Tuesday, April 5, 2011

Analytic Appropriations of Univocity

Check out the following two links: Proslogion and Alexander Pruss. For Scotus' texts on univocity see our fundamentals post.

But let's consider whether Aquinas and Scotus disagree.

We know the following:

Scotus thinks there are two conditions for a univocal concept.
1. to affirm and deny with respect to the same results in a contradiction.
2. It can be used as a middle term in a syllogism without there being a fallacy of equivocation.

Aquinas defines univocity (see for example Summa Ia q. 13 and De unioni verbi a. 2 ad 4) as when two things have the same name and the same definition. This is Aristotle's definition from the Categories.

Aquinas also thinks (Summa Ia q. 13 a.5?) that analogical concepts are such that they can serve as the middle term in a syllogism without there being a fallacy of equivocation.

Now some notes about the history of equivocity/univocity. We have seen Aristotle's view of univocity. His view of equivocity is when the name is the same but the definition is different. In the Metaphysics he admits of a kind of equivocity that is "focused" or has related meanings, and uses the health example. This is Aquinas' analogy. Aristotle's analogy shows up in the Ethics and consists of a proportion, and always involves four terms (A:B::C:D). Scotus' definition of univocity allegedly comes from Simplicius' commentary on Aristotle's categories (I say allegedly because I've never been able to find it. Of course, I haven't looked very hard either).

Scotus sees only two options: univocity or equivocity. There is no medium. Analogy is a subset of equivocity, and as such will result in a fallacy of equivocation. Aquinas sees three options: equivocity, univocity, and analogy. Mysteriously, he thinks that analogy is a medium between the extremes and so does not commit a fallacy of equivocation. I tend to side with Scotus on this point, given the history of the problem.

In rather annoying (perhaps, truly Scotistic?) fashion, Scotus also thinks there can be analogical concepts, and never bothered to attack Aquinas' notion of analogy (what may be important today was not necessarily seen as such in the 13th century), save in Collatio 23, which doesn't have a resolution. So we can fault Aquinas for confusing analogy with equivocity, and Scotus for not telling us how univocal concepts relate to analogical ones and for not analyzing Aquinas' position.

What does all this mean? Well, given the 700 year history of this debate, my readers should not be surprised that I arrive at no definitive conclusions. But if we ignore for the moment Aquinas' belief that analogical concepts avoid fallacies of equivocation and focus on his definition of univocity, a way of harmonization presents itself. For it is clear from Scotus' account that he is primarily interested in concepts, and there is no "real" correspondence between the univocal concept of being and being outside the mind. But Aquinas' definition of univocity concerns two things; and given all his other discussions of analogy in which it is clear that univocity is impossible because of the nature of the divine causality (ie., its equivocal), it's clear that Aquinas is primarily concerned with the "real", and that any analogical concepts are isomorphically related to their real foundations (hence, he has to say analogical concepts don't cause fallacies, because otherwise there would be no systematic theology, only mystical experience a la David Burrell's "theology is a dance"). So, to conclude, we could harmonize our medievals by the claim that they are in fact complementary, for Scotus thinks univocity is on the level of the concept, while Aquinas thinks that analogy is on the level of the real.


Deacon David said...

On Dr. Pruss's blog, linked above, commenter Brandon cites Scotus's definition of univocity, from Op. Ox. I, dist III, q i (Philosophical Writings, Wolter, tr., p. 20): "I designate that concept univocal which possesses sufficient unity in itself, so that to affirm and deny it of one and the same thing would be a contradiction. It also has sufficient unity to serve as the middle term of a syllogism, so that wherever two extremes are united by a middle term that is one in this way, we may conclude to the union of the two extremes among themselves."

This leaves me confused. It seems that on this definition, there can't be univocity in the concept of, e.g., wisdom, in the statements "God is wise," and "Socrates is wise;" or even between "Socrates is wise," and "Thomas is wise." In God, wisdom is perfect, infinite, and identical with being, and therefore it is true that God cannot be both wise and not wise. But in any creature, like Socrates or Thomas, wisdom is necessarily imperfect, finite, and not identical with being; and therefore for any creature, it is not a contradiction to be both (imperfectly) wise and (to the same extent) not wise.

But this is the opposite of where Scotus goes with this, isn't it? What am I missing?

Michael Sullivan said...

Deacon David,

as a pure perfection wisdom, according to Scotus, is a transcendental, and so indifferent to finity or infinity, etc. See my recent post on the transcendentals, where I discuss this very example:


Lee Faber said...

Also, note that these two conditions just tell you what Scotus means by "univocity". They are not an argument showing that something is univocal to God and creatures. As it turns out Scotus thinks human beings can form an univocal concept of a perfection that is indifferent to the perfect, infinite way it is in God and the imperfect, finite way it is in creatures.

Note, however, that it is a contradiction to say that Socrates is wise and Socrates is not wise. Wisdom is a perfection in the genus of quality that inheres in the soul. One may not be very wise, but if one has any wisdom, one has the quality; basically, it is a matter of degree.

Hence the Thomist complaints: it looks like God is just a degree of being and so a being among beings blah blah blah. This ignores Scotus' claims regarding infinity, which insures that there is still an infinite gap in reality between God and creatures (also note that the univocal concept of being does not correspond to any extramental reality).

Brandon said...

Strictly speaking, Aquinas's view of equivocation is that the name is the same but the definition totally different (totaliter diversa or omnino diversa), and his view of univocation is that the name is the same and the definition totally the same (omnino eadem). When he drops the totaliter or omnino, it's always still pretty clear that it's understood: he reads the terms of the definition in a strict sense. (He also explicitly recognizes in places that this is a narrower use of the term equivocation than we find in Aristotle himself.) This is why he thinks it's obvious that there has to be a mean between the two: nothing requires that the rationes of names be either totally different or strictly one and the same, and in examples like healthy or predicating 'animal' of both a real animal and a painted animal (also used by Aquinas on occasion, and I think a more illuminating example), there are some cases that are fairly plausibly not either.

Lee Faber said...

Brandon, I agree. There was a tradition originating with Boethius that divided equivocals into those that had nothing in common but a name (chance equivocals) and those that had related meanings as well (conventional equivocals).

Sadly, the largest medieval treatise on univocity and analogy has yet to be edited: Peter thomae's De ente, which in critical edition would probably be over 400 pages. he gives an extensive discussion of Aquinas, the Boethian tradition, the arabic contributions, as well as contemporaries from his own time like Auriol and Gerard of Bologna.

Brandon said...

the largest medieval treatise on univocity and analogy has yet to be edited

That's pretty much the standard thing with medieval thought, though. What you guys really need to do is starting drafting people at gunpoint to help do editorial work and translation; otherwise you'll never finish.

Lukáš Novák said...

I must say that I more disagree than agree with you, Lee. For one thing, it is difficult to speak of "Aquinas's theory of analogy", because Aquinas simply does not have any coherent theory of analogy underlying his various rather ad hoc applications. The "definition" of analogical things that, beside having the same name, have partly the same and partly different "ratio" is quite unsatisfactory. Almost any arbitrary homonyms have "ratio's" which are "partly the same".

Now regarding Scotus: I am quite convinced that his statements favorable to analogical concepts expressed in his Questions on Metaphysics are very early and incompatible with his later view.

When Scotus says that there is no reality corresponding to the concept of being, it must not be understood so that the concept does not capture anything real and common to God and creature at all. It most surely does, since any real, non-fictive concept conceives something real. It just means that that real which is conceived "a parte rei" is not a reality or formality, that is, a formally distinct generic metaphysical grade, perfectible by means of a differentia, but a mere common aspect of the proper realities of God and Creature, conceived inadequately (that is, non-exhaustively), viz. without their proper intrinsic modes of infinity and finitude respectively. Scotus can say that God and creature are "primo diversa in realitate" only in this sense: meaning that they do not share any formally distinct, and therefore adequately conceivable, formality/reality.

Now it is true that Aquinas speaks of things whereas Scotus of concepts; but this still does not leave any room for reconciliation, IMHO. First, because Aquinas defines analogical things by means of their partially-diverse definitions or "rationes", which correspond to Scotus's concepts; while for Scotus real concepts are always abstracted from reality and capture something of the reality itself, even in case of the concept of being. Therefore both for Aquinas and for Scotus real analogy must result in conceptual analogy, and vice versa. Second, because for Scotus univocity, defined as capability to function correctly in elementary logical relations and operations, entails perfect unity of the respective concept, its "indifference" toward, i.e. perfect abstraction from, its subordinates. This is incompatible with Aquinas's notion of analogy.

The quarrell between the thomists and the scotists is ultimately over the definition of univocity - for it would be futile to attempt to defend usefulness of any kind of concepts that do not stand up to the elementary requirements of logic.

According to the thomists, Scotus's definition is too broad, as it comprises also analogical concepts. Cajetan elaborates his theory of analogical concepts that have a lesser kind of unity
(proportional unity), precisely in order to be able to show that perfect univocal unity is not required to satisfy Scotus's definition. But that is Cajetan, not Aquinas; Aquinas's own notion of analogy seems to be simply too vague to allow any clear comparison with Scotus.

Suárez tried to harmonise the two views, but in doing so he had to redefine both univocity and analogy: his notion of univocity is narrower than Scotus's, and his "analogy of intrinsic attribution" would be classified as perfect univocity by Scotus (in fact, I think that it is just Scotus's teaching on essential order as a transcendental attribute of being under disguise).

Lee Faber said...

Thanks Dr. Novak. I haven't read the texts on whether there is a corresponding reality or not in some time. I will have to look into this.

But I am reluctant to abandon the division of univocity and analogy into the real vs. conceptual, or the logical-metaphysical-physical (though this may not be the same, and in any case, it may indeed be useless for harmonizing Aquinas and Scotus). This played a strong role in Scotus' development, and appears to have survived among the 14th c. Barcelona scotists as well.

Anonymous said...

I'm not really sure to which blog post I should append the following question, if to a blog post at all; but here it goes: Would one of you--Mr. Faber or Mr. Sullivan, or both--happen to have any idea as to which grad programs--either philosophy or medieval studies Ph.D. programs, specifically--make it easiest for their students to engage in a serious, intensive study of the thought of Scotus, and to critically compare his work with that of St. Thomas? Thank you very much for any information with which you are able to provide me.

Lee Faber said...


I dont think you need to look at Medieval studies programs at all. Generally, these are dominated by history and literature faculty. I suppose it depends if 1. you want to get an academic job and 2. whether you would rather teach in a philosophy dept. or a theology dept.

While there are a number of places to study Aquinas, I don't think there are many where you could study scotus unless you are self-motivated.

I think there are basically three choices:
Notre Dame
Univ. of Toronto

ND is probably the best option. You have two major Scotus scholars (Richard Cross, Stephen Dumont) as well as the presence of the Scotist Commission. Even if you're not interested in textual matters, a number of scotus scholars will be around, including Noone from CUA.

CUA has Noone, a co-head of the scotist commission, as well as Tobias Hoffmann, who has written extensively on both Scotus and Aquinas, plus a strong Thomist contingent.

Toronto has Peter King, and other late medieval phil. people such as Pickave. I don't know about thomism there, now that Maurer has passed away. There are modern thomists at the jesuit college in Toronto, but that's not quite as useful for comparing Aquinas and Scotus.

So that's my gut response. If you have languages, you could study at Bonn or Koln; there are a number of medieval philosophy centers out there. And there's always Oxford, though their main Scotus scholar went to ND.

Michael Sullivan said...


I took my PhD at CUA. My doctoral committee had on it Timothy Noone, one of the continent's preeminent Scotists, John Wippel, one of the continent's preeminent Thomists, and Michael Gorman, an analytic philosopher/theologian who has written on both Thomas and Scotus. As Faber mentions, Hoffman is also there as well as a number of other Thomist or Thomist-leaning scholars. So I think that CUA served me just fine for the purposes you're asking about.

This is not to denigrate Notre Dame at all, but Faber has more experience with them than I do. I would also note that unlike Notre Dame and most places the overall atmosphere of the philosophy department at CUA, while not hostile to modern and contemporary philosophy, and while not being a "history of philosophy" department exactly, is much more conducive to studying other periods on their own terms than most other departments.