Monday, August 27, 2007

Divine persons and unity-making relations

Scott Williams has made some very helpful comments on a previous post.

One of the points that came up concerns why the divine essence and a personal property together count as a 'person'. Scott has pointed out that the features of the divine essence and the features of a personal property, when grouped together, satisfy the list of features required for something to be a person.

This is certainly right, but it raises a different question in my mind. It is true that all the features of the divine essence and the features of a personal property together satisfy the list of features required for something to be a person, but is this a sufficient reason to think that we should group the divine essence and a personal property together in the first place?

I think the answer is no. This situation here seems to be like this. Suppose that some x is an F if it exemplifies ten particular features. So something is an F if it exemplifies all ten of those features. Further, let's call the first five of those features 'feature-group A', and let's call the last five of those features 'feature-group B'. Thus, something will be an F if it exemplifies both feature-group A and feature-group B.

But now suppose that some x exemplifies feature-group A (but not feature-group B), while some y exemplifies feature-group B (but not feature-group A). If we group x and y together, then x + y together exemplify feature-group A and feature-group B. But does x + y then count as an F? I think the answer is no.

Let me try to explain with an example. Suppose that a chemist comes up with way to combine paper and gelatin to make gel-paper. This gel-paper would be a neat kind of paper that basically has all the features of paper, and all the features of gelatin. You can write on it and file it and so forth for all the things you do with paper, but you can also eat it and flavor it and so forth for all the things you do with gelatin.

But now suppose that I throw the following items into a pile: a piece of wadded up paper, an old glass marble, a brick, and a lump of gelatin. I could show you my pile, group the paper and gelatin together, and then say: 'look, that exemplifies all the features of paper, and all the features of gelatin, and that's the list of features required for something to be gel-paper. I have gel-paper!'

If I said that, you would probably look at me like I was crazy, and you might pull out a piece of gel-paper and show me the real stuff. When we compare the two, it becomes clear that even though my little heap of gelatin and paper exemplifies all the features required for something to be gel-paper, my heap is not gel-paper.

The reason is that in my pile, the paper and the gelatin are just grouped together in an arbitrary way. They are grouped together arbitrarily because I could just as easily group the brick and the gelatin, the brick and the marble, and so forth. In the pile, there is no reason that one constituent should be grouped with this constituent rather than that constituent, so any groupings are just arbitrary.

The only way I could legitimately group my paper and the gelatin together is if they were related to each other in a way stronger than they would be related to the other constituents of the pile. For example, if there were some kind of glue sticking between the paper and the gelatin but no glue sticking between the paper and the other constituents in the pile, then the paper would be related to the gelatin in a way stronger than it would be related to the other constituents.

But given that there is no such glue or anything like that in my pile, it is entirely arbitrary to group the paper and the gelatin. So in my pile, my 'gel-paper' is just an arbitrary grouping of paper-features and gelatin-features. By contrast, in the real gel-paper, the paper-features and the gelatin-features are not just arbitrarily grouped together. They have some stronger kind of relation holding them together.

So I cannot say some x and some y count as an F just because they jointly exemplify all the features required for something to be an F. My gelatin and my paper jointly exemplify all the features required for something to be gel-paper, but they are not gel-paper. I can only say that x and y count as an F if they are joined together in some way in the first place.

Likewise in the case of the trinity. Just because the divine essence and a personal property jointly exemplify all the features required to be a person, it does not follow that the divine essence + a personal property counts as a person, just as it does not follow that my piece of paper and lump of gelatin count as gel-paper. It is true that all the features of the divine essence and all the features of a personal property together exemplify the list of features required for something to be a person.

The fact that the divine essence and a personal property together exemplify all the features required for something to be a person is only relevant if the essence and property are joined or fused together in the first place. Otherwise, it's just two entities (neither entity needs to be a res or an object) that, when grouped together arbitrarily, exemplify feature-group A and feature-group B separately.

Obviously, the answer here is to establish a unity-making relation between the divine essence and a personal property. That's what Henry and Scotus need to do to get me past this little bump. But this is precisely the bit that I'm having trouble being persuaded of.

Wednesday, August 22, 2007

A 15th century commentary on Scotus, Ord. 1.5.2.un. nn. 126-128

A 15th century commentary on Scotus: Peter Tartaretus (Pierre Tartaret). This is his exposition of the second difficulty Scotus brings up in the Ordinatio 1.5.2.un. nn. 126-128. You can find a translation of this bit from Scotus, as well as my thoughts on it, in my last post.

[From Peter's Lucidissima Commentaria, bk. 1, dist. 5, quest. 2, art. 3, p. 202 of the 1583 Venetiis edition.]

The second difficulty is that although the divine essence is not called potentiality, at least the relation ought to be called an act. This is argued as follows. What pertains to an act pertains to distinguishing, or this pertains to it insofar as it is an act. But a relation in the divinity distinguishes, so it pertains to that relation insofar as it is an act. Therefore, there will still be something that [operates in the divinity] as potentiality or as matter.

The major [premise] is clear from the Philosopher: to distinguish is [an act] of form, just as distinguishing pertains to acts.

In response, Scotus makes a distinction. An act is, as will be clear enough in distinction 27, quidditative or personal. A quidditative act is that by which something is said to be quidditatively, and in this way the divine essence is a quidditative act because the Father is God by deity, and the Son is God by deity. A personal act is that by which something is said to be, in a certain way, incommunicable, just as filiation is that by which the Son is the Son, and Socrateity is that by which Socrates is formally called Socrates. Similarly, paternity in the divine Father is that by which the Father is the Father. Joining this distinction together, we get this: although the personal, or hypostatic, relation of origin is called a personal act, or a hypostatic act, nevertheless the divine essence is a quidditative act.

It is argued [against this] that an act of origin is said to be an act of essence, for an act which distinguishes is an act of that which does not distinguish if both concur in the constitution of something. A relation of origin distinguishes, and the divine essence is not distinguished in a person, so therefore a relation of origin will be an act of the divine essence. It follows that the essence will not be a quidditative act, it will rather be quasi matter, just as Henry said.

Scotus responds that when it comes to an act which distinguishes, or that which distinguishes an act, that act can be understood in two ways. In the first way, it can be understood as that thing which does not itself distinguish but is enumerated and divided, just as the humanity in Socrates, which does not itself distinguish, is distinguished, divided, and enumerated into its singulars. In the second way, it can be understood as that which does not itself distinguish, but is not divided, enumerated, or multipiled into its many supposita, and this is the case for the divine essence.

To the form of the argument, Scotus concedes the major [premise]. When it comes to that which itself does not distinguish but is enumerated, divided, and multiplied, then what distinguishes is an act of that. But Scotus denies this for that which does not distinguish, is not enumerated, nor is divided, as is the case for the divine essence. Otherwise, Scotus concedes the major [premise] in creatures. The argument is clear there, because a created nature is multiplied into singulars. But this is denied in the divinity because the divine nature is not divided, nor is it multiplied.

Tuesday, August 21, 2007

Scotus on the generic view, Ord. 1.5.2.un. nn. 126-128

Scotus, Ordinatio 1.5.2.un. nn. 126-128 [Vat. 4: 72-73].

(Note that when Scotus says 'relation' here, he means a personal property. Although Scotus is famous for toying with the idea that the personal properties might be absolute (monadic) properties, here he is sticking to the traditional claim that the personal properties are just the relations of being a father, being a son, and being a spirit.)
[n. 126] The second difficulty concerns how the the relation but not the essence could be a person distinguisher, since a relation does not have the nature of an act, and distinguishing is an act, according to Metaphysics 7.

[n. 127] I concede that the relation is a personal act, not a quidditative act, because it distinguishes personally and not quidditatively. However, the essence is a quidditative act and a quidditative distinguisher. The quidditative act is simply perfect, since it is infinite, but the personal act is not in this way formally infinite of itself.

[n. 128] If you say that "a distinguishing act is an act of that which does not distinguish", this is false unless that which does not distinguish is distinguished by a distinguishing act, just as in creatures. Humanity is distinguished in Socrates and Plato by a and b, and for this reason the distinguishing act – even individually – is an act of that which does not distinguish, because that distinguishing act distinguishes this nature, which does not [itself] distinguish. But this is not the case here, because the personal property does not distinguish the essence, nor does it contract or determine it.
According to Scotus, the divine essence establishes what kind of thing the persons are, while the personal properties establish which person each of the persons are. To explain this, he makes a distinction between quidditative acts and personal acts. The divine essence is a quidditative act, and the personal properties are personal acts.

A quidditative act occurs when something exemplifies a quidditative property. A quidditative property is a kind-maker in the sense that it is a property K which, if exemplified by some x, is sufficient for x to belong to K-kind. For example, the property being human is a property which, if exemplified by some x, is a sufficient condition for x to belong to human-kind (that is, it is sufficient for x to be human). Likewise, the divine essence is a quidditative property for the divine persons, so it is the property which, when exemplified by each divine person, makes each person divine.

Similarly, a personal act occurs when something exemplifies a personal property. In this context, a personal property is a person-maker in the sense that it is a property P which, if exemplified by some x, is sufficient for x to be person P. For example, the property being Socrates is a property which, if exemplified by some x, is sufficient for x to be the person Socrates. Likewise, in the trinitiy, the personal properties being a father, being a son, and being a spirit are personal properties, so they are the properties that, when exemplified individually, make each person a particular person – Father, Son, or Spirit.

But there is more to the story. Part of what it means to be a person is to be a particular person, and so part of the function of a personal act is to distinguish one person from another. How exactly does this work? It's tempting to think the personal properties are like individual properties, haecceities, in that they individuate quidditative properties. This would fit nicely with Scotus's position that in the natural world, quidditative properties are individuated by individual properties.

Consider Socrates and Plato. They each exemplify the quidditative property being human, and this is what makes them human. Yet even though they both exemplify the same quidditative property, they are not one and the same human. When we count Socrates and Plato, we count two humans. The property being human must be divided, as it were, so that it can be exemplified separately by Socrates and Plato, since if it were not divided for Socrates and Plato, there would only be one quidditative property shared between them, and then we'd only count one human, not two.

According to Scotus, individual properties divide or individuate quidditative properties by making them individual for those entities who exemplify them. For example, the property being Socrates individuates the property being human for Socrates, and the property being Plato individuates the property being human for Plato. At the end of the day, after the property being human has been individuated for Socrates and Plato, the tally reads two humans: one individuated in Socrates, and the other individuated in Plato.

As I said a moment ago, it is tempting to think the divine personal properties individuate the divine essence in this manner. Indeed, this would distinguish the divine persons from each other because it would make each of them a distinct divine person, just as individual properties make Socrates and Plato distinct humans.

But Scotus thinks the personal properties do not individuate the divine essence into separate exemplifications in each divine person. The Father, for example, does not exemplify one instance of deity, while the Son exemplifies another instance of deity. On the contrary, each divine person exemplifies one and the same divine essence. In the trinity, then, the personal properties do not distinguish or divide the divine essence in any way.

So what do the personal properties distinguish, if not instances of quidditative properties? In the case of the divine persons, there are only two options: the divine essence and the personal properties. We'd already seen that Scotus rules out the divine essence: the personal properties do not individuate the divine essence. That leaves only the personal properties. In other words, the only thing the personal properties can distinguish are themselves.

For Scotus, one personal property is distinct from another personal property because there is something intrinsic to it that prevents it from being compatible with any other personal property. In Scotus's terminology, the properties are repugnant to each other, for they each include something in their definition that prevents them from being compatible with any other personal property.

In Scotus's picture of the divine persons then, a personal act can only be an act that distinguishes one personal property from another. If Scotus thinks the personal properties do not distinguish or individuate the divine essence, it follows straightforwarldy that they cannot distinguish the divine essence. The only other thing they can distinguish in the trinity are the personal properties, so they simply distinguish themselves: one personal property is not any other personal property.

All this makes for a puzzling picture. In the natural order, individual properties individuate quidditative properties into distinct instances of those quidditative properties. Socrates' and Plato's personal properties, for example, individuate the property being human into two instances of being human – that is, the personal properties makes two humans. Consequently, there is a one-to-one correspondance between the number of individual properties and the number of instances for quidditative properties. For Socrates and Plato, there are two personal properties, and there are two instances of humanity.

If there were no one-to-one correspondance, that is, if quidditative properties were not individuated, it would be difficult to count how many humans there are. The reason is that the only things we count two of are those things that individually belong to each of the two things – those things there are two of. If something is shared between two things, then there's not two of those, there's only one, so we'd only count one shared thing.

Consider Socrates and Plato again. If we want to count two humans, we need two instances of humanity, not one shared instance. If there were only one shared humanity, the only things there'd be two of are Socrates' and Plato's personal properties, and then we could only count two personal properties. We could not count two humanities because there wouldn't be two of those, there'd only be one.

Scotus's picture of the trinity seems to lead to the same point. If the personal properties do not individuate the divine essence for the divine persons, then what are we counting? On Scotus's view, the divine essence is numerically one, it is not multiplied or divided for each person. If we try to count three somethings there, it seems that all we could count are personal properties. If you asked quid tres?, it appears that we should only say three personal properties.

Sunday, August 12, 2007

More on Scotus vs. Henry 2

Yet another post on this Scotus vs. Henry thing. I think I finally figured out what I didn't understand in my last post. Here's my exposition of Scotus's argument in n. 73, from his Ordinatio 1.5.2.un.

On Henry's view, the divine essence plays a causal role in producing the Son. For Henry, of course, the divine essence plays the role of material cause, but for our purposes, it can be any causal role. So let's say that the divine essence plays some causal role, call it C, in producing the Son. When the Father produces the Son, then, C applies to the divine essence as it is in the Father. But, Scotus asks, is C a universal feature of every person that possesses the divine essence, or is it only a feature of a particular person (namely the Father)?

Consider, as an analogy, the property being intelligent. This property necessarily entails certain causal powers (such as the power to think) that every intelligent individual will possess. For example, if Socrates has the property being intelligent, then Socrates can think, if Plato has the property being intelligent, then Plato can think, and if any other individual possesses the property being intelligent, it too can think. But the property being intelligent can also entail other causal powers that will not necessarily be true of every individual that possesses the property being intelligent. For example, it requires intelligence to shop at the Athenian market, but not every intelligent individual can shop at the Athenian market. After all, there could be a possible world with no Athenian market, but there cannot be a possible world where intelligent individuals cannot think.

Scotus wants to know whether C – namely, the causal role that the divine essence plays in producing the Son – is like the causal role that being intelligent plays in thinking, or whether it's like the causal role that being intelligent plays in shopping at the Athenian market. In other words, does the divine essence necessarily entail C for every individual that possesses the divine essence, or is C true only of a particular individual (such as the Father) who possesses the divine essence?

If C is of the first sort, then we obviously end up with a circular explanation for the Son's production. For causal roles of the first sort, it follows that every individual who possesses the divine essence will play (or at least have the power to play) the causal role C in producing the Son, and since the Father, Son, and Spirit all possess the divine essence, C will necessarily be true of each of them. But if that's the case, then the Father, Son, and Spirit will (or can) play the causal role C in producing the Son, and that leaves us with the consequence that the Son is (or can be) caused, in the manner of C, by all three persons, and thus the Son is (or can be) caused by himself.

To avoid such a circular consequence, we are therefore pressured to take C as the second sort of causal role, namely the kind of causal role that is only true for a particular person who possesses the divine essence, namely the Father. In other words, if we want to say the divine plays some causal role in producing the Son, then if we want to avoid explaining the Son's production in a circular way, we need to say that the divine essence plays this causal role only when it exists in the Father.

But then the question is, why would it only play some causal role in the Father? What reason could be given for this? It is tempting to say, as many medieval theologians like Henry and Scotus do say, that the Father has the power to produce the Son in virtue of possessing the divine essence. We can use human intelligence again as an example. If a human person has the property being intelligent, then we could say she has the power to think in virtue of possessing that property. That seem sensible enough. But unfortunately, this doesn't get us very far. If the Father has the power to produce the Son in virtue of possessing the divine essence, then any other person which possessed the divine essence would have that power too, so why don't the other persons who possess the divine essence produce a Son? This is, in effect, the same question posed in a different way. Why is it that the divine essence plays a causal role in producing the Son only in the Father? Henry and Scotus each have strategies for getting around this, but I won't discuss those. My purpose here is just to figure out the force of Scotus's argument in n. 73.

So, Scotus has essentially shown that if the divine essence plays some causal role in producing the Son, then it can play that role only in the Father. From this, Scotus further argues that if the divine essence can play this causal role only when it is in the Father, and if this causal role is being a material cause as Henry has it, then it follows that the Son is produced in the Father. Imagine for a moment that the divine essence is a lump. If the divine essence is the material cause, as Henry holds, then this means the divine essence is the lump-in-which the Son is produced. But since the lump can only be the lump-in-which when it is in the Father, that is, when it is the Father's lump, then the Son can only be produced in the Father's lump. As Scotus sees it, then, Henry's view ends up with the consequence that the Son is produced in the Father, and that's a consequence Scotus sees as inappropriate.

I must admit that I find it hard to see why this should be a problem for Henry. If the divine essence is a substratum for the Father and the Son in the way that a lump of clay is the substratum of a statue, then of course the Son would be produced in that lump (the divine essence), and that lump would be in the Father. So it would not at all be inappropriate, on Henry's view, for this to be the case.

Wednesday, August 8, 2007

More on Scotus vs. Henry

More on Scotus's argument against Henry's substratum view. Here's my translation of the relevant bits.
[Ord. 1.5.2.un, my translation]

[n. 72] Further, a second argument against the principle claim [of Henry that the divine essence is the material term of the Son's production]. It is necessary to assign some being to the [divine] essence as it is that from which the Son is generated, because to be the causal basis of some true entity, in whatever genus of causal basis, does not belong to anything except a real entity.

[n. 73] Therefore, I ask what sort of being belongs to the essence as it is 'that from which the Son is generated' (by an impression). Either it is (a) precisely being ad se, which is the being of the essence qua essence – and then the Son is from the essence qua essence, and in this way [he is from] the three persons – or (b) it belongs to it to be in some subsistent [person]....

[The rest of n. 73 and n. 74 are translated in a previous post.]
The bit I have trouble with is (a).

Let me explain how I'm understanding this. For Henry, the divine essence is the material term of production, while for Scotus, it is the formal term. Consider a clay statue. We have a sculptor and a lump of clay. The sculptor takes the lump of clay, and fashions it into a statue by giving it the form of a statue. The sculptor is the producer, the form of the statue is the formal term of production, the clay is the material term of production, and the whole statue is the primary/adequate term of production. Saying the clay is the material term of production basically means that the clay is the stuff the statue is produced from. So the statue is related to the clay by a produced-from relation.

By analogy to the divine case, Henry thinks we can say the Son is produced-from the divine essence in the sense that the divine essence is the subject in which the Son is instantiated, similar to the way that a lump of clay is the subject in which a statue is instantiated. To put it in terms of causality, the divine essence plays the role of material cause in the Son's production.

Scotus says: okay, so if the divine essence plays the role of material cause, then are we talking about the divine essence as it is in all three persons, or as it is in the Father? If we go with the first option, then it follows that the divine essence as it is in all three persons will play a causal role in the Son's production. Scotus doesn't explain why this is a problem, but presumably, the consequence is that if the divine essence as it is in all three persons plays a causal role in producing the Son, then the divine essence as it is in the Son will play a causal role in producing the Son, and that's circular.

But what's the problem with this? I don't really see why this is a problem. We're just talking about material causality here, not efficient causality here. If the divine essence were the efficient cause (or played some efficient role) in the Son's production, then it makes sense. If the divine essence were the efficient cause of the Son, and if the divine essence were a part of the Son, then the Son would be efficiently caused by (a part of) himself. But we're not talking about efficient causality. We're talking about material causality, and I don't see why this is in any way odd.

Consider a clay statue. A statue is produced-from a lump of clay, and so the clay plays the role of material cause, but nobody would say the statue is produced-from a part of itself (or, if we allowed such a phrase, it wouldn't be problematic, for we all know what we mean: the statue is just produced from the lump of clay). On Scotus's reading, if the Son is produced-from the divine essence, then the Son is produced-from a part of himself, and that's circular. But if that's right, then this should be true for clay statues as well: if a statue is produced-from a lump of clay, then the statue is produced-from a part of itself, and that's circular. But surely we don't take the case of the clay statue as problematic, so why should we take the divine case as problematic?

I'm tempted to think Scotus is equivocating between efficient and material causality. As I hope is clear from what I've just written, Scotus's argument seems completely sensible if we think 'produced from' implies efficient causality, but it makes no sense if we think 'produced from' implies material causality. So it would be natural to think Scotus was just thinking about efficient causality when he wrote this bit.

But unfortunately, he seems to suggest in n. 72 that this argument applies, no matter what kind of causal role the divine essence plays (as Scotus puts it, 'in whatever genus of causal basis'). So apparently, Scotus thinks this works for any genus of causal role: efficient, material, formal, or final. So I'm stumped here. I can't seem to see why it would be a problem for material causality.

Besides, if Scotus thinks the argument works for any genus of causal basis, then it applies to the formal cause, and Scotus thinks the divine essence is the formal cause of the Son. So his argument would apply to his own theory just as much as to Henry's, if it were successful. Yep, I'm stumped.

Perhaps we could appeal to time here. A statue is produced from clay, but the clay is not a part of the statue until after the statue is produced, so it is not true, strictly speaking, that a statue is produced from a part of itself. But in the divine cause, time is not an issue. The Son is supposed to be eternally produced, so whenever it is true that the divine essence is a part of the Father, it is a part of the Son too. Thus, if the Son is produced-from (via material causality) the divine essence as it is in the Father, it is also true that the Son is produced-from (via material causality) the divine essence as it is in the Son. Then it would be circular.

But does that even work? I don't know. Something about all this seems very hocus pocus.

Tuesday, August 7, 2007

The divine attributes

The topic of the divine attributes focuses on the nature of God's properties. A divine attribute is a property of God. So what is the list of properties that belong to 'the divine attributes'?

Traditionally, such properties must be compatible with being God. The properties being good or being wise or being awesome are compatible with being God, but most of wouldn't say the property being addicted to child pornography is. So when it comes to the list of divine attributes, we're just including God-compatible properties on the list, attributes like 'good', 'wise', 'perfect', and so forth. (Which properties, I would suggest, are not PC-compatible, but only Mac-compatible).

In any case, this question has its locus classicus in Aquinas's discussion of the divine attributes. Aquinas's position runs as follows.

When we observe and ponder the world, we notice two types of entities: objects, and the properties of those objects. Consider a red car. The car is the object, but its red color is a property of the car. The car has other properties too, like being fast, weighing such and such, and so forth.

Now, would we say that an object is identical with any of its properties? Is a car identical -- the very same thing as -- its property of being red? Aquinas, at least, would say no. The car can still be the car without being red, so being a car and being red don't seem to entirely be the same thing. Likewise, the property being red can apply to things other than cars (like lamp shades and sunsets), so being red and being a car don't seem to be entirely the same thing.

Aquinas concludes from this that objects are different from their properties. Objects have properties, but objects are not identical to their properties.

Now apply this to properties that we include in our list of divine attributes. Take 'wise' for example. If I say 'Dan is wise', I seem to be talking about an object (Dan), and a property (being wise). On Aquinas's analysis, Dan is not identical to his property of being wise. After all, Dan can exist without being wise, and being wise can apply to objects other than Dan. Thus, Dan has wisdom, but Dan is not identical to his wisdom.

But what about when we say 'God is wise'? As Aquinas sees it, the situation is different. God does not have wisdom, God is wisdom. God is completely identical to his properties (i.e., the divine attributes). In other words, in the created realm, objects and properties are different, but in the divine case, the object in question (God) is identical to his properties (the divine attributes).

(Notice that this entails that God's properties will always be true of God, but this is not the case for creatures. Dan can exist without wisdom, but God cannot exist without wisdom. Dan can acquire or lose (if, perhaps, he had a lobotomy) wisdom, but God cannot acquire or lose wisdom.)

Aquinas, famously, argues that our minds work in accordance with the created order, so when I say something like 'John is wise', my mind understands John as having the property of being wise. So statements like 'John is wise' are understood as 'John has the property of being wise'. Aquinas thinks that our minds think this way because the world is that way, and God made our minds to understand the created order.

But God is not like the created world. God does not have the property of being wise. He is wisdom. As Aquinas sees it, we cannot understand it, because any statement like 'God is wise' gets translated in our brains as 'God has wisdom'. Thus, we always don't really understand the true meaning of our statements about God. We cannot then, understand what it is to be wisdom. We can only understand what it is to have wisdom.

(The same goes for existence. We can't understand what it is for God to be existence, we can only understand what it means to have existence. But existence is slightly more complicated than the other divine attributes like 'wisdom', because it is not clear that existence should be construed as a property. In any case, I won't discuss existence more here.)

As one author puts it, for Aquinas, we know that our words refer to God, but we don't understand the true sense of those words, since we can't understand what it is for God to be wisdom. We know our God-talk has reference, but it doesn't have sense. Aquinas is, in this respect, firmly in the negative theology tradition.

Thursday, August 2, 2007

Some thoughts on the divine essence as a substratum

I've been thinking some more about this little debate between Henry and Scotus on the divine essence as a substratum for the persons. Here are some of my (rough) abstract thoughts on the matter. Maybe later I'll clean this up if I get a chance.

In book 7 of De Trinitate, Augustine ponders two ways to explain how the divine essence is common to all three divine persons: as a shared genus or species, as for example if three people shared one form of humanity, or as a common material or substratum, as for example if three statues were made from the same lump of gold. I will call the first of these views the 'generic view', and the second the 'substratum view'.

One might wonder if it really makes much difference which of these we choose. After all, on both views, the divine essence is just something shared by the three divine persons, so the relations of identity and distinction work out in more or less the same way. It doesn't matter much whether we say the divine essence is a common 'form' or a common 'substratum', because either way, the divine essence is common to all three persons.

Here I will try to show that there is one reason why it might make a difference which of these views we choose. Although the identity and distinction relations work out in more or less the same way for each of these views, when we consider how the Father produces the Son, the generic and substratum views have to part ways. And then, as I will argue below, it turns out that the substratum view (but not the generic view) pressures us to say that the divine essence belongs more to the Father than to the Son, and that sounds a lot like subordinationism.

1. The generic and substratum views

There are many ways to compare and contrast the generic and substratum views, but for our purposes, I would like to think of this in terms of instantiation. From this angle, the difference between the generic and substratum views is that on the generic view, the divine persons instantiate the divine essence, while on the substratum view, the divine essence instantiates the divine persons. I will say more about this in a moment, but first I want to explain what I mean by 'instantiation'.

Instantiation, as I understand it here, is a technical way of talking about 'where' properties occur. The idea is that properties don't just float around, detached from objects. They occur in objects. The property being red, for example, is not just a free-floating patch of color. Rather, it occurs in particular objects such as walls and fast cars. These objects are 'where' the property being red occurs. In the technical jargon, each occurrence of the property being red is an instance of being red, and the object where a property instance occurs is said to instantiate the property (or, to use the passive tense, the property is instantiated by the object). So when I say that some x instantiates F, I mean that the property F occurs in the object x.

This description of instantiation might seem a little vague, and it is, deliberately so. The reason is that I don't intend it to depend very much on particular ontological commitments. Instantiated properties can be described in a variety of ways, e.g., they can be tropes, immanent universals, qualities, and so forth, but we can still talk about these in terms of instantiation. The same goes for the objects which instantiate properties. These can also be understood in various ways – as bundles of tropes or properties, an independent substance, a substratum, and so forth – and we can still talk about this in terms of instantiation.

To maintain the serviceability of instantiation, all we need to do is meet two conditions. First, for any x that instantiates F and any y that does not instantiate F, F is ontologically connected to x but not to y. If we consider a red wall and a green water tower, the color red is ontologically connected to the wall in a way it is not connected to the water tower. This 'ontological connection' can be construed in many ways, but however we explain it, we need to maintain that the color red is ontologically connected to the wall rather than the water tower

Second, if some x instantiates F, then F does not instantiate x. In other words, the relation must be asymmetric. The color red comes to exist in walls and fast cars, but fast cars and walls do not come to exist in the color red. Again, there are many ways to explain how or why the ontological connection of objects and their properties is asymmetric, but so long as we can maintain that walls instantiate the color red and not the other way around, the notion of instantiation is serviceable. (If we deny the asymmetry of x and F, then the substratum and generic views are not different at all, because the divine essence is just the overlap of the three persons.)

So how does this apply to the trinity? As I said at the beginning of this section, on the substratum view, the divine essence instantiates the divine persons. Thus, the divine essence is 'where', as it were, the divine persons occur, so we end up with one object which instantiates three persons. To borrow from Augustine's analogy, if you can imagine three statues occurring simultaneously in the same lump of gold, you have a good idea of the substratum view.

The generic view, on the other hand, maintains the opposite position: the divine persons instantiate the divine essence. Thus, the divine persons are 'where', as it were, the divine essence occurs, so we end up with three objects that instantiate (we could also use the word 'exemplify' here to mean the same thing) one divine essence. If you can imagine a single statue occurring simultaneously in three lumps of clay (perhaps in the way that a polyadic property might occur simultaneously in multiple objects), you have a fair picture of the generic view.

As I suggested early on, it may not seem like there is any significant difference between the substratum and generic views. Indeed, on both views, the divine essence plays the role of something common to all three. It is just the 'overlap' of the persons. Whether we say that this 'overlap' occurs because the divine essence instantiates the persons or because the persons instantiate the divine essence, the identity and distinction relations turn out to be exactly the same. However, in the next section I will argue that although this is true for the identity and distinction relations, when we consider how these two views explain the production of the Son, the substratum and generic views have to part ways.

2. The production of the Son

Thus far I have suggested that we can define the generic and substratum views in terms of instantiation: on the generic view, the divine persons instantiate the divine essence, while on the substratum view, the divine essence instantiates the persons. Each of these views entails a different statement about how the divine essence is included in a divine person.

On the generic view, the divine essence is instantiated by the divine persons. Putting it this way might make it sound like we can first have a divine person which then instantiates the divine essence, and thus there is the possibility that a divine person could exist without instantiating the divine essence. But this needn't by the case. We could insist that a divine person cannot fail to instantiate the divine essence, just as a human cannot fail to instantiate the property being human. Thus, on the generic view, some x is a divine person only when it instantiates the divine essence.

On the substratum view, things are reversed: a divine person is instantiated by the divine essence. Again, although putting it this way might make it sound like the divine essence could exist without exemplifying a person, this needn't be the case. We could insist that the divine essence cannot fail to instantiate at least one divine person, just as some clay cannot fail to instantiate at least one lump. Thus, on the substratum view, the divine essence is a divine person only when it instantiates a person, just as a lump of clay is only a statue when it instantiates a statue.

These two different conceptions of what makes a person determines how we explain the production of the Son. On the generic view, some x is a divine person only when it instantiates the divine essence, so producing the Son entails that however we explain the inner machinery of production, the end result must be that the Father causes some x (the Son) to instantiate the divine essence. On the substratum view, the divine essence is a person only when it instantiates a person, so producing another person entails that however we explain the inner machinery of production, the end result must be that the Father causes the divine essence to instantiate a person. In short, on the generic view, the Father must cause the divine essence to come to exist in the Son, while on the substratum view, the Father must cause the Son to come to exist in the divine essence.

It is at this point that we start to encounter the difference between the substratum and generic views, but to arrive at a clear picture of this, we need to consider the causal relations entailed by production. Of course, there are many different causal relations that we could discuss here. For example, we could talk about how a stone falling is causally related to the physical forces that act upon the stone, or how the fire that burned down the neighbor's house was causally related to the circumstances leading to the fire.

But these are not the sort of causal relations that make a difference for the production of the Son. The reason is that when the Father produces the Son, there is nothing there to cause the Son except the Father. The Father must therefore be the sole and sufficient causal source for the Son. No matter how we construe the Father's causal or productive role in producing the Son, the end result will therefore be that the Son is directly emitted or elicited – produced – from the Father. The Son will come 'out of', if you will, the Father. There is nowhere else the Son could come from.

Thus, when we say that the Son is produced from the Father, this produced-from relation must have the sense of 'coming directly from'. The Son will not be related to any causal source other than the Father. There is not an external artisan who shapes the Son out of the Father, there is no physical force that causes the Son to spring forth from the Father, nor is there anything else. The Son comes from the Father, plain and simple.

But here is where the generic and substratum views start to part ways. On the generic view, the Son is produced from the Father but not from the divine essence, while on the substratum view, it's hard to see how the Son could fail to be produced from the divine essence. On the substratum view, the Son is produced from the Father, but the divine essence just is the substance of the Father. The divine essence just is that object 'where' the Father occurs. If the Son comes from the Father in the sense that the Son directly issues forth from the Father, then surely the Son comes from that object 'where' the Father occurs, namely the divine essence. It would be odd to think that the Son issued forth from the Father, but not from the object where the Father is instantiated.

Consider a clay statue. Suppose that, perhaps in some other worldly dimension, clay statues can produce products in such a way that the products directly issue forth from the producing statues. In this scenario, it's hard to see how such a product could fail to be produced from the clay. If the clay just is the substance of the statue, the object 'where' the statue occurs, then surely the product will issue forth from the clay. It would be odd to think that such a product would issue forth from the statue, but not from the clay where the statue is instantiated.

On the generic view, however, the Son does not come from the divine essence in this way. Rather, the Son directly issues forth from the Father. Of course, on the generic view, the Father instantiates the divine essence, so the Father is 'where' the divine essence occurs, but nothing turns on this. Consider the clay statue again. A product issuing forth from a clay statue would come from the clay in which the statue is instantiated. It would be odd to think the product came from the statue but not from the clay where the statue is instantiated.

As it turns out then, the substratum and generic views begin to part ways when we consider where the Son issues forth from for each of these views. On the substratum view, the Son would issue forth directly from the divine essence, while on the generic view, the Son would issue forth directly from the Father. In the next section, I will argue that this particular difference pressures the substratum view to distance itself even further from the generic view.

3. The substratum view and the divine essence

Saying that the Son comes from the divine essence pressures the substratum view to say that the divine essence belongs more to the Father than it does to the Son. Otherwise, it turns out that the Son is produced from himself. The idea here is that if the divine essence is equally the object which instantiates each divine person, then it follows that the Son is produced from the object which instantiates himself. And since the Son just is that object (in which the Son is instantiated), the Son is produced from himself, and that's circular. Let me try to explain this with some analogies.

Imagine if you and I jointly owned a lump of gold which we then cast into a statue. If someone asked what the statue was produced from, I couldn't say it was produced only from my gold, because it's just as much your gold as it is my gold. The statue would be produced from the gold that belongs to each of us. In order to say the statue was produced from my gold, the gold would have to belong more to me than to you.

Or, to use an example closer to the divine case, imagine a possible world where the only material is clay, and the only way such clay can be arranged is statue-wise (in such a world, any bit of clay is a lump and a statue). If you pointed to one of these statues and asked me what it was produced from, I couldn't say it was produced from the lump's clay any more than the statue's clay, since the clay is equally the lump's and the statue's. In order to say the statue was produced from the lump's clay, the clay would have to belong more to the lump than to the statue.

Similarly, if the divine essence is equally the substance of each divine person in the sense that it is the object which instantiates each divine person equally, then we cannot say the Son is produced from the Father's substance, because the divine essence is just as much the Son's substance as it is the Father's. To put this another way, whenever it is true that the divine essence is the substance of the Father, it is also true that the divine essence is the substance of the Son. Consequently, whenever it is true that Son is produced from the Father's substance, it is also true that the Son is produced from his own substance. The only way the Son could be produced from the Father's substance (and not his own substance), is if the divine essence belongs more to the Father than to the Son and/or Spirit.

The conclusion here is that if we want to say that the divine essence is equally the substance of each divine person, then we have to accept the circular consequence that the Son is produced from the substance of each divine person (and thus, in part from himself). In order to avoid this problem, we are pressured to say the divine essence belongs in some way more to the Father than it does to the Son. Only then will it follow that the Son is produced from the Father's substance rather than from his own.

But if we do that, it seems we end up with a text book case of subordinationism. If the divine essence belongs more to the Father than to the Son and Spirit, then it obviously follows that the Son and Spirit possess the divine essence in a way less than the Father does. Since the Son and Spirit are therefore less than the Father (in whatever way we define 'belonging less to' here), it's hard to see how this sort of thing would not be a case of subordinating the Son and Spirit to the Father. If we hold to the substratum view, we would therefore want to find a way to explain how the divine essence could belong more to the Father than to the Son in such a way that did not amount to subordinationism.