February 24, 2020

Making (Non)Sense of Apophatic Theology

Recently, I've been trying to sort out the historical context of Berkeley's remarks on the divine attributes—and particularly the doctrine of analogy—in Alciphron 4. As this text shows, early modern philosophers were much more knowledgeable about, and influenced by, medieval philosophy than is often assumed. So I've been reading up on medieval understandings of analogy and apophaticism. Unrelatedly, I've also been reading through Plantinga's Warranted Christian Belief. This is a hugely influential book, and I'd read—and even taught—the crucial central portion of the book but (I must admit) this is my first time reading it cover to cover.

In part 1 of the book, Plantinga offers some rather obvious criticisms of some particularly silly modern versions of apophaticism. I'm not familiar enough with the theology and religious studies literature to know whether these are strawmen. In any event, Plantinga attributes the (alleged) prevalence of these views to the influence of a particular bastardized version of Kant, and it is certainly true that there is a certain amount of popular lore in theology and religious studies about "what Kant has taught us" that doesn't align at all with how philosophers (including Kant scholars!) tend to think about the matter. However that may be, I think there is something else going on here: when apophatic theology is divorced from its Neoplatonic context, it tends (generally, but not inevitably) to devolve into nonsense.

The line of thought behind traditional (medieval) apophatic theology is something like this. In the standard or paradigm cases of affirmation, like 'Socrates is wise', the subject term picks out a substance (e.g., Socrates) and the predicate term picks out an accident (e.g., wisdom) that exists in that substance. Both substance and accident are understood in a realist sense—there is a metaphysical structure (inherence) mirroring the syntactic structure of predication. Socrates and (Socrates's) wisdom are really existing things that stand in a certain metaphysical relation. Now, insofar as God is Oneness Itself, Being Itself, etc., this kind of metaphysical structure has no application to God.

It follows from this that, in this paradigm sense of affirmation, no affirmations are true of God. Furthermore, insofar as all affirmations, taken in the paradigmatic sense, are false of God, it follows that all negations, taken in the corresponding paradigmatic sense, are true of God. Thus there is a perfectly good sense in which God is not wise, God is not powerful, and perhaps even God does not exist. That is, neither wisdom, power, nor being are accidents of God, since God has no accidents. However, these negations are misleading because, insofar as substance-accident structure is inapplicable to God, negative predicates also can't be applied to God in the paradigm sense: that is, in the paradigm sense, God is not foolish, powerless, or non-existent either.

Fortunately for our ability to talk about God, there are plenty of other examples of language that don't match the paradigm. So the question of religious language then becomes: what kind of non-standard use of language is involved in our talk about God? It is generally assumed that this still involves (successfully) using our concepts—concepts derived from the created world—to think about God and convey those thoughts in speech. Further, even if many uses of religious language turn out not to describe how God is in Godself, but rather to describe the product of God's activity (i.e., the created world), nevertheless the theological description explicitly considers the world as such a product and in this way makes reference to God. Thus, even if (to simplify an example from Maimonides) 'God is wise' is in some sense equivalent to 'the world is orderly', the former statement explicitly directs us to consider the orderliness of the world as a product of God's activity in a way the latter does not.

The Neoplatonic picture that motivates traditional apophaticism, then, includes three crucial components: (1) a realist theory of substance and property; (2) a sparse theory of properties; and (3) the claim that God's absolute oneness (simplicity) prevents the realist theory of properties form applying to God. In (1), we include the view about how language relates, in paradigm cases, to this metaphysics. By a sparse theory, in (2), I mean the view that, even in the created world, there are true affirmations whose predicate term does not pick out a property. The most obvious case here is that of negative predicates.

In short, the key assumption is that true affirmations sometimes, but not always, directly reflect metaphysical structures in the world. The cases where it does so reflect are the paradigms, and the cases where it doesn't are in some way deviant, but this does not necessarily preclude truth. When this assumption is rejected, then we get silly formulations of apophaticism like 'none of our concepts apply to God', which can be refuted by silly objections like 'what about the concept of being that to which none of our concepts apply?'

Doubtless some people have tried to pin this kind of silliness on Kant. (But, again, I suspect this might be a strawman version of what's going on in recent theology and religious studies.) Nevertheless, I think that neither the association of this view with Kant nor any particular argument derived from Kant is responsible for the attractiveness of this kind of mistake. One piece of evidence is that we can find instances of this tendency well before Kant. In fact, both William King (1710) and Peter Browne (1729) write extensively about analogical God-talk without ever treating divine simplicity as a motivation, and at one point Browne even makes rather dismissive comments about that doctrine. Neither King's nor Browne's accounts run into problems quite as obvious as the one with the concept of being that to which none of our concepts apply, and Browne's theory, in particular, is quite sophisticated. Nevertheless, I think they can serve as instructive examples.

If King and Browne are not motivated by divine simplicity, what does motivate them to deny that we can make literal and univocal affirmations of God?* It is quite clear that both King and Browne employ analogy defensively. That is, both think there are arguments for atheism (and, in Browne's case, also deism and Socinianism/Unitarianism) that depend on taking certain religious affirmations literally and univocally, and these arguments can be blocked by taking them analogically. King's primary interest is in arguments for the incompatibility of divine foreknowledge and providence with human free will, but he clearly sees this as also connected with the problem of evil (and this is even clearer if we read him against the background of Bayle, as Anthony Collins suggests we should).

But now the motivation for interpreting theism analogically has come apart from the motivation for endorsing theism in the first place, in a problematic way. Commenting on King, Berkeley writes, "there [is] no argument that I know of for [God's] existence, which does not prove him at the same time to be an understanding, wise, and benevolent Being, in the strict, literal, and proper meaning of those words" (Berkeley to Percival, 1 March 1709/10). For King, we initially argue on the basis of design for the existence of a wise and powerful God, then subsequently back off of this assertion, in the face of objections, by claiming that God is wise and powerful only in some analogical sense.

This shift creates two kinds of problems. In the first place, if apophaticism becomes just a defensive maneuver, and does not have independent motivation internal to theism, then it's hard to know exactly where to stop. What kinds of affirmations can we make about God and how should we understand them? If we can't figure out where to stop, we may run into the kind of obvious incoherence Plantinga points to, or we might end up making all God-talk into a poetic metaphor for the beauty of the world or some such. Second, and relatedly, once apophaticism starts to look like some kind of retreat from the full-blown theism with which we began, one begins to wonder—with Collins, Berkeley, and Plantinga—whether we have retreated all the way to atheism.




Note


* I formulate apophaticism as the denial that we can make literal and univocal affirmations of God because of another point of contrast between these 18th century theories and certain medieval theories. Aquinas, for instance, explicitly argues that there are literal ('proprie') affirmations of God, and treats analogy as a mean between univocal and equivocal speech. Both King and Browne treat analogy as a mean between literal and metaphorical speech, and (as critics, including Browne, immediately noticed) King totally fails to distinguish analogy from metaphor.

Posted by Kenny at February 24, 2020 11:03 AM
Trackbacks
TrackBack URL for this entry: http://blog.kennypearce.net/admin/mt-tb.cgi/870

Post a comment





Return to blog.kennypearce.net