Parfit and Persons
Peter Forrest in his recent book God without the Supernatural: A Defense of Scientific Theism, as its title suggests, attempts to present a version of theism in which appeal to the supernatural is not necessary. (1) Forrest raises many interesting issues in his book, one of which is his view of the afterlife. While he does not specifically refer to Derek Parfit's groundbreaking book Reasons and Persons in his section on the afterlife, it is obvious that Forrest is building his approach around Parfit's innovative theories regarding personal identity. (2) Parfit offers such a myriad of distinctions, arguments, and examples that space does not allow complete description of them all. After surveying the implications of Parfit's central argument about personal identity to Forrest's scientific theism, this paper will evaluate the contributions of Parfit and Forrest.
Parfit raises four interrelated questions about the nature of persons and personal identity: (3)
(1) What is the nature of a person?
(2) What is it that makes a person at two different times one and the same person?
(3) What is necessarily involved in the continued existence of each person over time?
(4) What is in fact involved in the continued existence of each person over time?
Parfit employs his well-known science fiction teletransporter stories to frame these questions about personal identity. In the first story, illustrating simple teletransportation, someone is teletransported to Mars by a scanner which destroys his brain and body, records the exact states of his cells, transmits that information to a replicator on Mars which creates an exact replica of the person on earth. In the second story, illustrating the branch-line case, the same process occurs, but the scanner is defective, creating both a person on Mars and retaining the person on Earth, only to discover that the person on Earth has been fatally harmed and faces imminent death. The replica on Mars offers to pick up where the earthling's life left off. (4)
Parfit provides several terminological distinctions to clarify the discussion of personal identity. He distinguishes two kinds of identity--qualitative identity (exact likeness) and numerical identity (being the same person). He also contrasts psychological connectedness (holding particular direct psychological connections) and psychological continuity (holding overlapping chains of strong connectedness). Parfit argues that personal identity is transitive, but strong connectedness is not. (5)
Parfit initially proposes two explanations for his teletransporter examples. The standard physical criterion understands identity to lie in the spatio-temporal continuity of the object. The psychological criterion defines identity in terms of the continued existence of some purely mental entity. Three versions of the psychological criterion differ over the right kind of cause for psychological continuity: the narrow version (which insists on the normal cause), the wide version (which utilizes any reliable cause), and the widest version (which could have any cause). (6)
The physical and the psychological criteria, however, turn out to be reductionist, according to Parfit, because they make the following two claims, one or both of which a non-reductionist rejects: (7)
(1) that the fact of a person's identity over time just consists in the holding of certain more particular facts, and
(2) that these facts can be described without either presupposing the identity of this person, or explicitly claiming that the experiences in this person's life are had by this person, or even explicitly claiming that this person exists. These facts can be described in an impersonal way.
In the non-reductionist account persons are separately existing entities whose existence is non-branching psychological continuity and connectedness. Personal identity (understood as a separately existing entity) is "what matters," because the entity "owns" various experiences. Personal identity is all-or-nothing or determinate (admitting of a yes or no answer). In the reductionist account persons are not separately existing entities, are not absolute, but are indeterminate. It is relation R--psychological continuity and connectedness and/or continuity with the right kind of cause that "matters" in understanding identity. (8)
Whereas most people might opt for the non-reductionist account, Parfit endorses the reductionist account. Although he acknowledges that persons may exist, Parfit asserts that one could give a complete description of reality in terms of brains and bodies without resort to persons. The description of reality in terms of persons, Parfit argues, is redundant on the physicalist account he offers in terms of brain and body. He believes that use of the words "mental state" is misleading, because a state must be the state of some entity. Since the reductionist account rejects such entities, Parfit interprets experiences in terms of events. (9) Following Hume, Parfit utilizes the analogy of a nation to suggest that just as nations do not exist apart from its citizens, so persons do not exist as separate entities apart from their experiences. The existence of persons is not a deep metaphysical truth, but a heuristic device to interpret certain phenomena. Parfit claims that just as (6) and (8) are not inconsistent, so (3) and (5) are not inconsistent: (10)
(6) A nation's existence just involves the existence of its citizens, living together in certain ways, on its territory.
(8) A nation is an entity that is distinct from its citizens and its territory.
(3) A person's existence just consists in the existence of a brain and body, and the occurrence of a series of interrelated physical and mental events.
(5) A person is an entity that is distinct from a brain and body, and such a series of events.
Bernard Williams' objection to the psychological criterion (11) evokes from Parfit a discussion of the psychological spectrum, the physical spectrum, and the combined spectrum. The psychological spectrum, which is Parfit's reconstruction of Williams' argument, holds the body fixed while the psychological continuity and connectedness range from near (continued normal persistence) to far (a radically different person). Parfit counters with the physical spectrum, in which the psychological continuity and connectedness are held fixed while the body ranges from near (continued normal existence) to far (a radically different body). The combined spectrum ranges from near (continued bodily and psychological continuity) to far (radically discontinuous continuity). This results in something of a Sorites problem; there is no clear place along the spectrum to say when the future person is the same or different. This seems to support Parfit's claim that personal identity is ambiguous and indeterminate, rather than a clear, all-or-nothing entity. (12)
Traditional Theism, Persons, and the Afterlife
There is significant disagreement within traditional theism (13) concerning the nature of persons and their persistence into the afterlife. To answer Parfit's first basic question, three anthropological models predominate, which view persons variously as a trichotomy, dualism, or a psychosomatic unity. In the trichotomy perspective, persons are divided into body, soul, and spirit. In the dualism view, persons consist of a spiritual/soulish aspect and a bodily aspect. In the psychosomatic unity perspective, a person is a unity, an embodied soul/spirit, and various anthropological terms describe the whole person from particular points of view. The psychosomatic unity view appears to have the upper hand in current theological discussions. But from all three perspectives, a person is an embodied soul. The answer to Parfit's second, third, and fourth questions would be that personal identity consists in the persistence of this embodied soul over time.
These three anthropological positions predispose one toward particular positions of pareschatology (the intermediate state) and eschatology. The trichotomous and dualistic views usually anticipate a temporary division of soul/spirit from the body during the intermediate state, to be reunited with the body in the resurrection. The psychosomatic unity perspective does not countenance such a division, proposing instead either the creation of an intermediate body or soul sleep with the body. Most Christians anticipate a bodily resurrection and the transformation into an immortal body at the resurrection, although some less orthodox theologians think of the resurrection as symbolic of the continued immortal existence of the disembodied soul. Some process theologians affirm subjective immortality, in which immortality consists in being remembered in the mind of God.
It is difficult to reconcile the anthropology and eschatology of traditional theism with Parfit's perspective. Almost every theistic anthropology defines personhood in terms of an embodied soul/agent. Belief in a resurrection entails that there will be both physical and psychological continuity between this life and the next. Parfit himself suggests that even those whose bodies have been cremated could be reconstituted in the resurrection, just as the pieces of a watch could be taken apart and later reassembled. (14) But this requires the physical continuity which Parfit denies. Perhaps Parfit could take some comfort in a pareschatology which separates soul from body. In this scenario, the physical spectrum is (temporarily) on the far side, while the psychological spectrum is on the near side. But this is a temporary state which will be corrected in the resurrection. The psychosomatic views, however, insist on an unbroken unity of personal identity to avoid just such a division in personal identity as Parfit imagines. The bodily change which takes place in the resurrection might also seem to support Parfit's view, in that both the physical spectrum and the psychological spectrum move from near to far. But this movement is precisely a relatively continuous entity identified as a soul, spirit, or agent. Even the less orthodox disembodied immortal soul and subjective immorality views require the continued existence of a personal entity (if being a thought counts as being an entity). The resurrection requires not qualitative identity (because the body and mind will experiences significant changes along the combined spectrum), but numerical identity.
Forrest, Persons, and the Afterlife
While Forrest is not opposed to substance dualism, he proposes an anthropology in terms of a functionalist double aspect theory, a version of attribute dualism distinct from both reductive physicalism and from substance dualism. Forrest affirms the positive unity of the mental, that is, that the mental is a single mind rather than a collection of mental states. But he contrasts his view with the Cartesian view of a single mental substance. In Forrest's account, the unity of the mental is not an all-or-nothing affair, but a "messy" unity with degrees of integration and even the possibility of overlapping minds. He speaks of the self only with a small "s", and of consciousness as neither a new category nor even an entity at all. Human consciousness is just a part of Forrest's notion of God as unrestricted consciousness. (15)
Forrest's speculations about the afterlife are central to his theocentric understanding of life, for without the afterlife there would be the unpalatable conclusion that a loving God would have brought many people into the world whose uncompensated suffering outweighed their positive experiences. The afterlife is thus for Forrest a sharing of the divine joy and a compensation for negative experiences in this life. Forrest proposes a number of speculations about the afterlife, ranging from rather physicalist understandings to more mentalistic understandings. These speculations, roughly listed from more physicalistic to more mentalistic, include (a) a brief beatific vision as an effect of a dying brain, (b) the evolutionary development through emergent order of a paradise, (c) the creation of a paradise replica of earth which does not violate the laws of nature, populated with human replicas, (d) the survival of the person in a split universe discontinuous with this world, and (e) a disembodied existence in which the dead live on in the collective minds/brains of the living. (16)
Although Forrest does not refer to Parfit by name, his use of many of Parfit's terms reveals that he is at least aware of Parfit's concerns. Forrest's anthropology is thoroughly consistent with Parfit's reductionism. In particular, the idea of a messy unity of the mental, the possibility of group minds and collective consciousness which blurs the lines of personal identity, and the assertion that the mind is not an all-or-nothing substance or entity parallel Parfit's view. Furthermore, several of Forrest's speculations about the afterlife virtually inculcate Parfit's simple teletransportation and branch line cases. Case (c) involving the paradise populated with human replicas is an eschatological version of the simple teletransportation case. Case (d) of the split universe and case (e) of living on in the collective consciousness of others appear to be similar to Parfit's branch-line case.
A Response to Parfit and Forrest
If Parfit and Forrest are right about anthropology and the afterlife, then the corresponding views in traditional theism are wrong. I shall argue that, though the views of Parfit and Forrest are logically possible, there is sufficient evidence that they are not the case. Personal identity is the persistence of a numerically one agent with a unified consciousness over time.
(1) The Explanatory Power of the Teletransportation Case. The simplest course of action is to dismiss the teletransportation story as science fiction. Parfit's fantastic science fiction anecdotes may be evidence that his position is logically possible, but offers little evidence that it is actual. We have no such teletransporters, scanners, and replicators, and we have no evidence that we ever shall. An argument is stronger when it is based in reality than in science fiction; by analogy to the known rather than by analogy to the unknown. Parfit discounts the advice of W. V. Quine that while "[t]he method of science has its uses in philosophy, . . . I wonder whether the limits of the method are properly heeded. To see what is 'logically required' for sameness of person under unprecedented circumstances is to suggest that words have some logical force beyond what our past needs have invested them with." (17)
Had Parfit found empirical evidence in support of his position, it would have been far stronger than the hypothetical case he offers. He does appeal to the split brain phenomenon in support of his bundle theory, claiming that the two independent hemispheres are functioning as multiple spheres of consciousness. (18) But his hypothetical examples based on split brain research go far beyond actual research--he proposes genuinely split persons. Parfit appears to be profoundly mistaken about the current state of split brain research. While he makes much of the fact that the corpus callosum is severed, he overlooks the fact that other ipsillateral neural pathways remain intact. These remaining neural connections account for how one hemisphere attempts to do the work of the disabled hemisphere. Shown a naked lady, one side will bring about a blushing response, but the subject cannot account for his embarrassment. As Thomas Nagel has argued persuasively, (19) the empirical evidence shows that the result of a split brain is not two persons but one numerical identity. (20)
Even if one allowed Parfit's teletransportation story, however, one could reject its applicability on the grounds that it is not transportation but death. A person shot through space at light speed would be transportation, but the destruction of the person's brain and body is death. The replicator on Mars may produce a clone of the person, but that person is dead as far as this life is concerned. What is on Mars is something of a Stepford wife, not the same person.
(2) The Premature Erasure of the Personal Agent. The accounts of both Parfit and Forrest reject the existence of a self as a separately existing entity or substance. Parfit goes so far as to deny mental states, for mental states are states of something. The self or agent of which the mental states are states of is precisely the entity I want to affirm. One could cite the evidence that is usually cited on behalf of this claim--the "I" derived through the Cartesian method of doubt, the sense of continuity provided through our memory, the introspective evidence and intuitions that we are a self, and our sense of ownership about our own experiences. One could claim, for example, that Galveston beach has many potential experiences to offer, but it doesn't become my experience until I go there and actually experience it myself. Potential experiences "float around," in other words, until they are experienced by an experiencer.
The Humean view of the self as merely a bundle of perceptions fails to do justice to the introspective evidence. While the analogy of a nation is intended to lend support to the doctrine of a self as not a separate entity, one could also think of a nation in terms of its government. Each nation has a chief executive, a cabinet of advisers, and the citizens who both have input into the government and carry out the nation's purposes. Even so, the person has a chief executive to govern (the will/self), a cabinet of advisors to provide counsel (desires, judgments, beliefs), and members (the parts of the body) who both send input and carry out plans. One could also counter the nation analogy with other analogies, such as the Vienna Circle or the Alabama Crimson Tide, whose name and reputation far exceed their membership, and whose ever-changing membership are difficult to capture.
Christine Korsgaard argues effectively (21) that one must begin to address the personal identity question not by considering identity across time, as Parfit does, but what makes one a personal unity now. A person must be a unified agent because of the necessity of conflict resolution and the introspective reality that we deliberate and choose from among various desires from a particular standpoint. Even in split brain cases such as Parfit describes, without a unified agent there would be no way of choosing between the options recommended by the two brain halves, or by various desires. Only a unified personal agent could make such an executive decision.
If one is a personal unity now, then it is merely a small step to Robert Nozick's proposal (22) that personal identity through time consists in a maximal connection from the closest predecessor to the closest available continuer. And, although Nozick is not willing to apply transitivity to personal identity, it seems only another modest step to combine a series of near approximations along the combined spectrum to assert a numerical personal identity across time. Admittedly, such transitivity would not be logically necessary. But it does seem that such a claim could be validated by empirical and psychological evidence. Even in extreme cases of split brains (a severed corpus callosum) or split minds (dissociative phenomena and multiple personality disorder) does not indicate that one person is divided into two persons, but that one person is experiencing unusual phenomena which might affect his/her personality but not the numerical identity of his/her personhood. (23)
Parfit and Forrest, of course, can define the experiencer as the brain, and can offer their stories and speculations on behalf of such a bundle view. Perhaps there is no definitive way to arbitrate between these two intuitions--the self as the focal point of identity and the selfless flow of experiences. It seems very difficult, however, to deny the self without risking self-refutation. Parfit and Forrest might could claim that brains, bodies, or streams of consciousness deny the self, but that seems a very strange claim to make. Surely one seeking the inference to the best explanation, as Forrest claims to be doing, (24) would find a separately existing entity to have more explanatory power in understanding human experience than a mere bundle of perceptions.
(3) The Role of the Afterlife in Forrest's Project. It is understandable that Parfit sees his own view as compatible with the Buddhist denial of the self and the affirmation instead of merely a stream of consciousness. (25) Indeed, it is difficult to speak meaningfully of an afterlife unless it includes the continued existence of an entity, because otherwise it would be the life of another entity rather than the afterlife of the same entity. This is not much of a problem for Parfit, but an indeterminate self is a problem for Forrest's eschatology, because the afterlife plays such a significant role in his account. It is the afterlife that God provides to compensate those who have suffered incommensurately in this life. But if there is no continued numerical identity of an entity, theodicy problems are multiplied instead of resolved. Person P suffers but P clones, P replicas, or split P's get the compensation.
Forrest's speculations (c) of the replica persons in a replica world, (d) of an afterlife in a discontinuous split universe, and (e) of survival in the minds/brains of the living do not provide the requisite compensation because the one who suffers is not the one who is compensated. Option (e) seems to be something of a pyramid scheme (not unlike Social Security) in which the well-being of one generation is cast on the shoulders of the next generation. Nor does option (a) with its brief beatific moment seem to compensate adequately for a life of suffering. Of course, such a dramatic experience is to be measured qualitatively rather than quantitatively or temporally, but it is difficult to imagine how one brief moment could provide the requisite compensation for a Holocaust victim. So whereas Parfit's anthropology may help Forrest to maintain a non-supernatural anthropology, it does violence to his eschatology and thus his entire project.
(4) How to Be a Non-Reductionist. Parfit states that one can be a non-reductionist by denying either or both of reductionist theses (1) and (2). I propose that both theses be rejected. Thesis (1) should be rejected because personal identity consists in more than merely holding certain particular facts. Profound amnesiacs might forget all facts about themselves but still possess numerical identity with who they were before the amnesia. The answer to question (2), then, (concerning what it is that makes a person at two different times one and the same person) is not holding certain facts but possessing numerical identity.
Thesis (2) should be rejected because all three of the reductionist denials listed in it are unconvincing. That one could account for the experiences of life without reference to the numerical identity of a person or even without the claim that a person exists seems very strange, as odd as explaining how a nail got hammered without reference to the hammer or how the sounds got recorded without reference to the tape recorder. More precisely, it is difficult to account for such phenomena without resort to one who hammers using the hammer intentionally and one who records using the tape player intentionally. To use Aristotle's example, the stick moves the rock, and the hand moves the stick, but it is the personal agent who moves the hand. Of course, the reductionist could reply that the brain accounts for these experiences without resort to a soul or agent. But they can do so only through an impoverished impersonal account which neglects the introspective evidence of our ownership of experiences. We feel keenly that our experiences are not merely a passing stream, but something we own. It was our first date, our senior prom, our wedding, not just an event we passively observed. One could develop a parallel impersonal account, of course, but not one which does justice to the introspective evidence. If one must choose between the personal and impersonal accounts, the introspective evidence tilts the scales decisively in favor of the personal account unless there is significant evidence to the contrary. Introspection does not guarantee a unified personhood, but it is the best evidence we have.
1. Peter Forrest, God without the Supernatural: A Defense of Scientific Theism, Cornell Studies in the Philosophy of Religion, ed. William Alston (Ithaca: Cornell University Press, 1996), 127. Forrest is perhaps best known as the theistic apologist who convinced naturalist David Armstrong that naturalism was conceptually incoherent.
2. Derek Parfit, Reasons and Persons (Oxford: Clarendon Press, 1984). Forrest refers to Parfit in relation to ethics, but he does not specifically address how Parfit's views of personal identity would impact his own theistic presentation of the afterlife. Parfit's larger purpose is to express personal identity in indeterminate terms so that he can employ it as an argument against S, the self-interest theory, which claims that persons should always do what is in their own self-interest, often foregoing immediate gains because of future negative consequences. If Parfit can weaken the link of personal identity between now and the future, the future consequences will be of less import.
3. Parfit, 202. Excellent overviews of the issues surrounding personal identity include C. Blakemore and S. Greenfield, Mindwaves (New York: Basil Blackwell, 1987); Daniel Kolak and Raymond Martin, eds., Self and Identity: Contemporary Philosophical Issues (New York: Macmillan, 1991); J. Perry, ed., Personal Identity (Berkeley: University of California Press, 1975); and David M. Rosenthal, ed., The Nature of Mind (New York: Oxford University Press, 1991).
4. Ibid., 199-201.
5. Ibid., 201-202, 206. Psychological connectedness is a matter of degree, but Parfit suggests that the number of direct connections which hold should be at least half of the connections for every actual person. Strong connectedness requires even more direct connections.
6. Ibid., 202-209.
7. Ibid., 210.
8. Ibid., 209-213. According to Parfit, the reductionist view allows that a person may be appropriately spoken of as an agent, a thinker, what has experiences, and the subject of experiences. But it cannot countenance speaking of a person as a separately existing entity distinct from a brain and a body and a series of physical and mental events.
9. Ibid., 210-211.
10. Ibid., 211-213.
11. Williams' objection, in the form of a counterexample, goes against Parfit's central project. A patient (in great pain) is being operated on by a crazed neurosurgeon who causes the patient to have amnesia and then gives him a whole new series of memories, say of Napoleon's. Williams' point is that even if there were such a radical change in personal identity, the person would still fear future pain, and hence the psychological criterion is false and the self-interest theory is true. See Parfit, 229-230.
12. Ibid., 223-243.
13. My discussion of theism will be primarily from a Christian point of view, however, these views are shared to some degree by other (particularly Western) religious traditions.
14. Parfit, 204.
15. Forrest, 181-185, 191-200.
16. Ibid., 56-64.
17. W. V. Quine, review of Milton K. Munitz, ed., Identity and Individuation, in The Journal of Philosophy 69 (1972), 490, cited in Parfit, 200. Parfit also quotes Ludwig Wittgenstein to the same effect.
18. Parfit, 245-280.
19. Thomas Nagel, "Brain Bisection and the Unity of Consciousness," Synthese 20 (1971); reprinted in his Moral Questions (Cambridge: Cambridge University Press, 1979), 147-164.
20. If split brain research indicated what Parfit believes it does, it would certainly be a surprise to Roger Sperry, the neurosurgeon with dualistic commitments who pioneered split brain research. See Roger Sperry, "Hemisphere Deconnection and Unity in Conscious Awareness," in Self and Identity, ed. Kolak and Martin, 55-68; Arthur Custance, The Mysterious Matter of Mind (Grand Rapids: Zondervan, 1980); and Laurence W. Wood, "Recent Brain Research and the Mind-Body Dilemma," in The Best of Theology, 2 vols., ed. J. I. Packer and Paul Fromer (Carol Stream: Christianity Today, 1988), 2:203-241.
21. Christine Korsgaard, "Personal Identity and the Unity of Agency: A Kantian Response to Parfit," in Self and Identity, ed. Kolak and Martin, 323-338.
22. Robert Nozick, "The Closest Continuer View," in Self and Identity, ed. Kolak and Martin, 213-226.
23. A thorough discussion of split brain and split mind research is in Self and Identity, ed. Kolak and Martin, 55-162.
24. Forrest, 31.
25. Parfit, 502-503.