Split Brains and The Godhead

I believe in the Holy Trinity. So I believe that there are three divine persons—Father, Son, and Holy Spirit—and one God. Now the mere claim that there are three of one thing and one of another is logically unproblematic. After all, there is no problem with the claim that, for example, there are three musketeers and one Eiffel Tower. But the Doctrine of the Trinity says more than just that there are three divine persons and one God. It seems to say that each of these three persons is the one God. Thus the Athanasian Creed:

...there is one Person of the Father, another of the Son, and another of the Holy Spirit...the Father is God; the Son is God; the Holy Spirit is God. And yet there are not three Gods, but one God.

There is more to the Doctrine of the Trinity than (1) through (7). (For example, (1) through (7) are silent on Who proceeds from Whom.) Nevertheless, I shall use ‘the Doctrine of the Trinity’—or just ‘the Doctrine’—to refer to the conjunction of (1) through (7). For my only aim is to defend the Doctrine from the charge that it entails a contradiction. And that charge is inspired by (1) through (7).

The charge is easy to motivate. Taken most straightforwardly and naturally (and given (1)), (2) implies that the Father is a person and the Son is a person and the Father is not identical with the Son. From this we get:

The most straightforward and natural reading of (5) entails that the Father is God and the Son is God and the Father is identical with the Son. This implies:

Obviously, (8) and (9) are contradictory. And similar reasoning easily generates contradictory statements about the identity of the Son with the Spirit and of the Spirit with the Father.

I shall defend the Doctrine of the Trinity from the charge that it is contradictory. But before presenting my own arguments, I shall examine two other ways one might try to defend the Doctrine, one involving “relative identity” and the other “social trinitarianism.” My own defense does not require that these familiar defenses fail. But—I shall argue—they do fail. And, in the course of arguing for this, it will become clearer what a successful defense must do.

The claim that the Father is the same God as the Son seems to entail that the Father is identical with the Son. This entailment seems to hold because, in general, A’s being the same F as B seems to entail that A is identical with B.

More carefully, this entailment seems to hold for relations like being the same dog as or being the same tree as or being the same human as or being the same God as. But it does not seem to hold for relations like being the same shape as or being the same size as or being the same height as. We are happy to accept, for example, that A is the same height as B while denying that A is identical with B.

Indeed, A’s being the same height as B not only fails to imply that A is identical with B; it also fails to imply that A is a height and that B is a height. A’s being the same dog as B, however, seems to imply not only that A is identical with B but also that A is a dog and that B is a dog. From now on, when I make a claim about A’s being the same F as B, I shall have in mind only those cases where this entails that A is an F and B is an F. And in those cases, it seems obvious that A’s being the same F as B entails that A is identical with B. [1]

At least, it seems obvious to me. Defenders of relative identity, however, deny just this entailment. They typically insist that, for example, A’s being the same tree as B does not imply that A is identical with B. (Paradigmatic relative identity theorists insist on this because, they say, there is no such thing as absolute identity to be entailed; more on this below.) Relative identity is most closely associated with Peter Geach (1972, 238-249 and 1973). But it may not have originated with him. Geach himself claims to find it in Aquinas (Geach, 1961, 118). Moreover, Richard Cartwright reports finding relative identity endorsed by both Anselm and the Eleventh Council of Toledo (Cartwright, 1987, 193).

Whatever its provenance, relative identity promises to free the Doctrine from contradiction. For relative identity tells us that the Father’s being the same God as the Son does not entail that the Father is identical with the Son. [2] If this is right, then of course the Doctrine does not imply the contradiction noted in the previous section.

This theological benefit notwithstanding, I think we should reject relative identity. To begin to see why, note that John Perry (1970, 185) compares the view that identity is relative to the thesis that being a left-handed brother of does not entail being a brother of. That thesis seems flatly false. And if it is not false, then we have no idea what the relation of being a left-handed brother of is supposed to be. Similarly, the claim that, for example, being the same dog as does not entail being the same as (i.e., being identical with) is either false or renders being the same dog as unintelligible. That is the first objection to relative identity.

Believers in relative identity do not typically think that something special about, say, trees precludes an analysis of being the same tree as in terms of being a tree and being the same as. Rather, they think that all identities—being the same tree as, being the same electron as, and so on—are relative and so fail to entail being the same as. For they typically deny that there is any such thing as being the same as to be entailed. In other words, and as noted above, they typically deny there is any such thing as absolute (i.e., classical, non-relative, plain old) identity.

Insofar as relative identity implies that there is no absolute identity, then it is false. For surely there is absolute identity. Surely there is something that is identical with itself. This is my second objection to relative identity. Of course, this is no objection to relative identity on its own terms. Geach would not take the rejection of absolute identity to be a reductio of his view; rather, he takes it to be his central insight. Nevertheless, I think this second objection is decisive. And, at any rate, it is the principal reason I (and I think many others) reject the view that all identity is relative. So I conclude that no defense of the Doctrine of the Trinity is successful if it requires denying that there is something that is identical with itself.

But suppose someone claimed only that identity was sometimes relative. So suppose he conceded that there is such a thing as absolute identity and there is something that is identical with itself. But suppose he went on to insist that not every identity is absolute; some identities are relative. Suppose he said, for example, that while being the same tree as entails being the same as, being the same God as does not.

This attenuated version of relative identity is immune to my second objection. And this attenuated version may seem more attractive than full-throttle relative identity, especially if it postulates relative identity only in very unusual cases, cases where absolute identity might seem more trouble than it is worth. For example, one might claim that the logic of absolute identity—which is good enough for everyday purposes—“breaks down at the quantum level” or “breaks down when it comes to the very nature of God.”

Peter van Inwagen presents something like an attenuated version of relative identity in defending the Doctrine of the Trinity. He takes a relative identity reading of the relevant trinitarian claims. But he is careful to add: “...I shall assume neither that classical identity exists nor that it does not exist” (1995, 241). So van Inwagen’s solution, which invokes relative identity, is intended to be consistent with (but not entail) the existence of absolute identity. And so it is meant to be consistent with the claim that, for example, being the same tree as is analyzed as being a tree and being the same as (i.e., being identical with).

When first motivating the charge of contradiction, I said that, read most naturally and straightforwardly, claim (5)—the Father is the same God as the Son—entails that the Father is identical with the Son. Now those who (like Geach) insist that all identity is relative will disagree. They will object that the most natural and straightforward reading of (5) does not entail the Father’s identity with the Son. For they would say that the relative identity reading of (5) is the most natural and straightforward. After all, they will insist, in every paradigm case of “identity,” we have only one or another kind of relative identity, never absolute identity. And so Geach can, by his own lights, plausibly maintain that his reading of the Doctrine is the default one.

Van Inwagen endorses a relative identity reading of (5). Yet he cannot agree with Geach that that reading of (5) is the most natural and straightforward. For if—like van Inwagen—we do not deny that there is such a thing as absolute identity, we should say that the following is a perfectly intelligible reading of (5): The Father is God and the Son is God and the Father is identical with the Son. And we should surely say that that reading—again, assuming we do not reject absolute identity out of hand—is the most natural and straightforward.

The defender of attenuated relative identity cannot plausibly maintain that her reading is the default one. Rather, she recommends that we take a less-than-most-natural reading. But once we open the door to less-than-most-natural glosses on (1) through (7), there is—absent further argument—no reason to accept the relative identity gloss as opposed to some other.

Now perhaps the defender of attenuated relative identity will reply that no other gloss is as compelling as hers. Fair enough. But in order to make that point, she will have to do more than present her reading of the Doctrine; she’ll have to say something about how it is better than its competitors. And this shows that van Inwagen’s approach faces a hurdle that Geach’s does not. For, as we have seen, Geach can claim that his reading of the Doctrine is the default reading; nothing similar can plausibly be claimed of any of the “glosses,” including the gloss suggested by attenuated relative identity.

As noted above, some object that alleged kind-relative identity relations are unintelligible. But at least Geach can reply that, definitions of those relations aside, we are acquainted with kind-relative identity all the time. With respect to the relativity of identity, Geach would say, being the same God as is just like being the same tree as.

The attenuated relative identity theorist says that identity is relative only with respect to the Trinity—or only in cases far removed from common experience. So she cannot say that being the same God as is anything like being the same tree as. And so being the same God as, besides being undefined, turns out to be unlike paradigm cases of being the same F as, all of which involve absolute identity. In light of this, the objection that relative identity relations are unintelligible is even more compelling when made against attenuated relative identity than when made against Geach’s view.

Geach would say that, because there is no such thing as being the same as, being the same God as does not entail it. This would render the relation of being the same God as mysterious enough. But I think the mystery is increased if there is indeed the relation of being the same as, but being the same God as is allegedly too weak to entail it. After all, given the existence of being the same as, surely there is some relation that entails it and being God. If being the same God as is not that relation, then which relation is it? And what are we supposed to call the relation that entails absolute sameness and otherwise looks for all the world like it is being the same God as? Again, the charge that relative identity relations are unintelligible gets a leg up if relative identity is attenuated.

Attenuating relative identity exacerbates worries about the intelligibility of the relative identity relations, which worries were serious enough to begin with. This in turn makes it harder for attenuated relative identity to answer adequately the first question asked about it. That question was why—if we are to depart from the most natural and straightforward reading of the Doctrine—should we depart in the relative identity way. For this particular departure, of course, can be no more attractive than it is intelligible.

I say that the attenuated relative identity theorist cannot overcome these challenges. She cannot make the relevant relative identity relations intelligible and so she cannot persuade us that the right reading of the Doctrine invokes them. So I conclude that we should reject her defense of the Doctrine. [3]

My conclusion is based, in part, on the idea that if attenuated relative identity relations are unintelligible, a defense of the Doctrine in terms of such relations is unacceptable. The final move open to the advocate of this defense is to challenge that idea. So I close my discussion of attenuated relative identity by considering the following speech:

There is such a thing as absolute identity. So, to avoid contradiction, we must depart from the most natural and straightforward reading of some part of the Doctrine. Let’s depart from the most natural reading of claims invoking “being the same God as.” I depart by saying that such claims assert a relation—call it ‘relation X’—between the divine persons that does not entail absolute identity. I add that, whatever X is, it doesn’t result in a heretical reading of the Doctrine. But that is all I add. Note, specifically, that I don’t purport to make X “intelligible.” Now for some nomenclature: I will call X a ‘relative identity relation’ and I will call my view ‘attenuated relative identity’.

The view expressed in this speech is immune to my objections above. But it does not save attenuated relative identity. For, it’s “nomenclature” notwithstanding, this speech does not contain an attenuated relative identity defense of the Doctrine at all. Indeed, it contains no defense of any sort. Instead, it merely expresses confidence that there is some (non-heretical) defense or other. I think this confidence is praiseworthy. Nevertheless, to express such confidence is not the same thing as defending the Doctrine. (That’s why someone can, without contradicting himself, say he has no defense of the Doctrine but is confident that some defense or other is out there.) And it is a defense we are after in this paper.

Social trinitarianism emphasizes the interpersonal (or social) relationships among the divine persons. Social trinitarianism has many contemporary advocates. Moreover, its advocates credit it with a venerable history, finding its roots in the Cappadocian Fathers, including Gregory of Nyssa and Gregory Nazianzus (Morris, 1986, 212; Plantinga, 1989, 32; Brown, 1989, 55).

Its recent popularity and rich history notwithstanding, social trinitarianism is sometimes accused of falling into tritheism, one of the two principal heresies regarding the Trinity. Tritheism, obviously enough, says that there are three Gods. Tritheism does not do justice to claims (5), (6), and (7) of the Doctrine, claims like the Father is the same God as the Son. (The other principal heresy here is modalism, which denies that there really are three distinct divine persons. Modalism does not do justice to claims (2), (3), and (4) of the Doctrine, claims like the Father is not the same person as the Son.) As noted above, I want to defend the Doctrine from the charge that it is contradictory; let me now add that I won’t count as successful any heretical defense.

It is hard to know how to evaluate the charge that social trinitarianism is tritheistic. This is primarily because social trinitarianism itself is hard to define. Sometimes its defenders seem to equate it with the utterly unobjectionable claim that there really are three persons in the Trinity. Thus Cornelius Plantinga, Jr., says:

So the first defense of social trinitarianism against the charge of tritheism is this: to say that Father, Son, and Spirit are the names of distinct persons in a full sense of person scarcely makes one a tritheist. (1989, 34)

The most common objection raised against defenders of the social model for the Trinity like myself is that it must inevitably lead to tritheism, given its understanding of Father, Son, and Holy Spirit as three distinct persons. (1989, 48)

These comments (and others like them; see also Morris, 1986, 212-213) make social trinitarianism sound equivalent to the thesis that the Doctrine of the Trinity is true but modalism is false. There is nothing in this thesis to suggest tritheism, unless one has already cast one’s lot with the heretics by claiming that modalism and tritheism are the only options.

Sometimes, however, social trinitarians do seem tritheistic, their protests to the contrary notwithstanding. For example, C. Stephen Layman (1988) argues that because each divine person is itself a distinct substance, there are three divine substances. And Thomas Morris at least leans in this direction when he glosses claims about three persons as claims about three “divine beings” (1986, 217-218). Moreover, Richard Swinburne claims that “there are three and only three Gods,” saying that this way of putting things avoids the “traditional terminology” (1988, 234).

Given our purposes in this paper, it does not matter what exactly social trinitarianism amounts to or whether that theory—once clearly defined—is tritheistic. Instead, I want only to explore whether something like social trinitarianism offers an orthodox defense of the Doctrine against the charge of contradiction. Thus social trinitarianism is of interest to us only if it (or something like it) can block the charge that (1) through (7) lead to contradiction.

Let’s start by considering a version of social trinitarianism no one explicitly endorses. (Discussing this version will set up the main point I want to make about social trinitarianism as it is actually defended.) So consider absolutely pure social trinitarianism. It is “pure” because it claims that the unity among the divine persons is purely social. It claims that harmonious social relationships exhaust that unity. Let’s assume that perfect love is sufficient for every such social relationship. Thus pure social trinitarianism asserts that:

Pure social trinitarianism’s reading of (5) is surely consistent with (2), the claim that the Father and the Son are not the same person. And since the pure social trinitarian will read (6) and (7) along the same lines as (5), on her reading the Doctrine of the Trinity is definitely not contradictory.

Pure social trinitarianism renders the Doctrine non-contradictory. Nevertheless, we should reject it. For it is tritheistic. To begin to see why I say this, note that the pure theory implies that A’s being the same God as B is analyzed as A’s being divine, B’s being divine, and A and B’s loving each other perfectly.[4] This understanding of being the same God as implies that two or three or ten humans, when able to love each other perfectly, will be one in the same way that the Father and the Son are one. The relata will of course differ: in the one case it is humans, in the other divine persons. But the relation—the oneness, the unity—will be the same. Surely, something has gone wrong. Surely, the sense in which the divine persons are one is stronger than the sense in which, once freed from sin and its effects, you and I shall be one.[5]

Moreover, I believe that the Father loves each of us perfectly. Given the pure theory, the only thing keeping us from being one with Him in just the sense that the Son is one with Him is a failing of love on our part, a failing due to sin. But, again, something has gone wrong. For surely it is false that each of the redeemed in Heaven will enjoy exactly the same unity with the Father as that enjoyed by the Son.

And imagine Apollo, Zeus, and Ares resolving their differences, making amends, mending fences and so finally loving each other perfectly. What you are imagining, I insist, is a species of tritheism. Yet we have the relationship of perfectly loving holding among divine relata. Given the pure theory, each of these divine beings would “be the same God as” each of the others. That is, Apollo would be the same God as Zeus (and so on) in exactly the sense in which the pure theory says that the Father is the same God as the Son (and so on). And I think this shows that the pure theory is tritheistic. (Cf. Leftow, 1999, 232)

As noted above, no actual social trinitarian is pure. For example, Cornelius Plantinga, Jr. thinks that not only love unifies the divine persons, but also (among other things) the impossibility of each person’s existing without the other two (1989, 37). Presumably, there is no limit to the “unifying factors” the social trinitarian can add, just so long as she remains true to her central claims: modalism is false and the divine persons love each other perfectly. Indeed, she can even add that the divine persons are unified by being the same God as. Looked at in this light, it’s hard to see how any orthodox believer could fail to be a sullied social trinitarian.

Social trinitarians need not—must not—defend the pure theory. So they must allow that more than love unites the divine persons. But social trinitarianism as such does not say what this more is. As a result, social trinitarianism is not really a theory about what (1) through (7) mean.[6] As I noted above, it is hard to say exactly what social trinitarianism is. But I think that the following is in the ballpark and at least makes impure social trinitarianism more than the claim that modalism is false and the divine persons love each other. Social trinitarianism is the view that it is important, for theological and pastoral purposes, to articulate and emphasize the love and other interpersonal relationships among the persons of the Trinity. Thus understood, I hope it is clear that whatever its insights, social trinitarianism is not the place to look for a way to block the charge of contradiction.

At one time, a treatment for serious attacks of epilepsy was brain bisection or commisuratomy. (This may seem like a change of subject, but its relevance to the Doctrine will become clear below.) Brain bisection is the severing of the patient’s corpus callosum, a band of nerve fibers through which the brain hemispheres communicate directly with each other. Cutting the corpus callosum limits the spread of a seizure to one half of the brain. Yet it has a side effect, well known and beloved among philosophers of personal identity. After brain bisection, a distinct “sphere of consciousness” seems to be correlated with each brain hemisphere.

Moreover, the evidence indicates that each sphere has its own ways of getting information. Here are some examples. The left half of the patient’s visual field is accessible only to the sphere of consciousness associated with the right hemisphere (and vice versa). Similarly, the right-hemisphere-sphere gets tactile input from the left hand (and conversely). And each sphere of consciousness enjoys its very own nostril.

The evidence for these and similar claims is the strange behavior that can be elicited, in experimental situations, from patients who have had the surgery. Here is a representative account:

What is flashed to the right half of the visual field, or felt unseen by the right hand, can be reported verbally. [Typically, the left hemisphere of the brain controls speech.] What is flashed to the left half field or felt by the left hand cannot be [verbally] reported, though if the word “hat” is flashed on the left, the left hand will retrieve a hat from a group of concealed objects if the person is told to pick out what he has seen. At the same time he will insist verbally that he saw nothing. Or, if two different words are flashed to the two half fields (e.g., “pencil” and “toothbrush”) and the individual is told to retrieve the corresponding object from beneath a screen, with both hands, then the hands will search the collection of objects independently, the right hand picking up the pencil and discarding it while the left hand searches for it, and the left hand similarly rejecting the toothbrush which the right hand lights upon with satisfaction. (Nagel, 1975, 232)

Moreover, “if a split-brain monkey gets hold of a peanut with both hands, the result is sometimes a tug of war” (Nagel, 1975, 231). And a physician told me that, when he extended his hand to a split-brain patient, the patient responded by reaching out to shake with his right hand—and also with his left! (For detailed experimental data, see Gazzaniga, 1970.)

I agree with the general consensus that brain bisection results in “two spheres of consciousness.” But I add that brain bisection does not produce new people. When one person lies down on the table for the surgery, that same person (and she alone) gets back up. I think this is the right thing to say, in part, because of reflecting on the possibility of a temporarily disabled corpus callosum.

…been equipped with some device that can block communication between my hemispheres. Since this device is connected to my eyebrows, it is under my control. By raising an eyebrow I can divide my mind. In each half of my divided mind I can then, by lowering an eyebrow, reunite my mind. (1984, 246)

Parfit then imagines availing himself of this device while taking a physics exam. He imagines “dividing his mind” so that he can—in one sphere of consciousness and with one hand—work out one way of solving a problem and—in the other sphere and with the other hand—work out another. He “reunites his mind” ten minutes later.

Suppose this were to happen. Then it seems that Parfit—one person—would acquire a novel psychological ability. But it does not seem that he would acquire a novel way of reproducing; nor does it seem that Parfit would, ten minutes after thus reproducing—and with impunity—annihilate one of his recent offspring (or himself). So I conclude that only one person is involved in the case Parfit imagines. And I think that if there is one person in a case of a temporary division of consciousness, then there is one when the division is more lasting. After all, the number of persons involved should be fixed once the division occurs; whether there is (say) one person right now should not be a matter of what will happen in the future. Thus whether a division will be temporary, or instead be long lasting, is irrelevant to the number of persons resulting when it occurs.

…if the patient is permitted to touch things with both hands and smell them with both nostrils, he arrives at a unified idea of what is going on around him and what he is doing, without revealing any left-right inconsistencies in his behavior or attitudes. It seems strange to suggest that we are not in a position to ascribe all those experiences to the same person, just because of some peculiarities about how the integration is achieved. The people who know these patients find it natural to relate to them as single individuals. (Nagel, 1975, 238)

(According to Nagel, the “most notable deviation in ordinary behavior was a patient whose left hand appeared to be somewhat hostile to the patient’s wife” (1975, 233).) Brain bisection is not a philosopher’s fantasy. It really occurs. And we really treat those with split brains as a single person. And I think this is the right thing to do.

Anyone who denies that each split-brain patient is a single person must say that those closest to such patients are deeply confused. And I suppose that if there are two persons associated with each split brain, their friends and family (and the law?) should treat them as two persons, radically altering current practice. But I don’t think anyone would seriously recommend changing our practice in this way. I think all will agree that it is false that we should treat each actual split-brain patient as two persons. Yet the claim that two persons result from bisection implies this falsehood. Thus we have a second reason to reject that claim.

Neither of the considerations just noted, however, is the main reason that I say that brain bisection does not multiply persons. The main reason is that—so I say—each of us is a human organism. And I deny that brain bisection results in two human organisms where once there was one. So, I conclude, brain bisection does not result in two of us where once there was one. Instead, brain bisection divides the consciousness of a single human organism; that is, it divides the consciousness of a single person.

The substance dualist could defend a similar argument. Suppose each of us is a simple, immaterial soul; suppose further, as substance dualists typically do, that each of us (in this life at least) is associated with a particular body and brain. And suppose brain bisection does not bring about a new soul. Then brain bisection would not make a new one of us; it would not produce a new person. Instead, bisecting a brain would split the consciousness of its associated soul, the one soul that was there all along. [7]

I conclude that brain bisection does not give us two persons where before there was one. Instead, one person remains, but with two spheres of consciousness. This conclusion—and my argument for it—is controversial. For I admit that part of my argument relies on one or another disputed assumption about the nature of human persons. And I concede that I have gone along, uncritically, with the received view that brain bisection results in two spheres of consciousness. Thus there are a number of objections one might raise to my claims above. But I don’t need to respond to these objections. For, as we shall see below, all that matters for my purposes is that, in making my controversial claims, I have not contradicted myself.

Perhaps someone will charge that what I say above is in some sense contradictory. Perhaps they will say that I have contradicted some necessary or conceptual truth. Thus one might claim that it is a matter of necessity—or even analyticity—that one person cannot have two spheres of consciousness. I reply that this claim itself is controversial. And so its denial cannot be contradictory in the most straightforward way. It is at least a live philosophical option that a single human person can have two spheres of consciousness. After all, it is at least a live philosophical option that a human person is either an organism or a substantial soul.

Henceforth, when I consider whether something is contradictory, the issue is not whether it merely contradicts a substantive metaphysical thesis. Rather, it is whether it is contradictory in the sense in which the Doctrine of the Trinity is charged with being contradictory. The idea behind that charge, as we saw at the start of this paper, is not merely that the Doctrine contradicts a contentious metaphysical thesis. It is instead (and very roughly) that any clear thinking person can be shown that the Doctrine leads to a formal contradiction. My reading of split brain cases is not thus contradictory. My reading of these cases is philosophically defensible. My reading is a live philosophical option. And, as we shall see, this is all that my defense of the Doctrine requires of my reading.

S’s corpus callosum is severed. S is one person. But she now has two spheres of consciousness, named (for obvious reasons) ‘Lefty’ and ‘Righty’. S decides that she might as well have a little fun with her condition. And so she engages in written correspondence in such a way that if Lefty is involved, Righty is not; and vice versa.

For example, she makes sure that letters sent to her are presented to only one side of her visual field, accessible to only one sphere of consciousness. And she replies to letters read by Lefty with a hand under only Lefty’s control, likewise for letters read by Righty. Moreover, she signs her letters not with her own name, but either as ‘Lefty’ or ‘Righty’, depending of course on the responsible sphere of consciousness. Lefty and Righty take turns on correspondence duty, alternating daily.

You enjoy a lengthy correspondence with S. Because your letter might be received on a day when (for example) Lefty is in charge of correspondence, you can’t assume that anything read by Righty will be available to S as she reads your letter. After all, if S reads the letter “as Lefty,” she won’t read it in the full knowledge of what she previously read “as Righty.” So your letters are to a large extent redundant. Just in case.

In what follows, it is important that, in the story just told, Lefty and Righty take turns corresponding. It is important that, in corresponding with S, you thereby correspond with Lefty or with Righty. And, so I say, this is what happens. But someone might object as follows:

In your story Lefty and Righty do not take turns corresponding. Only persons correspond and Lefty and Righty, according to your story, are not persons. And the way S signs her letters—‘Lefty’, ‘Righty’, ‘S’, ‘Willard van Orman Quine’— is irrelevant to who actually wrote them.

Let me drive this point home with a story of my own: S* has two eyes, Lefty* and Righty*. Sometimes she tapes Righty* shut, reading and responding to letters making use of only Lefty*; she then signs her letters ‘Lefty*’. I hope you agree that this does not imply that sometimes one corresponds with S’s left eye.

My reply begins by returning to some of the evidence for the claim that brain bisection results in two spheres of consciousness. The patient is told to search for whatever is flashed on the screen. On one side the word ‘pencil’ is flashed, on the other ‘toothbrush’. One of the patient’s hands searches for the pencil but not the toothbrush, the other for the toothbrush but not the pencil.

The natural assumption here—the assumption embodied in the claim that brain bisection results in two spheres of consciousness—is that the sphere of consciousness associated with one hemisphere knows that the word ‘pencil’ was flashed on the screen but not that the word ‘toothbrush’ was; the other sphere knows that the word ‘toothbrush’ was flashed, but not ‘pencil’. The natural interpretation of the experimental data is that each sphere knows something the other doesn’t.

Moreover, the spheres could communicate with each other. If the sphere controlling speech shouted out “I saw the word ‘toothbrush’ on the screen,” then both spheres would know that ‘toothbrush’ was flashed on the screen. Indeed, just as Lefty could communicate with Righty by shouting, so Lefty could communicate with Righty by writing. Lefty could use the right hand to write a letter, a letter which is then projected on the side of the visual field accessible to Righty. And if Lefty can write letters to Righty, she can write them to you.

Nothing remotely like the claims just made about Lefty and Righty are true of Lefty* and Righty*. It is false that Lefty*, a mere left eye, knows something that Righty* does not. Nor could Lefty* correspond with Righty*. Lefty* and Righty* were intended to illustrate an objection to my claim that, in corresponding with S, you correspond with Righty or with Lefty. Instead, they illustrate that that objection to my claim fails. Its failure is illustrated by the fact that the case of Lefty* and Righty* is simply not analogous to the case of Lefty and Righty.

We can reinforce the relevance of the disanalogy by noting that no one defends the claim that an eye is a person. So no one would say that Lefty* and Righty* are persons. And anyone who did would be beyond the philosophical pale. But some insist that spheres of consciousness are persons. (We shall see later that some social trinitarians seem to say this.) So some will insist that Lefty and Righty are persons. For reasons noted in Section IV, I disagree with them. But their view is not crazy. They are not beyond the pale. Unlike eyes, spheres of consciousness are at least somewhat person-like.

They are “person-like.” For, as noted above, there seems to be some sense in which spheres of consciousness, again unlike eyes, know things and can even correspond. Indeed, let’s add that S’s corresponding is somehow analyzed in terms of Lefty’s corresponding or Righty’s corresponding. (Or perhaps vice versa.) This analysis is most plausible—and most clearly non-circular—if the sense in which S corresponds is a different sense from the sense in which Lefty or Righty corresponds. So let’s add that. Let’s add that although there is a sense in which S corresponds and a sense in which Lefty corresponds, there is no univocal sense of ‘correspond’ in which both S and Lefty correspond. We can also add that Lefty’s and Righty’s failing to correspond—or to know or to think or to love or to hope or to believe—in the same sense in which S does partly explains why Lefty and Righty fail to be persons.

Recall your lengthy correspondence. The letters from S pile up on your desk. A colleague leafs through them and asks: “Who has been writing to you?” You say: “S; remember her?” He says: “Sure. But”—glancing at the signatures on the letters—“who is Lefty? Who is Righty?” You think to yourself that letters from Lefty just are letters from S, likewise letters from Righty. You think that to write to S is nothing other than to write to Lefty or to write to Righty. You think that to hear from S just would be to hear from Lefty or to hear from Righty. And so you say: “Lefty is S. Righty is S. They are both our friend S.” To clarify that there is just one person (i.e., S) authoring the letters, you add the following:

Read in the most natural and straightforward way, (A) is false. For thus read, (A) implies that Lefty is a person and Righty is a person and Lefty is identical with Righty. Yet neither Lefty nor Righty is a person; each is, instead, a sphere of consciousness; nor is Lefty identical with Righty. Nevertheless, (A) seems like an appropriate thing to say. You may not have told your colleague the whole story; but you haven’t been obscurantist, either. (A) is a pretty good first stab at the situation.

You aren’t sure exactly how to go beyond the first stab. For you aren’t quite sure what more to say about Lefty and Righty. It might help if they were physical objects, like brain hemispheres. But you know that can’t be right. For (we now add to our story) substance dualism is true. So rather than physical objects like brains or brain hemispheres, it is immaterial objects—souls—that have mental properties. You know all this. So you know that each “sphere of consciousness” is not “had” by a brain hemisphere, but by S’s soul (i.e., by S herself). So you can safely rule out the possibility that Lefty and Righty are themselves brain hemispheres. You can likewise rule out the possibility that Lefty and Righty are proper parts of S. For S has no proper parts at all.

S is a soul. So she has physical properties only (at best) in an extended and relational sense. For example, the claim that she is over five feet tall is (at best) a shorthand way of saying that she is appropriately related to a body that is over five feet tall. But not all of S’s properties are thus extrinsic; not all of her properties are relations to her body. Her mental properties are intrinsic.

Because S’s mental properties are intrinsic, it is possible for S to have her mental properties without standing in relations to contingent things outside of herself, to things like her body. This is not to deny that S’s intrinsic mental properties are typically related in important ways to her body. For example, stimulate her body in the right way, and S’s soul will feel pain. Sever the corpus callosum in her body’s brain, and S’s consciousness will be divided. But any relation here is presumably causal and so contingent. So if S is a soul, it is possible for her to feel pain even if disembodied. And if S is a soul, her consciousness could be divided even if no split brain belongs to her.

Let’s alter the story a bit. S has not undergone brain bisection. Indeed, S could not undergo brain bisection. No brain belongs to her. For, we are now imagining, S is a disembodied soul who has never had a body or a brain. Nevertheless, S has a divided consciousness of the sort typically induced by brain bisection. And S is somehow able to communicate with the embodied. She can somehow control a pen so as to write letters. And she can somehow read letters written to her.

You didn’t know anything about S until you saw her ad in the personals. You like what the ad says (no picture, though), so you begin to correspond. Or at least you try to. You are frustrated to find that your letters are sometimes answered by Lefty and sometimes by Righty but—so it seems to you—never by S. (Lefty and Righty presume to speak for S—indeed, they write as if they were S—so you assume that they are secretaries, acting under S’s direction.)

You are annoyed by the—so it seems to you—impersonal nature of this arrangement. And, to make matters worse, apparently Lefty and Righty don’t communicate with each other or with S very well. For example, you tell S in one letter that you like long walks on the beach and fruity rum drinks. But, in a letter S writes (by way of Lefty) several weeks later, she asks if you are a teetotaler. You begin to wonder whether your letters are reaching S at all.

You write several times demanding to correspond with S directly, not via Lefty or Righty. But S (in a letter from Lefty and also in a letter from Righty) replies that you demand the incoherent. Your demand, she says, presupposes that Left and Righty are intermediaries between you and her, intermediaries that can somehow be circumvented. But that presupposition, she continues, is all wrong. Rather, to write to Lefty or to write to Righty just is to write to her. To correspond with her is nothing other than to correspond with Lefty or with Righty. S tells you that your asking to correspond with her but with neither Lefty nor Righty is like her asking to correspond with you but not with your mind.

S realizes that these claims will seem odd to you. So she tries to cast light on them by explaining her somewhat peculiar nature. So she says things like: “I am one immaterial person but two spheres of consciousness.” She is careful to insist that she is not two immaterial persons. And she emphasizes that Lefty and Righty are not merely roles she occupies.

Taken most naturally and straightforwardly, (A) is false; for thus taken it entails that Lefty is a person, Righty is a person, and Lefty is identical with Righty. But, in light of the story I have just told, I think there is a fairly natural reading of (A) that comes out true.

Above I tried to motivate the way in which (A) seems true. But let me say more. Note that if you want a relationship with S, you have a relationship with Lefty or Righty. To interact with Lefty or Righty is to interact with S. And for S to love you just is for Lefty to love you or for Righty to love you. Likewise for S’s hating you or talking to you or issuing a command to you or… For the purposes of friendship and interaction—indeed, for all practical purposes—Lefty is the same person (that is, S) as Righty.

Moreover, when (for example) S issues a command and Righty issues that command, the command is not issued twice over. For S’s commanding does not duplicate Righty’s commanding. Rather, S’s acting in any way at all is somehow analyzed as either Righty’s acting or Lefty’s acting. (Or perhaps it goes the other way, and Righty and Lefty’s acting are analyzed in terms of S’s acting.) Thus Lefty’s acting is the same person’s acting—S’s acting—as is Righty’s acting. Lefty is the same actor—that is, S—as Righty. (That is, Lefty is the same actor as Righty in the sense of ‘actor’ in which we have one actor: S. In another sense of ‘actor’, we have two: Lefty and Righty. But in no univocal sense of ‘actor’ do we have three.)

Similarly, for S to believe a proposition (for example, the proposition that the word ‘pencil’ flashed on the screen) just is for Righty to believe that proposition or for Lefty to believe it. Righty’s believing something is the same person’s believing it—S’s believing it—as is Lefty’s believing something. Righty is the same believer—when by ‘believer’ we mean person who believes—as is Lefty.

For reasons like those just noted, I conclude that (A)—Lefty is the same person as Righty—is an appropriate and fairly direct way to express a truth. Indeed, I don’t know of any better way to express that truth. We can get at it by multiplying examples like those above. Yet all those examples seem to support or indicate or gesture at a peculiar relationship between Lefty and Righty, a relationship that I can’t better express than by saying that they are, in some very important sense, the same person.

Given the story I have told, the following is true taken straightforwardly and naturally. Lefty and Righty are both spheres of consciousness and:

In light of the above, I say that our story about disembodied S shows that (A) and (B) are non-contradictory when appropriately understood. [8] Moreover, I say, (A) and (B) are not obscurantist or misleading. They do as good a job as any pair of claims could at getting at what is going on in the story. And if all that is right, then I suggest we should—because of the obvious analogies—say something similar about the following two claims. I say we should conclude that the following need be neither contradictory nor obscurantist and misleading:

Indeed, I think the analogy between the story of S and the Doctrine of the Trinity is even stronger than (A) and (B) and (5) and (2) suggest. To begin to see why I say this, note that orthodoxy requires us to say that the three divine persons are not three substances. What, then, are they? In presenting his social theory of the Trinity, Plantinga says:

[A social theory of the Trinity] must have Father, Son, and Spirit as distinct centers of knowledge, will, love, and action. Since each of these capacities requires consciousness, it follows that, on this sort of theory, Father, Son, and Spirit would be viewed as distinct centers of consciousness or, in short, as persons in some full sense of that term. (1989, 22; emphasis in original)

Similarly, Morris, another social trinitarian, often uses ‘centers of consciousness’ as a synonym for ‘persons’ (1986, 210-218).

The social trinitarian cannot accuse us of modalism if we defend the claim that there are three divine persons in what she takes to be the relevant sense of ‘person’. And so even modalism’s most emphatic opponents should have no objection to our glossing (2) as:

Of course, something similar can be said about Lefty and Righty. So let’s say it:

As already noted, the three divine persons are not three divine substances. There is only one such substance, God. And the claim that each divine person is God is standardly taken to be equivalent to the claim that each is this one divine substance. With this in mind, we can endorse:

The divine persons are not substances. But S is. Being a soul, she is an immaterial substance. Thus we could recast (A) as:

There is a striking analogy between (A*) and (B*) and (5*) and (2*). And (A*) and (B*), as I have explained them, are not contradictory. So I conclude that we are not compelled to think that (5*) and (2*) are contradictory; likewise for the Doctrine as a whole.

Someone might object that I don’t really endorse (A*), since I reject its most natural and straightforward reading, since I deny the claim that Lefty is a substance and Righty is a substance and Lefty is identical with Righty. And so, someone might suspect, I don’t really endorse (5*).

In reply, return to the point that opened this paper. If each of the Doctrine’s claims is read in the most straightforward and natural way possible, the Doctrine is contradictory. So, assuming that the Doctrine is non-contradictory, at least some of the Doctrine’s claims should not be read in the most straightforward and natural way possible. There must be a way to endorse the Doctrine while rejecting the most natural and straightforward reading of at least some of its claims. To borrow language used earlier, one must “do justice” to all those claims, even if one does not endorse each and every claim in the most straightforward way possible. With all this mind, note that the my story “does justice” to the claim that Lefty is the same substance as Righty, although of course I reject that claim’s most natural and straightforward reading.

The story of disembodied S includes something controversial about the nature of human beings. Perhaps the claim that human persons are immaterial souls is false. (I think it is.) And perhaps if it is false, it is necessarily false. And so perhaps the story I told is impossible. And so—one might object—nothing I have said suggests that the Doctrine of the Trinity is possibly true.

If a triune God exists, then this is (presumably) a matter of necessity. And so to show that (1) through (7) are possibly true would be tantamount to showing that they are in fact true. Yet surely we do not need to show that the Doctrine is true to defend it from the charge that it is contradictory. And so surely we do not need to show that the Doctrine is possibly true to defend it from that charge. So although I have not shown that the Doctrine is possibly true, I do think the analogy defended above shows that we are not forced to conclude that the Doctrine is contradictory. As I put a similar point earlier in the paper, the Doctrine of the Trinity is at least a live philosophical option.

I claim only that (A) and (B) and (A*) and (B*) are appropriately analogous to the Doctrine of the Trinity. I do not claim to have presented a theory of the Trinity. I have not defended a gloss or account or analysis of ‘being the same God as’ as it is used in formulating the Doctrine. I do not claim that each divine person is a sphere of consciousness and that God is an immaterial substance akin to S’s soul.

One reason that I don’t claim this is that, even if there were human souls, God and souls would not be kindmates. Of course, God and souls would be alike in being immaterial. But this alone does not make them members of the same kind, lest that kind include, for example, abstract objects as well. And the other obvious way in which God is like a human soul—having mental properties—does not suggest a theory of the nature of God, lest it suggest that all beings with mentality (humans, God, angels, demons, dolphins, dogs) have the same nature.

Similarly, I do not think that reflections on Lefty and Righty yield an analysis of what it is to be a divine person. One reason is that, beyond the sorts of things already said in this paper, I don’t know what a center of consciousness is. And I certainly do not purport to have an informative account of being a center of consciousness that applies univocally to Lefty and to the Father. Nor do I claim to have an account of being a person that applies univocally to S and to the Father. Indeed, I think it is an open question whether there is any such account.[9]

Of course, I must insist that ‘being a person’ does not mean being a center of consciousness in just the way Lefty is a center of consciousness. (For if it did mean that, Lefty would be a person.) But I do say that a human being who has not undergone brain bisection is in some other sense (and only contingently) a single center of consciousness. And just as Lefty is in this way like a non-split-brain human person, so I think Lefty is appropriately analogous to the Father.

Still, we are left wondering what it is to be a person. We are left wondering what it is to be a center of consciousness. We are left wondering whether we are persons in exactly the same sense that the Father is a person. We are left wondering how the way in which a normal, non-split-brain human is a center of consciousness resembles the way in which Lefty is a center of consciousness. There is no clear answer to any of these questions; at least, there is no answer from the logic of identity or the law of non-contradiction or merely knowing how to count. And so believers in the Trinity can happily admit not knowing how to answer these questions. Happier still, we can do so without feigning ignorance about one, three, or identity.

I claim that I have defended the Doctrine from the charge of contradiction. But I also deny having a theory of the nature of the Trinity. Someone might object that I can’t make my defense without such a theory. For, so this objection goes, unless we know exactly how to interpret (1) through (7), we have no right to say that (1) through (7) are non-contradictory. And unless we can rightfully say that, we have no defense against the charge that the Doctrine is contradictory.

I concede that I have not proven that the Doctrine, rightly interpreted, is not contradictory. But such a proof is not the only way to defend the Doctrine from the charge of contradiction. One could, instead, argue that there is no compelling reason to believe the Doctrine is contradictory. It is this sort of defense I have presented.

Let’s return to my defense one last time. So suppose the story I told about disembodied S is true. Even in such a (comparatively) mundane case, mysteries persist. We don’t know the nature of S’s soul, other than its being non-physical and mental. (So we do not know what makes S a different kind of thing from angels and demons and God.) Nor do we know what exactly spheres of consciousness are.

Suppose, again, that the story about disembodied S is true. But suppose further that we (like the Church Fathers) are unfamiliar with brain bisection and its odd effects. Suppose moreover that the ideas of a soul (much less a disembodied one) and of a sphere of consciousness have never occurred to us. And finally suppose that S decides to reveal her nature to us so as to help us interact with her rather than to teach us metaphysics. Then (A) and (B) and (A*) and (B*) would get at something non-contradictory in about as clear and direct a way as we could hope for. Then (A) and (B) and (A*) and (B*) would be in equal measure appropriate and puzzling and—when rightly interpreted—non-contradictory.

And so, I say, it goes for claims (1) through (7). A full-blown theory of the metaphysics of the Trinity would tell us what the divine substance is and what the divine persons—Father, Son, and Holy Spirit—are. I don’t have that. But I do think we have seen enough to conclude that (1) through (7) could be wholly appropriate and puzzling and—when rightly interpreted—non-contradictory. In other words, there is no reason to conclude that they are in fact contradictory. And so the charge of contradiction fails to stick.

Anscombe, G.E.M. and Geach, Peter (1961) Three Philosophers (Ithaca, NY: Cornell University Press).

Brown, David (1989) “Trinitarian Personhood and Individuality” in Feenstra and Plantinga.

Cartwright, Richard (1987) “On the Logical Problem of the Trinity” in Philosophical Essays (Cambridge, MA: MIT Press).

Davis, Stephen T., Kendall S.J., Daniel, and O’Collins S.J., Gerald (eds.) (1999) The Trinity (New York: Oxford University Press).

Feenstra, Ronald J. and Plantinga, Jr., Cornelius (eds.) (1989) Trinity, Incarnation, and Atonement: Philosophical & Theological Essays (Notre Dame, IN: University of Notre Dame Press).

Gazzaniga, Michael (1970) The Bisected Brain (New York: Appleton-Century-Crofts).

Geach, P.T. (1973) “Ontological Relativity and Relative Identity” in Milton K. Munitz (ed.) Logic and Ontology (New York: NYU Press).

Layman, C. Stephen (1988) “Tritheism and the Trinity,” Faith and Philosophy 5: 291-298.

Leftow, Brian (1999) “Anti Social Trinitarianism” in Davis, Kendall, and O’Collins.

Morris, Thomas V. (1986) The Logic of God Incarnate (Ithaca, NY: Cornell University Press).

Nagel, Thomas (1975) “Brain Bisection and the Unity of Consciousness” in John Perry (ed.) Personal Identity (Berkeley, CA: University of California Press). (Originally published in Synthese 22 (1971): 396-413.)

Plantinga, Jr., Cornelius (1989) “Social Trinity and Tritheism” in Feenstra and Plantinga.

Swinburne, Richard (1988) “Could There Be More Than One God?,” Faith and Philosophy 4: 225-241.

Van Inwagen, Peter (1995) “And Yet They Are Not Three Gods But One” in God, Knowledge, and Mystery (Ithaca, NY: Cornell University Press.)

* Thanks to Mike Bergmann, Jeff Brower, Jim Cargile, Tom Crisp, Chuck Mathewes, Mark Murphy, Mike Murray, Mike Rea, Ted Sider, Donald Smith, Thomas Williams, and Dean Zimmerman.

[1] I suppose that we could use (for example) ‘being the same dog as’ to express a relation that holds between A and B if and only if A is a dog and B is a dog and A is heavier than B. If we adopted this usage, then of course ‘A is the same dog as B’ would express a proposition entailing that A and B are dogs but not entailing that A is identical with B. (Indeed, since nothing is heavier than itself, it would entail that A is not identical with B.) But, obviously, this has nothing to do with the question of whether A’s being the same dog as B entails that A is identical with B.

[2] I shall focus on relative identity and (5), the claim that the Father is the same God as the Son. But, of course, each of (2) through (7) is subject to a relative identity reading.

[3] Van Inwagen (1995) meticulously presents the formal properties of some relative identity relations; but I believe this falls short of telling us what those relations are.

Someone might claim that the explicitly stated formal properties are all there is to the relevant relation. That is, someone might claim, for example, that (5) says only that the Father is related to the Son by a relation with the relevant explicitly stated formal properties. But, obviously, one must defend this highly technical gloss on (5). One must give a reason to believe this is the right way to understand (5).

[4] The only other way to take the pure theory is as identifying the relation of being the same God as with the relation of loving each other perfectly. Thus taken, the pure theory implies that, once conformed to the image of Christ, you will be the same God as I. And I shall be the same God as you. And each of us will be the same God as the Father. This is a reductio.

[5] Perhaps some social trinitarians would object. Morris offers the following as partial support of the social theory of the Trinity:

When Jesus...is represented in the gospel of John (17:21) as praying to the Father concerning his disciples and other followers “that they may all be one; even as thou, Father, art in me, and I in thee,” he was surely not asking that there be only a single, solitary Christian. He was asking for unity among numerically distinct individuals, not for numerical identity here, and thus he was implying that he perceived the oneness between himself and the Father not to be that of numerical identity, as that between, say, Cicero and Tully, but rather to be that of some sort of harmonious unity between ontologically distinct individuals. (1986, 209-10).

[6] Of course, some particular social trinitarian may have a theory about what (1) through (7) mean. That’s a different point.

[7] The claim that a soul “explains the unity of consciousness” is inconsistent with a soul’s having two spheres of consciousnesses. But that claim is not an essential part of substance dualism. Substance dualism itself is consistent with one soul’s having more than one sphere of consciousness.

[8] Some might object that there is a contradiction in my story’s claim that disembodied S corresponds with embodied humans. In reply, if it is contradictory for a non-physical thing to interact in these ways with the physical world, then theism itself—with its creator God—is itself contradictory. But theism is not contradictory. And if it were, there would be no point to defending the trinitarian species of theism.

[9] Important early discussions of this question are found in Boethius (Contra Eutychon, III) and Aquinas (Ia. IIae. Q.30 a.1