The definition of constitution

Leftow begins by quoting my most recent formal definition of the constitution relation:

In the application to the Trinity, ‘x’ will designate the divine nature, ‘y’ the trinitarian Persons. Leftow has three main objections to this use of constitution in trinitarian theorizing:

A possible counterexample

1. 1. My definition of constitution is subject to a counterexample. Following Lynn Baker, I gave as an example of constitution an ice cube, constituted by a mass of water molecules. Leftow situates the ice cube in a ‘frozen universe’, and claims that there would then be no ‘causal activity’ such as is required for constitution, and therefore my definition fails.

As Leftow realizes, I don't see this as an example of constitution at all, and therefore not as a counterexample to my definition. If x constitutes y, then x and y must be of different ‘primary kinds’, which will only be true if y has ‘whole classes of causal properties that x would not have had if it had not constituted anything’ (Hasker (Reference Hasker2021a), 526). Leftow's attempt to show that this would be the case in his frozen world is laboured and unconvincing. But without those novel causal properties, there are no distinct ‘primary kinds’ and therefore no constitution – and no counterexample.

The problem of improper parts

1. 2. Leftow claims that there is no viable candidate for the ‘unity condition’ for x and y, the condition which makes us want to say that ‘there is only one thing there’. In the most recent definition, the condition (1*) is stated as ‘x and y have all their parts in common at t’. Since both the divine nature and the trinitarian Persons are mereologically simple, this must be taken to include ‘improper parts’. Leftow, however, claims that there is no coherent way of understanding the notion of improper parts that will serve my trinitarian purpose.

At this point a bit of background is called for. In Baker's definition of constitution, on which mine is modelled, the unity condition is given as ‘x and y are spatially coincident at t’. Baker was giving a definition of material constitution; since neither the divine nature nor the divine Persons are material, it seemed to me that this condition could not serve for the Trinity. (I have since come to doubt this, however; of this more later.) I therefore substituted the requirement for ‘all parts in common’, which also accords with Baker's thinking. Leftow argues that my reliance on improper parts introduces circularity: constitution is defined in terms of improper parts, and improper parts are defined in terms of constitution. This, however, need not be so. Consider the following definition of improper parts:

So far as I can see, this can be combined with the unity condition (1*) without generating circularity. A question that can arise here is, what is the relation between an object and its sole improper part? It might seem that a plausible candidate relation is identity, but this would have the consequence that the constituting object and the constituted object are identical, and if so, the constitution relation would not apply. The answer, in any given case, is first to compare the two putative relata of the constitution relation, to see if they exemplify different primary kinds. If not, then they are identical and the issue of constitution does not arise. If they do exemplify different kinds, then the constituting object is identical with its improper part, and the constituted object is not identical with its proper part but is rather constituted by it. This is nicely illustrated by my example of Simple-Hunk and Simple-Athena. Simple-Hunk is a quantity of material that is similar in most respect to clay, but that is impossible to divide into separate parts. (‘No matter how sharp the knife, that last thread of material connecting the two nearly separated parts will always prove impossible to sever’ (Hasker (Reference Hasker2021b), 558).) Simple-Athena differs from Simple-Hulk in the ways ordinary Athena differs from ordinary Hunk; they clearly will be of different primary kinds. Simple-Hunk, then, will be identical with its sole improper part; Simple-Athena will not be identical with that part but will rather be constituted by it.

Nevertheless, I can't easily dismiss the possibility that, given additional time, Leftow might find a problem with this proposal as well. (It's always a mistake to underestimate his metaphysical ingenuity!) So let's see if there is a way to dispense with improper parts.

Omnipresence and spatial coincidence

I did suggest one possible way to do this. According to the common understanding of the concept of God, both the divine nature and the divine Persons are omnipresent. Furthermore, it is highly plausible that nothing else can be omnipresent in the way God is. If that is so, then perhaps being omnipresent can be understood as a way of being ‘spatially coincident’, and if so Baker's formulation of the unity condition can be left standing. There is, however, a complication here. In my previous response, I interpreted omnipresence as meaning God's ‘complete awareness of everything that exists and occurs at any point in space, and the capacity to act at any point in space’, while leaving it undecided whether God is literally present in space. Leftow plausibly objects that ‘Complete awareness and capacity to act do not suffice for literal spatial location’ (Leftow (Reference Leftow2022), 4). I find upon reflection that I must agree with this. It follows that, if I am to affirm spatial coincidence of nature and persons, I shall have to affirm that God literally occupies space. To forestall what may seem an obvious objection, this does not contradict the doctrine that God is not essentially a spatial being. It may be that, whereas God is not essentially spatial, God is such that, necessarily, if there is space, then all of that space is occupied by God. (Compare: God is not essentially a creator, but necessarily, if there exist concrete beings other than God, then God has created them.)

Some philosophers, however, are reluctant to affirm that God literally occupies space. According to Joshua Hoffmanand Gary Rosenkranz ,

This, however, requires argument, unless of course it rests on a stipulative definition of ‘purely spiritual being’. (The stipulative definition will not be accepted by those who hold that a substantial soul can be spatially extended, such as emergent dualists and some Thomists.) All the more is argument required when they assert, ‘It seems plausible that any physical being is to some extent subject to the laws of nature.’ If the supposedly ‘physical being’ is God, the omnipotent creator of all, why should God's presence in space entail God's subjection to the laws of nature, which exist only because God has ordained them?

In contrast to this, Hud Hudson has argued that a conception of omnipresence according to which God literally occupies space may well be viable (Hudson (Reference Hudson, Flint and Rea2009), 205–211). It will not be the case, on this conception, that part of God is present in each part of space; rather, God is present as a whole at each and every point in space. It would seem, then, that Leftow needs a convincing refutation of Hudson's defence of a ‘literal occupation account’ of omnipresence, if he is to close off this avenue for the constitutional trinitarian.

Leftow, however, has another objection at this point. Omnipresence applies only if there is space, but there need not have been space. Where does this leave my statement of the unity condition in terms of omnipresence? I had suggested that without space, the divine Persons would be omnipresent by default, since there would be no space not occupied by the Persons. Leftow has convinced me, however, that this is not a promising way to go. There is a preferable alternative: the divine nature, and the divine Persons, are necessarily such that they are possibly omnipresent, and this is something that is not true of any other being. Plausibly, only a divine being can be omnipresent in the relevant sense, and therefore only a divine being can be possibly omnipresent. And necessarily, there cannot be more than one divine nature, so unity is guaranteed.

Constitution as requiring conventions

Leftow, however, has yet another arrow in his quiver. He offers a ‘deflationary account’ of constitution, such that

1. 3. In its legitimate instances, the constitution relation is dependent on conventions. But the Trinity is not convention-dependent, so constitution cannot be used to explain the Trinity.

In arguing that constitution depends on conventions, Leftow selected as an example works of art. (He has already taken the relation between Athena and Hunk as a paradigm for constitution.) Leftow claimed that, by making a mark on a rock he could, in virtue of our conventions concerning works of art, produce a ‘Marked Rock’ sculpture. I pointed out that, were Leftow actually to produce a Marked Rock, most people would not agree that it was a work of art. In addition to this, Leftow's way of looking a works of art has other disadvantages. It would mean that we never have any positive reason to suppose that a prehistoric society had produced any works of art, since we don't know enough about them to say that they had the relevant concepts, including the concept of a work of art. (I personally am far more strongly convinced that some cave paintings are works of art, than I am that Leftow's analysis of the concept is the right one.) And on the other hand, in our own society candidates for ‘work of art’ status would be all too numerous. Turn a group of children loose on a rocky hillside with magic markers, and we may soon have dozens of Marked Rocks. Leftow will now have to interview each of the children, to learn which (if any) of them intended to be producing a work of art! Clearly, Leftow's concept is not one we operate with in our ordinary dealings with works of art.

In his latest attack on constitution, Leftow has not pursued his case for works of art as conventional. But it's not clear that he can afford to give in on this point: If at least some works of art are constituted in a way that does not depend on conventions, this undermines his claim that constitution in general requires convention-dependence. Perhaps he now thinks that the fact that a work of art is ‘mind-dependent’, in the sense of being humanly created, is enough to rule it out as an analogy for the Trinity.Footnote ² But why so? As I observed previously, ‘One would rather expect that created things would depend for existence upon their creators, and that an uncreated being would not so depend. Why is this supposed to rule out the application of constitution to the Trinity?’ (Hasker (Reference Hasker2021b), 555).

There is another sort of example often cited as cases of constitution: living creatures, such as animals, are constituted by masses of tissue; for example, a cat is constituted by a mass of cat-tissue. Here we have neither a human creator, nor a convention. Leftow barely takes notice of this possibility; in a footnote, he acknowledges he is committed to the view that ‘there is no such thing as the mass of cells or particles that makes animals up’ (Leftow (Reference Leftow2022), n. 5). Rather, ‘The cells and particles can do their work as pluralities’ (ibid.). This seems inconsistent with the earlier acceptance of Hunk: why should masses of clay be all right, while masses of cat-tissue are ruled out? Nevertheless, excluding such ‘masses’ from one's ontology may be a defensible philosophical position. But it is a controversial position, one the constitutionalist will not accept, so the result is a standoff. And as already noted, a standoff is a win for the constitutionalist, because it means Leftow has not succeeded in his aim of ruling out constitution as a way of thinking about the Trinity.

What can be concluded from this third round of Leftow's assault on constitution? Readers will draw their own conclusions, but I believe a reasonable judgment would be that constitutionalism is still on its feet; it has taken some punches, but it has not gone down for the count. And remember that the constitutionalist, in order to win, has only to avoid defeat. This follows from Leftow's claim to eliminate constitution as a viable way of formulating trinitarian doctrine, a claim which requires him to accept the burden of proof. The constitutionalist, in contrast, makes no claim to ruling out other ways of understanding the relation between the divine nature and the divine Persons. Others are welcome to reject the constitution account, and to offer their preferred alternatives. Strong claims require strong support, so Leftow fails if he does not conclusively refute the constitution account.

Leftow, indeed, does offer an alternative account of the relation between nature and Persons, and that is what we have to examine in the final section of this response.

Identity between Father and Son?

Leftow's account of the Trinity has a striking feature that has not been sufficiently emphasized in earlier discussions, including my own.Footnote ³ Namely this: on his view, all of the Persons of the Trinity are identical. The Father is identical with the Son and the Holy Spirit, the Son is identical with the Father and the Holy Spirit, and the Holy Spirit is identical with the Father and the Son. The Father is ‘God-living-the-Father-life-stream’; the Son is ‘God-living-the-Son-life-stream’, and the Holy Spirit is ‘God-living-the-Spirit-life-stream’. These facts explain why it is appropriate in various contexts to refer to God as Father, Son, or Spirit respectively. It is, nevertheless, the same identical person, God, who is referred to in each of these different ways. This places Leftow's account in conflict with other recent writers on the doctrine. For example, none of the other philosophers and theologians considered in my Hasker (2013) would accept that the Persons of the Trinity are identical with one another. This is also the case, so far as I can tell, with the authors other than Leftow represented in McCall and Rea (Reference McCall and Rea2009). (By no means are all of these authors advocates of Social trinitarianism.) This by itself, to be sure, does not immediately demonstrate that Leftow's views are heterodox; it remains possible, at this point, that there is some satisfactory explanation for this disagreement. I submit, however, that this feature of his view should put us on the alert to consider whether in fact his trinitarian views are orthodox and acceptable.

In bringing out what I take to be a conflict between Leftow's views and standard trinitarian orthodoxy, I have focused on Jesus' words from the Cross: ‘My God, why have you forsaken me?’ (Matthew 27:46, Mark 15:34). I proposed that, on Leftow's view of the matter, what Jesus ought to have said is, ‘Why have I-as-Father forsaken myself-as-Son?’ Leftow has provided a summary of our previous discussion centred on this text, and I need not repeat that summary here. In his most recent salvo, he ‘takes a deep breath’ and sets out at greater length his own approach to this passage, in a way that (for me at least) adds substantial clarity to our previous conversation. In fact, I find myself virtually starting all over again to understand his views on this topic.

An important feature of Leftow's discussion is the distinction between interpreting some words of the biblical text, and giving a metaphysical account of the situation those words represent. I agree that this distinction is important. It would be a mistake to interpret any biblical writer as expressing the full Nicene doctrine of the Trinity, let alone as giving the sort of metaphysical analysis that might be offered by a medieval or modern philosopher. I myself never thought of attributing to Jesus the conception of the divine nature as I understand it, let alone the notion of constitution. And I was mistaken to suppose that Leftow was offering his own conception of a trinitarian Person as a possible interpretation of Jesus' words. I would however add – something I would hope Leftow will not deny – that not just any metaphysical analysis can be offered as giving an account of the situation as portrayed in a given text.Footnote ⁴

I propose to conduct this renewed examination of Leftow's views by posing two questions concerning the biblical text in question. The first question is

1. 1. To what extent does the depiction in this text of the relation between Jesus and the Father constitute a firm basis for our own reflection on the subject?

Surprisingly, Leftow's answer to this question amounts to, ‘Not all that much’. I had supposed that, while Jesus' self-understanding undeniably went through a process of development, that process had reached a stable resolution well before his death, so that his views as expressed then will remain as a fixed point around which our own thoughts on the matter should be developed. Not so, according to Leftow. Things might be as I had supposed, of course, but there are other possibilities. It could be that Jesus' process of reflection was cut short prematurely by his death, as was indeed his entire life. Or it could be that he had reached a point of maximal clarity at some previous time (unfortunately, we are given no indication of when this might have been, or where we should look for the evidence), but for various reasons (including, no doubt, the pain, suffering, and anxiety of the crucifixion), the clarity was lost at the time of the words from the Cross. Since we have no way of knowing which of these possibilities may actually be the case, we are unable to take these words as a fixed, authoritative point of reference for our own views. It is well to know of these thoughts of Jesus, to the extent we are able to ascertain them, but in the overall picture they remain only one data point among others.

But leaving to one side the question of authority, there is another question:

1. 2. To what extent are we able to tell, on the basis of the cry of desolation and other words of Jesus, how he at that time conceived of his own nature and his relationship with the Father?

Once again, Leftow's answer is surprising; it amounts to ‘Not very much’. Jesus was ‘dulled by fatigue, with pain blasting away’. As already noted, he might at an earlier point have achieved deep insight, but now he ‘may not even be able to comprehend what once seemed transparent’. Furthermore Leftow, building on his time-travel story, proposes that Jesus, even if fully aware of his own identity with the Father, might still have addressed the Father using ‘you’, because ‘“You” makes better sense when addressing someone across the room’. Thus, ‘the words of desolation could be exactly as they were even if (improbably) Jesus was then consciously and fully aware of all Trinitarian niceties, and I was right about all of them’. Leftow triumphantly concludes, ‘there is just no way from the biblical text to any claim that Jesus then disagreed with my view’ (Leftow (Reference Leftow2022), 8).

I now have to say that I am absolutely astonished by Leftow's answers to these two questions. It would be difficult for me to overstate how strongly I disagree with those answers. To me, it seems somewhere beyond bizarre for a Christian to take the view that Jesus' own mature understanding of his own nature, and of his relationship with the Father, is just one view on these topics among others that need to be taken into consideration. And the profound agnosticism concerning the interpretation of biblical (and other) texts that is implied by Leftow's treatment of the cry of desolation is inimical both to sound exegesis and to good theology. Particularly egregious is the suggestion that, on the basis of his time-travel scenario, Jesus' understanding when he uttered those words (and, I may add, many, many other words that are recorded in all four Gospels), may have been precisely the opposite of what the words themselves convey. This amounts to taking a remote, speculative possibility to undermine the plain meaning of the biblical text, and to prevent that text from communicating the meaning that, on the basis of all available evidence, it actually bears.Footnote ⁵ This cannot possibly be a sound interpretative procedure.Footnote ⁶ But while I cannot agree with what Leftow says here, I can in a way understand it. If one's aim is to leave a large space open for metaphysical speculation, it is well to keep the meaning of the biblical text as indeterminate as possible.

What then of my own view? I should not have thought of myself as a naïve biblicist, but perhaps in comparison with Leftow that is what I am. I do think the cry of desolation has a clear meaning at least in one respect, though we may be far from plumbing the depths of what was expressed in those words. But one thing about the text is crystal clear: Jesus thought of God, his Father, as a person distinct from himself. Here I do not appeal to any metaphysical niceties that are the province of a philosophical analysis. The idea of a person, a rational subject of cognitive, affective, and volitional experiences, was surely part of Jesus' basic conceptual equipment, as it is for each one of us. And there is abundant evidence, spread throughout all four Gospels, that Jesus thought of his Father as another person. The vast extent of the evidence ought to remove once and for all the temptation, as shown in Leftow's time-travel example, to employ some ingenious metaphysical strategy in order to evade the implications of any single passage. Jesus and the Father are two distinct persons: this is the core assertion of Social trinitarians, both in the proto-Social writings of the early Fathers and in the explicit Social trinitarianism of the twentieth and twenty-first centuries. To be sure, there is more that needs to be said. In particular, Social trinitarians have the task of explaining how, even though they are distinct persons, Father, Son, and Holy Spirit are together just one God. One way of addressing this is to say that each of them is constituted by the one divine nature.