Divine Simplicity II: Divine Attributes

This post covers another aspect of divine simplicity, namely, how it can be reconciled with a plurality of divine attributes. This problem itself is an expression of the more general problem of the relation between divine transcendence and human language. Divine simplicity is one way to ensure that God is unlike everything in the created world, for it is, after all, a negative doctrine: God is not composed, does not have parts, etc.

The scholastics came up with three solutions to the problem. The first was largely semantic. God is so transcendent and ineffable that he cannot be grasped by human thought or captured by human language. Divine attributes, such as wisdom and justice, are all one in God; when these are predicated of God, they signify primarily the divine essence as one. But there is also a secondary sense of these terms, which connotes the created realm as an effect of God. Only in the second sense are they considered distinct. Most of the twelfth-century thinkers held this view, and it was revived by Ockham and Auriol in the fourteenth century.

The second solution was primarily concerned with elaborating the role of the human intellect. Divine attributes are distinct only as a result of the operation of the intellect (that there are divine attributes is generally assumed based on the Dionysian via eminentiae). The intellect is too weak on its own in its present state to directly grasp God, so it requires a plurality of concepts. This plurality of concepts corresponds to the plurality of attributes. This second solution was authored by Bonaventure and Aquinas; or, more accurately, Bonaventure sketched it out and Aquinas developed it more fully. But he could never make up his mind about it, and one of his students that held one of his views was secretly investigated, and in general, Aquinas' changing views caused lots of problems for his would-be followers (such as, what "causes" the attributes, how can the divine essence be the fundamentum in re, is a "ratio" just a concept in the human mind or does it have an objective correlate in God?). So we will omit any further discussion of Aquinas. And in any case, Aquinas is irrelevant for understanding Scotus on this issue.

The relevant thinkers are Henry of Ghent and Godfrey of Fontaines (and, to a lesser extent, Thomas of Sutton). Henry has very complicated views on attributes, and they probably do a lot more work in his system than any other medieval thinker I know. For our purposes here, it suffices to note that Henry had a view regarding the origin of the attributes similar to Scotus' theory of instants of nature (see all the posts labeled "intelligible being"). Henry basically applies the three acts of the intellect from the Aristotelian commentary tradition to God. So we have an instant of simple apprehension in which the divine intellect apprehends the divine essence as one simple thing or one simple nature. But "then" it starts to reason about the essence, and by doing this it generates the attributes (attribute=divine essence+ratio from the intellect). However, and this is important, it does not generate the will, even though it is a sine qua non cause of volitional acts. So in the third instant the will is actualized and begins to go through its own series of movements. In the end we have then two fundamental attributes that cannot be reduced to each other, and all other divine attributes are ordered to one of these primary attributes (incidentally, intellect and will serve as the principles for the emanation of the divine persons, but visit the "Henry of Ghent" blog for more on this).

Godfrey of Fontaines thought all this was bullcrap, and instead extended Aquinas' views on divine ideas to help out the problem of attributes. Basically, ignoring his arguments against Henry, Godfrey thinks that God can compare the divine essence to any creature, and since he is omniscient, and because creatures imitate the divine essence in various ways (hence the multiplicity of perfections that are attributed to God), God can compare his essence to the contents of the human mind and see that the human mind, because of its weakness, sees a plurality of attributes in God. So the distinction of attributes is not really in God at all, just the human mind, but God does know that in a derivative sense he has attributes. So in the end, Godfrey cannot avoid positing some movement in divinis either. [this is not entirely accurate, but I don't want to reread either Godfrey or that chapter of my diss.]

We turn now to Scotus. As is probably well known now to all readers of this blog, Scotus has two commentaries on the Sentences of Peter Lombard, a Lectura and an Ordinatio, as well as a series of student reports, Reportationes, the most trusty of which seems to be the one labeled I-A. In the first two commentaries, Scotus’ discussion of divine attributes is in d. 8 q. 4 in Reportatio I-A, although the doctrine is the same (save more possible variations regarding the formal distinction), the discussion of it has migrated to d. 45, which is about the divine will. Consequently, I will focus here on the Ordinatio. I do recommend reading the Reportatio, however, for it adds the notion of the propositio famosa, which holds that whatever is distinct in reason can be treated as if it were really distinct; Scotus uses this principle to help him escape from objections to his views based on the identity of indiscernables (as Scotus puts it, if a is the same as c and b is the same as c, then a is the same as b).

The basic point that Scotus argues is that the attributes are distinct prior to or apart from any operation of the intellect, whether the intellect in question is divine, human, or angelic. To start off, in the solution of d.8 q.4 Scotus accepts that there are distinctions of reason in God, as well as distinct formal objects, that is, between different modes of conceiving the same object. This suffices for distinctions between say ‘wise’ and ‘wisdom’, but not between entities like wisdom and truth. This is because God knows the divine essence intuitively (see here for intuitive cognition), and can only find these entities in the essence; he does not cause them by means of his intellect. Here is the argument to this effect:

Ordinatio I d. 8 q. 4 (ed. Vat. IV, 257):

“Praeterea, intellectus intuitivus nullam habet distinctionem in obiecto nisi secundum quod exsistens est, quia sicut non cognoscit aliquod obiectum nisi ut exsistens, ita non cognoscit aliqua distincta formaliter in obiecto nisi ut exsistens est. Cum ergo intellectus divinus non cognoscat essentiam suam nisi intellectione intuitiva, quaecumque distinctio ponatur ibi in obiecto – sive sit distinctiorum obiectorum formalium, sive ut rationum causatarum per actum intellectus – sequitur quod ista distinctio erit in obiecto ut actu exsistens est: et ita si ista est obiectorum formalium distinctorum in obiecto, erunt ista distincta formaliter (et tunc sequitur propositum, quod talis distinctio obiectorum formalium praecedit actum intellectus), si autem sit rationum causatarum per actum intelligendi, ergo intellectus divinus causabit aliquam intellectionem in essentia ‘ut relationem rationis’, ut est exsistens, quod videtur absurdum.”

Translation:
"Furthermore, an intuitive understanding has no distinction in an object except according as it is existing, because just as it does not know some object save as existing, so it does not know something to be formally distinct in the object unless as it is existing. Since therefore the divine intellect does not know its essence except by an intuitive intellection, whatever distinction is posited there in the object – whether it is of distinct formal objects or as definitions caused by the act of the intellect – it follows that that distinction will be in the object as it is existing in act; and so if that is of formally distinct objects in the object, they will be formally distinct (and then the matter at hand follows, that such a distinction of formal objets precedes the act of the intellect), if however it is of definitions caused by the act of understanding, therefore the divine intellect will cause some intellection in the essence, as a relation of reason, as it is existing, which seems absurd."

The result of this is that there is a distinction preceding the operation of an intellect, such that wisdom is in God and goodness is in God, but wisdom in God is not formally goodness in God. Scotus thinks he has an argument that proves this.

Ord. I d. 8 pt. 1 q. 4 n. 192 (ed. Vat. IV, 261)
Quod probatur, quia si infinita sapientia esset formaliter infinita bonitas, et sapientia in communi esset formaliter bonitas in communi. Infinitas enim non destruit formalem rationem illius cui additur, quia in quocumque gradu intelligatur esse aliqua perfectio (qui tamen ‘gradus’ est gradus illius perfectionis), non tollitur formalis ratio illius perfectionis propter istum gradum, et ita si non includit formaliter ‘ut in communi, in communi’, nec ‘ut infinitum, infinitum’.

Translation:
"This is proved: because if infinite wisdom would be formally infinite goodness, then wisdom in common would be formally goodness in common. For infinity does not destroy the formal ratio of that to which it is added, because in whatever grade some perfection is understood to be (which grade, nevertheless, is a grade of that perfection), the formal ratio of that perfection is not taken away because of that grade, so if it [wisdom], as in common, does not formally include [goodness] in common, neither [will wisdom] as infinite [include goodness] as infinite."

This is a pretty compressed argument, and I’m not at all sure what’s going on at the end. This is the clear part:

If infinite wisdom were formally infinite goodness, then wisdom in common would be formally goodness in common.

The likely interpretation of this is that Scotus has in mind his doctrine of ultimate abstraction from Lec./Ord. I d. 5. According to this notion, the mind can perform a series of abstractions from a material object and ultimately arrive at a pure quiddity or definition. With this in mind, the argument means that if wisdom and justice, qua infinite, are the same, then at the level of pure abstraction (that is, with infinity having been abstracted) wisdom and justice must also be the same. Scotus takes this to be false, and the remainder of the quoted passage supports the claim that infinity does not alter the definition of something, in this case, a pure perfection.

Scotus follows this argument with further considerations on what if means to be formally included in the definition of something:

Ordinatio I d. 8 pt. 1 q. 4 (ed. Vat. IV, 261-62)

Hoc declaro, quia ‘includere formaliter’ est includere aliquid in ratione sua essentiali, ita quod si definitio includentis assignaretur, inclusum esset definitio vel pars definitionis; sicut autem definitio bonitatis in communi non habet in se sapientiam, ita nec infinita infinitam: est igitur aliqua non-identitas formalis sapientiae et bonitatis, in quantum earum essent distinctae definitiones, si essent definibiles. Definitio autem non tantum indicat rationem causatum ab intellectu, sed quiditatem rei: est ergo non-identitas formalis ex parte rei, et intelligo sic, quod intellectus componens istam ‘sapientia non est formaliter bonitas’, non causat actu suo collativo veritatem hiuius compositionis, sed in obiecto invenit extrema, ex quorum compositione fit actus verus.

Translation:
I declare this, because ‘to include formally’ is to include something in its essential definition, so that if a definition of the including could be assigned, the included would be a definition or part of a definition; just as the definition of goodness in common does not contain wisdom, so neither [does the definition of] infinite [goodness contain the definition of] infinite [wisdom]. Therefore there is some formal non-identity of wisdom and goodness, insofar as they would have distinct definitions, if they were definable. A definition, however, does not only indicate the notion/definition caused by the intellect, but the quiddity of the thing. Therefore there is formal non-identity form the side of the thing, and I understand this in such a way that the intellect composing that proposition ‘wisdom is not formally goodness’, does not cause the truth of the proposition by its own comparative act, but it finds the extremes in the object, from the composition of which the act is made true."

The basic idea here is that none of the divine attributes include each other in their definitions or parts of definitions, and this is true apart from any operation of the intellect.

So there you have it. The attributes are distinct ex natura rei (which means they are distinct prior to the operation of any intellect, human or divine), a distinction that is formal (the formal distinction is doing most of the work here, so see the relevant post). In God the attributes all exist under the extrinsic mode of infinity, which safeguards divine simplicity (for more on infinity see the ‘natural knowledge of God’ post in this series). When ultimate abstraction is performed, the intellect discovers that these attributes are distinct because none of them fall into the definitions of the others.