A Defense of Univocal Religious Language
April 4, 2011 — 11:11

Author: Trent Dougherty  Category: Christian Theology  Tags: ,   Comments: 3

Last week, I had the great good pleasure of hosting Richard Cross for a number of events at Baylor. To my knowledge, he’s one of the few people willing to defend (not necessarily as his own view, but as a perfectly sensible position) Scotus’ thesis that there must be *some* univocal concepts involved in predications concerning God. This got me to thinking about religious language.
It’s been a decade since I studied this, but the following argument is one I find highly suggestive. It sides with Scotus and, as I recall, the followers of Cajetan, in arguing that religious language can’t be analogy “all the way down.” Here’s my simple (perhaps simplistic) reasoning.

I have in mind as paradigms “God is good,” “God is wise,” and “God is loving.” There might be other families of speech that operate differently (Like “God is a lion,” etc.).
I like the view that in such cases a univocal concept is predicated analogously. I think they call this “analogy of proper proportionalitiy” but all my books on this are at the office. I want to focus on the concepts applied, though, not the mode of predication.
The view I want to defend–and it seems to be anathema to the Radical Orthodoxy crowd, though I can never really be sure what they’re saying (or who they are)–is that when we say “God is good” the concept GOOD* expressed must be “mostly the same” as the concept–call it GOOD and we won’t assume that GOOD*=GOOD or otherwise–expressed in assertions like “Trent is good” (I didn’t say they had to be true!) or “Ted is good” (in case truth makes a difference).
By “mostly the same”–and here we come to the point–I have this in mind. I might put it like this: They must have a significant overlap in “content” in that most of what GOOD entails is also entailed by GOOD* (I take it the notion of concept entailment is sufficiently clear).
For example, if X satisfies GOOD then X will not punish people for no reason, etc. If GOOD* doesn’t entail most of the same stuff, then I don’t know what we’re talking about. That is, when you say “God is good” I can’t infer much of anything from that. For all I know, that would allow for God to punish people without any reason and so God’s “goodness” would be meaningless to me. It would certainly not, and this was Lewis’ beef, provide me a (non-deviant) basis for worshiping God.
I’m inclined to think that the “overlap” of GOOD and GOOD* needs to be pretty close to coincidence. In point of fact, I can’t see any reason at present not to think they are identical. Call this the identity thesis.
ID1 In predications of the sort under consideration, in “God is F” and “Socrates is F” the predicate “is F” expresses the same concept C, though the “mode” of predication might be different.
The mode of predication business is meant to allow for the fact that it wouldn’t be surprising if the language worked differently in the case of an uncreated, infinite, and necessary being (but maybe something like my reasoning above would rule that out in this class of predications, which wouldn’t bother me…actually, just had an idea about that for another post.
ID1 seems to me to capture what (I understand to be what) Scotus is getting at and for similar reasons. One question I have is whether Thomists would disagree with this and why. I take it they do, but I don’t really understand why, and I think I’ve got a theory as to how to do what they want done consistent with ID1. Unsurprisingly, it has to do with pragmatics, rather than semantics.

  • Trent:
    Interesting post!
    1. ID1 seems to do justice to Aquinas’ main worry about univocity, namely that divine attributes are differently predicated of God and of creatures because of divine simplicity. God’s wisdom is God, while Socrates’ wisdom is a mere feature of Socrates, so God’s relationship to God’s wisdom is not the same as Socrates’ relationship to Socrates’ wisdom. But this seems to have to do with the mode of predication.
    2. It occurs to me that I don’t remember anything in Aquinas that commits him to non-univocity in regard to negative predicates. It could, thus, be that “is a non-bicycle” is univocally applied to God and Socrates. Perhaps the same would be true for disjunctive predicates. Thus, perhaps “Is divine or is Greek” is univocally applied to God and Socrates? Of course that I don’t remember anything in Aquinas that s is only weak evidence that there is nothing in Aquinas that s.
    3. I’d like to hear what you think of this post of mine on Aquinas and Scotus on analogy.

    April 4, 2011 — 19:43
  • Interesting, and for me especially, since I’ve just begun rethinking the problem of divine attributes.
    My first question, I think, is that of whether the analogy is one of concept, of predication, or of being. Though hardly in a position to defend myself here, I’m inclined to think that Aquinas and Aristotle had the last in mind.
    That is, the analogy is one of being because, as Alexander just pointed out in his reply, given divine simplicity, “God’s wisdom is God,” while given human complexity, “Socrates’ wisdom is a mere feature of Socrates.” So, as he says, “God’s relationship to God’s wisdom is not the same as Socrates’ relationship to Socrates’ wisdom.” As I’d put it, in the former case, it is a real identity. In the latter, it is a real possession or having.

    April 5, 2011 — 16:16
  • Hi there, longtime lurker here. I don’t think Scotus or his medieval followers would understand what it is for univocal concepts to be predicated analogously. I confess I don’t understand it either. The view you outline in most of this post is basically a version of analogy, not univocity (though your ID1 I think is accurate). Scotus holds that one can form a concept (or, I suppose, form a proposition) of one of the transcendentals or the perfections that you mention, that can be applied in the same sense to God and creatures. The concept, as Cross likes to say, is a vicious abstraction, because it does not correspond to any extramental reality (ie., there is no extramental reality, even being, in which God and creatures agree). The univocal sense/concept is contracted to either God or creatures when one adverts to the intrinsic modes of each (infinity for God, finitude for creatures); intrinsic modes are extramental, real features of essences and appear to have a similar function to the Porphyrian differentiae. They contract the univocal sense of being to what is contained in it. [As far as I can tell, intrinsic modes are Scotus’ answer to the problem you are solving by distinguishing sameness of concept from mode of predication]
    Thomists, including Cajetan, generally object to this picture because they think it makes being a genus, makes God just another being among beings (Radical orthodoxy), and destroys divine simplicity (they can’t separate the definition of a perfection from its mode of being, such as the essence/definition of goodness from its accidental character in humans).

    April 5, 2011 — 23:51