as kinds of “seeing.”
Now the Thomistic doctrine of analogy tells us that when we correctly predicate some attribute of God, we are using the relevant terms, not in a univocal way, but in an analogous way. That is to say, when we say for example that God has power, we don’t mean that He has power in exactly the sense we do, though we also don’t mean that His power is completely unlike what we call power in us. Rather, when we call God powerful we are saying that there is in God something analogous to power in us. Or take a more clearly metaphysically loaded term like “being,” as used in a sentence like “God has being.” Accidents and substances can both be said to have being, but accidents lack the independent existence that substances have; material things and angels can both be said to have being, but material things are composites of matter and form while angels are forms without matter; created things and God both have being, but in created things essence and existence are distinct and in God they are not; and so forth. The being of an accident is analogous to that of a substance, that of a material thing is analogous to that of an angel, and that of a created thing is analogous to that of God; that is to say, it is neither completely identical nor absolutely incomparable.
When we bring the concept of analogy to bear on the doctrine of divine simplicity, we can see what is wrong with Craig’s bare assertion that the doctrine is unintelligible. For this assertion has whatever plausibility it has, I would suggest, only if we think of God as having an essence, as existing, and as having power, knowledge, etc. in the same or univocal sense in which we and other creatures have these things. For what we call power in us is clearly different from what we call knowledge in us; our essences are different from our “acts of existing” (to use the Thomistic jargon); and so forth. So to say that knowledge (in that sense) is identical to power (in that sense), etc. does seem unintelligible. But that is simply the wrong way to understand the doctrine of divine simplicity. Properly understood, the doctrine does not say that power, knowledge, goodness, essence, existence, etc., as they exist in us, are identical. Rather, it says that there is in God something that is analogous to power, something analogous to knowledge, something analogous to goodness, etc., and that these “somethings” all turn out to be one and the same thing. “Power,” “knowledge,” “goodness,” etc. are merely different, analogously used descriptions we use in order to refer to what is in God one and the same reality, just as (to borrow Frege’s famous example) the expressions “the morning star” and “the evening star” differ in sense while referring to one and the same thing (the planet Venus).
Precisely because God is simple, though, there is in Aquinas’s view a sense in which we cannot strictly know His essence. For we know things in the strict sense by being able to define them in terms of genus and specific difference, and since God is absolutely simple, there is in Him no distinction between genus and difference, and thus no way to define Him (again, in this technical sense of “define”). God is not merely a unique member of some general class of things; the fact that there is one God is not some metaphysical accident, but an absolute metaphysical and conceptual necessity. But precisely for that reason, precisely because He is so radically unlike anything in the created order, we simply cannot expect to comprehend Him with anything close to the sort of clarity with which we can understand the denizens of that order.
Now this is the God to which the arguments of classical natural theology – by which I mean arguments falling into the broad metaphysical tradition inclusive of Platonism, Aristotelianism, Augustinianism and Thomism – inevitably lead. For such arguments all tend to the conclusion that the ultimate explanation of the world can only possibly lie in what is pure actuality, or being itself, or the One, or that in which essence and existence are identical; and all such concepts entail the doctrine of divine simplicity. What all this leaves us with vis-à-vis Craig’s first two criticisms is this: The arguments of natural theology entail the doctrine of divine simplicity; and thus, since (many of us would claim) we can know that those arguments are sound, we can know also that the doctrine of divine simplicity is true. Furthermore, the doctrine of analogy undermines any prima facie case for claiming that the doctrine of divine simplicity is unintelligible; and any residual sense of mystery is adequately accounted for by the fact that, given His nature, God is not the sort of thing we should expect to understand with the sort of clarity with which we understand the natural order.
What, then, of Craig’s third criticism, to the effect that the doctrine of divine simplicity entails that God has no accidental properties but only essential ones, which (Craig says) conflicts with the evident fact that God could have created a different universe and thus known, loved, and willed different things than in fact He has?
Here, building on a distinction famously made by Peter Geach, we need to differentiate between real properties and mere “Cambridge properties.” For example, for Socrates to grow hair is a real change in him, the acquisition by him of a real property. But for Socrates to become shorter than Plato, not because Socrates’ height has changed but only because Plato has grown taller, is not a real change in Socrates but what Geach called a mere “Cambridge change,” and therefore involves the acquisition of a mere “Cambridge property.” The doctrine of divine simplicity does not entail that God has no accidental properties of any sort; He can have accidental Cambridge properties.
Now it was Aquinas’s position that “since therefore God is outside the whole order of creation, and all creatures are ordered to Him, and not conversely, it is manifest that creatures are really related to God Himself; whereas in God there is no real relation to creatures, but a relation only in idea, inasmuch as creatures are referred to Him” (ST I.13.7). As Barry Miller points out in his book A Most Unlikely God, this amounts to the claim that while the relation of creatures to God is a real one, the relation of God to creatures is a mere Cambridge one, so that (for example) God’s creating the universe is one of His merely Cambridge properties.
How can this be so? As Brian Davies points out in his chapter on divine simplicity in An Introduction to Philosophy of Religion (3rd edition), what is essential to acting is the bringing about of an effect in another thing, not undergoing change oneself as one does so. What is essential to teaching, for example, is that one cause someone else to learn, and not that one lecture, write books, or the like. Of course, in created things, bringing about an effect is typically associated with undergoing change oneself (e.g. for us to cause another to learn typically requires lecturing, writing, or the like as a means). But that is accidental to agency per se, something true of us only because of our status as finite, created things. We should not expect the same thing to be true of a purely actual uncaused cause of the world. Hence there is no reason to suppose that God’s creation of the world entails a change in God Himself.
Nor does anything about God’s other relations to the world entail that they involve anything other than Cambridge properties. For example, as Davies points out, God’s love for the world is not like our love, which typically springs from some need. God, as purely actual, needs nothing; it is not that He has some lack which He seeks to remedy by creating us or getting us to love Him, which would entail a non-Cambridge change in Him. Rather, God loves us in the sense of willing what is good for us, which He does changelessly. Similarly, God’s knowledge of things is not a matter of coming to know them. Rather, He knows all things by virtue of knowing Himself as timelessly creating them.
Obviously, much more could be said. But this much suffices to show that here, as in so many other contexts, seemingly damaging objections to traditional theological doctrines lose much or all of their force when the doctrines are understood in light of the classical metaphysical picture within which they were originally formulated. (For those who are interested, the writings by Miller and Davies cited above are good places to look for more detailed treatments of the topic of divine simplicity. I also say a little more more about it in Aquinas, and Eleonore Stump has a very useful chapter on the subject in her book Aquinas.)