Craig on Thomism and divine simplicity, Part I

In an earlier article, I discussed William Lane Craig’s treatment of the kalām cosmological argument in Volume IIb of his Systematic Philosophical Theology.  Let’s now take a look at Volume IIa, which is devoted to the divine attributes.  Specifically, I’ll address what he has to say about Thomism and the doctrine of divine simplicity.  This is part of a larger discussion of the doctrine, but I’ll confine what I say in this two-part article to his treatment of the Thomistic position, for two reasons.  The first is the practical one of keeping these posts to a manageable length.  The second is that, as a Thomist, I think the Thomistic approach to the issue is the most plausible and important.  Craig might agree with that much.  He says that “Thomas Aquinas [is] the preeminent defender of divine simplicity” (p. 104), that “historically, the doctrine of divine simplicity comes to its fullest expression in the theology of Thomas Aquinas” (p. 136), and again that “with Aquinas we have reached the highwater mark in the history of the development of the doctrine” (p. 143).

Mishandling Aquinas

Craig has nevertheless long been very critical of the doctrine of divine simplicity in its traditional form, especially as understood by Thomists, and his treatment in Systematic Philosophical Theology continues in this vein.  Unfortunately, some of the weaknesses of his earlier discussions recur in this latest treatment.  As I have pointed out before, some of Craig’s objections in previous work rest on serious misunderstandings of Aquinas’s views.  The discussion in Systematic Philosophical Theology also gets Aquinas and other Thomists wrong in several serious ways.

I am well aware that critics of Thomism often complain that Thomists frequently accuse them of misunderstanding Aquinas.  But if the accusation is just, it really doesn’t matter how many times it is leveled.  Moreover, critics of Thomism (including Craig) themselves often emphasize how very different Thomism’s general metaphysical premises are from the options with which contemporary philosophers tend to be more familiar.  They can hardly be surprised if they do in fact often fall into misunderstandings.

In any event, with much of what Craig says, the mistakes are unambiguous and it simply cannot reasonably be denied that he is getting Aquinas and other Thomists wrong.  For example, he claims that “contemporary Thomists recognize and celebrate Thomas’ transformation of Aristotle’s conceptual distinction into the so-called real distinction between essence and existence.  But they also recognize that Thomas nowhere tries to justify such a real distinction” (p. 148).  This is a bizarre assertion.  The reality is that Aquinas offers several arguments for the distinction and that contemporary Thomists commonly understand him to be doing so, most famously in his short work De Ente et Essentia.  There Aquinas supports the real distinction in contingent things by arguing that one can know such a thing’s essence without knowing one way or the other whether it exists, and that if there is something in which essence and existence are identical, it would have to be unique.  John Wippel devotes chapter V of his book The Metaphysical Thought of Thomas Aquinas to Aquinas’s arguments for the real distinction.  On p. 275, note 1 of his book Real Essentialism, David Oderberg cites Book II, Chapter 52 of Aquinas’s Summa Contra Gentiles as a source of arguments for the real distinction.

Many more examples could be given of Thomists who take Aquinas to provide arguments for the real distinction; to take just two recent ones, there is Gaven Kerr’s treatment of Aquinas’s views in chapter 1 of his book Aquinas’s Way to God, and my discussion at pp. 241-46 of Scholastic Metaphysics.  But the reason I cite Wippel and Oderberg in particular is that Craig himself cites them, on the very same page on which he asserts that “contemporary Thomists… recognize that Thomas nowhere tries to justify such a real distinction”!  How could he make such an egregious mistake?  The answer is that it seems clear that Craig did not actually bother to look at what Wippel and Oderberg wrote.  Instead, when summarizing their views he cites what James Dolezal says about them in his book God without Parts.  But Craig also gets even Dolezal wrong.  In footnote 157 on p. 148 of his book, Craig says:

Dolezal comments, “Thomas does not attempt to prove the real distinction between essence and existence in creatures, but instead simply presupposes it as background to his exposition of the divine essence as identical with the divine existence” (Dolezal, God without Parts, 99)

But if you look at pages 98-99 of Dolezal’s book, you’ll find that what he is referring to is not Aquinas’s work in general, but rather merely a specific passage from Aquinas’s commentary on the Sentences of Peter Lombard.  And Craig does not quote the entirety of Dolezal’s sentence, which actually begins: “In this passage Thomas does not attempt to prove the real distinction between essence and existence in creatures…” (emphasis added)!  By deleting those first three words, Craig has greatly distorted Dolezal’s meaning.  To make matters worse, on p. 97 of his book, Dolezal had already explicitly noted that “in De ente et essentia Aquinas famously makes the case for the real distinction between essence and existence in creatures,” and goes on to discuss what he characterizes as the “argument” Aquinas there gives for the distinction.  Hence even Dolezal, who seems to be Craig’s main source for what contemporary Thomists think, says the opposite of what Craig attributes to him.

Apparently, it is because Craig misunderstands Dolezal that he misunderstands Aquinas, and into the bargain also misunderstands contemporary Thomists like Wippel and Oderberg.  It seems that, because Craig mistakenly thinks that Aquinas gave no arguments for the real distinction, he assumed that the arguments from Wippel and Oderberg cited by Dolezal were their own arguments rather than Aquinas’s.  But in fact, in the passages quoted by Dolezal and then quoted from Dolezal by Craig, both Wippel and Oderberg are discussing arguments they derived from Aquinas.

A second egregious misunderstanding of Aquinas is implicit when Craig writes:

It has traditionally been argued that a perfect being cannot have any passive potentiality but must be fully actual and therefore immutable.  This argument is counter-intuitive and, as we shall see, has theologically unacceptable implications.  In the hands of advocates of divine simplicity, it entails that God lacks, not only any Aristotelian ability to change intrinsically, but, more radically, even the capacity to be different than he is.  It is enough for now to observe that a being who has infinite resources of untapped potential is plausibly greater than a being whose power is exhausted by the creation of the finite world.  What God has actually created could not possibly represent the full range of his power.  The ability to create an indefinite number of different universes is obviously a great-making property which must therefore be ascribed to God. (p. 144)

In the paragraph immediately before this, Craig refers to Aquinas’s views about the relationship between perfection and being composite, and in the paragraph immediately following the one just quoted from, Craig discusses Aquinas’s views about divine immutability.  And Aquinas does hold that God lacks any passive potentiality and is fully actual.  Hence the reader would naturally infer that what Craig is suggesting in the passage quoted is that Aquinas holds, or at least says things that entail, that God’s “power is exhausted by the creation of the finite world,” that “what God has actually created… represent[s] the full range of his power,” and that he therefore lacks any “ability to create an indefinite number of different universes.”

But Aquinas neither says nor implies any such thing, and indeed he says exactly the opposite.  In fact, he explicitly addresses the objection that if God is pure actuality and without potentiality, then he would be lacking in power.  As he notes, this objection rests on an equivocation on the word “potentiality”:

There are two kinds of power or potentiality, viz., (a) passive power, which in no way exists in God, and (b) active power, which one must especially posit in God.  For it is clear that each thing, insofar as it has actuality and perfection, is an active principle of something, whereas each thing is acted upon insofar as it is deficient and imperfect.  Now it was shown above that God is pure actuality, and that He is perfect absolutely speaking and in every way, and that imperfection has no place in Him.  Hence, it especially belongs to Him to be an active principle and in no way to be acted upon.  But the nature of an active principle belongs to an active power.  For as the Philosopher says in Metaphysics 5, an active power is a principle of acting upon another, whereas a passive power is a principle of being acted upon by another.  Therefore, it follows that active power exists especially in God.

Active power, far from being contrasted with actuality, is instead grounded in actuality.  For each thing acts insofar as it is actual.  Passive power, on the other hand, is indeed contrasted with actuality, since a thing is acted upon insofar as it is in potentiality.  Hence, passive power is excluded from God – but not active power. (Summa Theologiae I.25.1, Freddoso translation)

For this reason, Aquinas goes on to argue in the succeeding article that precisely because God is fully actual, his power is infinite (Summa Theologiae I.25.2).  And three articles after that, he explicitly addresses the claim that God’s power must be limited to creating the world that actually exists, and he argues that, on the contrary, “the divine wisdom is not so restricted to any particular order that no other course of events could happen.  Wherefore we must simply say that God can do other things than those He has done” (Summa Theologiae I.25.5, Fathers of the English Dominican Province translation).

Needless to say, Craig would no doubt disagree with the arguments Aquinas gives in these passages.  But the point is that Aquinas here argues for claims that are precisely the opposite of what Craig seems to attribute to him in the passage from Systematic Philosophical Theology quoted above.

A third misunderstanding concerns Aquinas’s view that God’s essence is identical to his existence but that we cannot give that essence a proper definition in terms of genus and differentia.  According to Craig, this entails that “we have no idea what that essence is” and that God’s being “is not conceivable by us” (p. 143), so that God’s nature is “unintelligible” (p. 139).  As a consequence, “Thomas’ doctrine thus leads to a profound agnosticism about God, a conclusion which is, in the words of Eleonore Stump, ‘religiously pernicious.’” (p. 160).  This last remark gives the impression that Stump agrees with Craig’s characterization of Aquinas’s position.  But in fact Stump’s view, like Dolezal’s, is the opposite of what Craig’s more unwary readers would suppose.  In reality, she explicitly argues that to think that Aquinas’s view entails agnosticism is to misunderstand him.  And the view she characterizes as “religiously pernicious” is not Aquinas’s own view, but a view she thinks rests on a misinterpretation of Aquinas.

The essay of Stump’s from which Craig quotes is “Simplicity and Aquinas’s Quantum Metaphysics.”  She devotes a whole section of the paper to rebutting the claim that Aquinas asserts or implies agnosticism about the divine nature.  As she notes, Aquinas makes many positive assertions about the divine nature.  For example, in the Prima Pars of the Summa Theologiae, Aquinas argues that among the divine attributes are pure actuality, goodness, omnipotence, intellect, and will (though Stump herself does not discuss all of these).  Indeed, as Stump emphasizes, Aquinas explicitly rejects Maimonides’ view that we can say only what God is not, and speak affirmatively only of his effects rather than of God himself. Hence, in Question 13 of the Prima Pars, Aquinas writes:

Absolute and affirmative names of God, as “good,” “wise,” and the like… signify the divine substance, and are predicated substantially of God, although they fall short of a full representation of Him… So when we say, “God is good,” the meaning is not, “God is the cause of goodness,” or “God is not evil”; but the meaning is, “Whatever good we attribute to creatures, pre-exists in God,” and in a more excellent and higher way.  (Summa Theologiae I.13.2)

Of terms like “good” when applied to God, Aquinas adds that “each one signifies Him in an imperfect manner, even as creatures represent Him imperfectly.”  But to signify and represent him imperfectly does not entail failing to signify or represent him at all. 

In the succeeding article, Aquinas adds that "’being,’ ‘good,’ ‘living,’ and the like” are here not to be understood merely metaphorically, but that “such names can be literally applied to God” (Summa Theologiae I.13.3).  It is true that he goes on to argue that these terms must also be understood analogically rather than univocally, and Craig is also critical of Aquinas’s account of analogical language.  The point of that account, though, is precisely to show how we can make intelligible, literal, affirmative statements about the divine nature without speaking either univocally or equivocally, despite our being unable to define that nature in terms of genus and differentia.  Craig’s discussion gives the impression that Aquinas either outright acknowledges that his position entails a radical agnosticism that makes the divine nature utterly unintelligible, or at least does little to try to avoid this result.  But that is precisely the opposite of the truth.

Part of the problem here is that Craig appears to misunderstand what it means to say that we cannot define the divine nature.  He speaks as if to “define” something is to attach some sort of meaning or conceptual content to it, so that if we cannot define the divine nature, that entails that our statements about God are without meaning or conceptual content, and thus unintelligible.  But that is not what Aquinas and other Thomists mean.  They use “definition” and related terms in a narrow technical sense which simply entails identifying the genus and differentia of the thing being defined.  There is no implication at all that a failure to identify genus and differentia entails that one’s words are meaningless or unintelligible.

This should be obvious enough just from everyday examples which have nothing to do with theology.  The average person could not give you a “definition” in this Thomistic sense of the essence of a star, or of salt, or of a virus.  He might even be unable to understand the concepts from physics, chemistry, and biology that would be necessary to giving such a definition.  But that does not entail that he “has no idea” what stars, salt, or viruses are, or that their natures are flatly “not conceivable” or “unintelligible” to him.  To be sure, our grasp of the divine essence is vastly more imperfect than our grasp of the natures of such things.  The point, though, is simply that the absence of a definition in terms of genus and differentia does not entail unintelligibility or agnosticism.

There are yet other serious misunderstandings of Aquinas.  For example, Craig attributes to him the view that “God has no essential properties, indeed no properties at all” but also “paradoxically, that God does have all perfections” (p. 138).  But it is highly misleading at best to say that Aquinas takes God to have “no properties at all,” and it is only if one badly distorts his position that it will sound paradoxical.  In support of his characterization of Aquinas’s view, Craig cites a remark Brian Davies makes to the effect that Aquinas “denies that, in a serious sense, God has any attributes or properties” (at p. 36 of Davies’ article “Simplicity,” in Charles Taliaferro and Chad Meister, eds., The Cambridge Companion to Christian Philosophical Theology).  But that this demands significant qualification is obvious from the points we saw Stump make.  Moreover, Craig leaves out crucial qualifying remarks made by Davies himself.  Davies immediately goes on to explain what he means, by saying that “normally, attributes or properties can be distinguished from the things that possess them.  Yet… Aquinas denies that God and his nature can be thought of as distinct” (p. 36).  He also goes on to write:

Aquinas thinks that there are positive reasons for employing different words (and meaning them literally) when it comes to talking about God.  He thinks, for example, that there are positive reasons for speaking of God as omnipotent and good.  But, in addition, he thinks that there are positive reasons for not taking God to be an individual exhibiting a variety of attributes… So he concludes… that goodness and power in God do not amount to different realities. (p. 39)

Craig seems to think that Aquinas denies that God has any properties or attributes in the sense that he is something like a bare or featureless particular.  If that were so, then it would indeed be paradoxical to go on to attribute goodness, omnipotence, and the like to God.  But as is obvious not only from what we’ve seen Aquinas himself say, but also from what Davies says, that is not what Aquinas means.  What he means instead is that while God does have attributes like goodness and omnipotence – and therefore is not a bare or featureless particular – his goodness and omnipotence are realities that are identical to each other and to God himself.  This makes his attributes unlike those possessed by created things, which are distinct from each other and from the substance that bears them.

In the second part of this article, I’ll turn to Craig’s treatment of the more general metaphysical disputes underlying the debate over divine simplicity, and to specific criticisms he raises against the doctrine and the arguments of its defenders.