So It Goes

  1. 1.  New York University

Change presupposes a certain position which I take up and from which I see things in procession before me: there are no events without someone to whom they happen and whose finite perspective is the basis of their individuality. Time presupposes a view of time. It is, therefore, not like a river, not a flowing substance. The fact that the metaphor based on this comparison has persisted from the time of Heraclitus to our own day is explained by our surreptitiously putting into the river a witness of its course. … Time is, therefore, not a real process, not an actual succession that I am content to record. It arises from my relation to things.

— M. Merleau-Ponty2

I believe that the existence of an enduring self is an illusion and that this illusion is the root of the suffering inherent in the human condition. I am not a scholar of Buddhism or a practitioner, and this lecture is not an exercise in Buddhist studies. I merely want to explore whether this particular Buddhist thought can be understood in terms familiar to analytic philosophy. How might the illusion of an enduring self lie at the root of human suffering?

One of my reasons for wanting to understand this thought is that it challenges an attitude shared by several philosophers who might otherwise seem sympathetic to the Buddhist conception of the self. Philosophers as diverse as Christine Korsgaard and Daniel Dennett have claimed that the self is something that we must invent or construct.3 But these philosophers believe that inventing or constructing a self is a wonderful accomplishment of which we should be proud, whereas the Buddhists believe that it is a tragic mistake that we should try to undo. Can Western philosophers make sense of the Buddhist attitude? That’s what I want to know.

One philosopher who professes to embrace the Buddhist attitude is Derek Parfit, reflecting on his own neo-Lockean theory of personal identity.4 Locke argued that our past selves are the people whose experiences we remember first-personally. Parfit points out that the experiences of a single person in the past might in principle be remembered by more than one of us in the present — if, for example, the hemispheres of the person’s brain had been transplanted into two different bodies. In that case, there would be more than one of us with a claim to a single past self, a situation incompatible with the logic of identity. Hence connections of memory do not necessarily trace out the career of a single, enduring object, and they are unsuited to serve as the integuments of an enduring self.

Parfit suggests that giving up our belief in an enduring self would be beneficial. Of the time when he believed in his own endurance, he says, “I seemed imprisoned in myself”:5

My life seemed like a glass tunnel, through which I was moving faster every year, and at the end of which there was darkness. When I changed my view, the walls of my glass tunnel disappeared. I now live in the open air.

Parfit elsewhere describes this liberation in less metaphorical terms:6

Egoism, the fear not of near but of distant death, the regret that so much of one’s only life should have gone by — these are not, I think, wholly natural or instinctive. They are all strengthened by the beliefs about personal identity which I have been attacking. If we give up these beliefs, they should be weakened.

Parfit explicitly notes the similarity between his view of personal identity and that of the Buddhists,7 but he does not directly compare the consolations claimed for these views. Such a comparison might have suggested to Parfit that he underestimates the revolution in attitude that his view of personal identity can produce. For he claims that the consolations of his view can be obtained by attending to the philosophical arguments for it,8 whereas the Buddhists believe that they can be obtained only through long and arduous meditational practice.

I will argue that shedding our belief in an enduring self would have consequences far more radical than Parfit has imagined — results that cannot be obtained by philosophical argument alone. Breaking out of a glass tunnel is not the half of it.

In order to understand how belief in an enduring self could lead to suffering, we have to understand the ontological status of the self believed in. What exactly would it be for the self to endure?

Metaphysicians have defined two distinct conceptions of how objects persist through time.9 Under one conception, objects are extended in time as they are extended in space. Just as a single point in space can contain only part of an extended object, a spatial part, so a single point in time can contain only part of a persisting object, a temporal part. The object fills time by having one temporal part after another, just as it fills space by having one spatial part next to another. An object that persists through time in this way is said to perdure.

Under the alternative conception, an object’s extension in time is different from its extension in space. Whereas only part of an object can be present at a single point in space, the object can be wholly present at a single point in time. An object that persists through time in this way is said to endure.

But what does it mean to say that the object is wholly present at a single point in time?10 To be sure, all of its spatial parts can be present at a single instant, but all of its spatial parts are conceived to be simultaneously present under the conception of it as perduring, too. And saying that the object is wholly present at a single point in time cannot mean that all of its temporal parts are present. For how can all of the object’s temporal parts be present at a single point in time if the object also exists at other times?

According to some philosophers, saying that an object is wholly present at a single point in time means that it does not have temporal parts at all. Yet what is to prevent us from considering the object as it is at a single moment and then denominating that aspect of it as a temporal part? If the object is extended in some dimension such as time, and that dimension is itself divisible into smaller and smaller regions, such as hours and minutes and seconds, then nothing can prevent us from abstracting temporal parts from the object by prescinding from its existence beyond one of those regions. The nature of endurance thus appears mysterious. And the suspicion arises that we couldn’t possibly believe in an enduring self, because we have no coherent idea what it would be for the self to endure.

These brief considerations fall far short of proving that no coherent idea of an enduring self can be found. But rather than pursue a coherent idea of an enduring self, we should consider the possibility that an incoherent idea will do. An incoherent idea will certainly do if the enduring self is just an illusion. Maybe if we figure out how such an illusion might arise, we will understand the resulting idea, coherent or not.

In my view, the idea of an enduring self arises from the structure of experience and experiential memory, just asLocke first suggested.11 When I remember a past experience, I remember the world as experienced from the perspective of a past self. My memory has an egocentric representational scheme, centered on the person who originally had the experience from which the memory is derived. That person’s standpoint lies at a spatiotemporal distance from the present standpoint that I occupy while entertaining the memory. But the mind is not especially scrupulous about the distinction between the subjects occupying these distinct points of view.

Consider, for example, my memory of blowing out the candles on a particular birthday cake in 1957. This memory includes an experiential image of a cake and candles as seen by a five-year-old boy. Now, if I invite you to imagine that you are that birthday boy, then you will conjure up a similar image in your imagination. You might report this thought experiment by saying, “I’ve just imagined that I am the birthday boy at David Velleman’s fifth birthday party.” The first occurrence of the pronoun ‘I’ in this report would of course refer to you, whoever you are: let’s say you’re Jane Doe. But what about the second occurrence of ‘I’? Have you imagined that you, Jane Doe, are the birthday boy? Surely, you haven’t imagined a bizarre scenario in which the five-year-old David Velleman is somehow identical with a completely unrelated woman (as we are supposing) named Jane Doe. Rather, you have simply imagined being the five-year-old David Velleman, by imagining the birthday party as experienced by him.12You have formed an experiential image whose content might be summed up by the statement “I am the birthday boy” as uttered in the imagined scene by the five-year-old David Velleman — a statement in which ‘I’ would refer to him, the one experiencing the scene, rather than you, the one who has imagined it.13 When you say, “I’ve imagined that I am the birthday boy,” you should be interpreted as saying, “I’ve imagined an experience with the content ‘I am the birthday boy’,” or “I’ve imagined ‘I am the birthday boy’,” where the first occurrence of ‘I’ refers to you but the second refers to him.

What then of my experiential memory? When I say, “I remember that I was the birthday boy,” I am making a report similar to yours. That is, I am reporting an experiential memory whose content would be expressed by the statement “I am the birthday boy” as uttered in the remembered scene by the five-year-old who experienced it. But whereas you may be aware that you haven’t imagined the birthday boy’s being you, Jane Doe, I am strongly inclined to think that I have remembered his being me, the present subject of this memory.14 I thereby conflate my remembering self with the self of the experience remembered. When I say, “I remember that I was the birthday boy,” I take myself to be referring twice to my present self. I who remember the experience and the “I” of the experience thus become superimposed, so that a single self appears to be present in both.

The selves superimposed in this appearance are two momentary subjects: I in my present capacity as the subject of memory, existing just in the moment of remembering; and the “I” of the remembered experience, who existed just in the moment of the experience. In either case, I am conceived as wholly present at a single point in time, either as me-here-and-now, entertaining the memory, or as “me”-there-and-then, having the experience. Superimposing one of these momentary subjects on the other yields the illusion that they are numerically identical — that the subject whose existence was complete in the moment of the experience remembered was one and the same as the subject whose existence is complete in the moment of remembering. This appearance is already incoherent if one and the same thing cannot have its existence confined to each of two different moments. The incoherence is compounded by the thought that this momentary subject has persisted through the interval between the original experience and the memory, existing in its entirety at each intervening moment.15

The same effect is produced by experiential anticipation, in which I prefigure a future experience from the perspective that I expect to occupy in it. A single self appears to have its full existence both now and later, because I who anticipate the experience and the “I” of the anticipated experience become superimposed.

For a spatial analog of the resulting idea, think of the scene in which Woody Allen plays a spermatozoon about to be launched from the loins of … Woody Allen.16 In reality, of course, a person occupies different points in space with different parts of himself, none of which is identical to any other part or to the person as a whole. We might say, then, that a person pervades space. In this scene, however, Woody Allen occupies different points in space with a smaller self that plays the role of each spatial part of his own body. We might say, then, that he invades space rather than pervading it. Incoherent, to say the least. Yet experiential memory leads me to think that my own temporal extension is composed of a single momentary self playing the role of each temporal part of my existence.

I am tempted to say that all of my temporal parts are present at a single point in time because I tend to think of myself as my present self — a momentary subject whose existence is indeed complete in the here-and-now. I am tempted to say that I nevertheless persist through time because I tend to think of this self, complete in the moment, as nevertheless existing at other moments. And because I therefore conceive of each moment in my temporal extension as containing my complete self, I am tempted to deny that it contains a mere temporal part of me. There I am, all of me, at my fifth birthday party; here I am, all of me, remembering that party; there I will be, all of me, on my seventy-fifth birthday — as if one and the same momentary subject can play the several parts of my five-year-old, sixty-two-year-old, and seventy-five-year-old selves. I think of myself as all of me, all the time, just as Woody Allen is all Woody Allen in every one of his cells.

What would be the consequences of truly shedding our sense of being enduring objects and learning to conceive of ourselves as perduring instead? I want to suggest that the existence of an enduring self, if it is indeed an illusion, is one of two illusions that go hand-in-hand. A consequence of shedding the one illusion would be to shed the other as well. The other illusion of which I speak has to do with the nature of time.

The concept of perdurance for objects is most at home in a conception of time known as eternalism. According to eternalists, all of the temporal facts can be expressed in terms of the temporal relations between events. One event can occur earlier or later than another, or simultaneously, and it can be closer to or further from the other in time. The relations among events as earlier, later, or simultaneous, and closer or further apart, exhaust the temporal facts, in the eyes of eternalists: there is no more to time than these relations.

The philosopher J. Ellis McTaggart argued that the temporal relations among events are not sufficient to satisfy our concept of time, although he also argued that the concept is incoherent.17 Temporal relations among events do not change, and so McTaggart argued that they cannot account for the passage of time — that is, for the way events draw nearer from the future, until they occur in the present and, having occurred, recede into the past. When we say that a future event is always drawing closer and closer, eternalists must understand us as meaning only that the event is nearer to our second utterance of the word ‘closer’ than it was to the first. And these temporal relations are as they always were and always will be; or, rather, they exist timelessly, constituting time itself. The future event that we describe as drawing closer and closer not only stands closer to the last word of our description than it does to the earlier words; it always has and always will stand in those relations, or it stands in them timelessly. Such unchanging relations cannot constitute time, McTaggart argued, because time requires change — specifically, the change that consists in an event’s approaching from the future, arriving in the present, and receding into the past.

Yet the change thus required by our concept of time struck McTaggart as paradoxical and hence impossible. An event’s changing from future to present to past must unfold in time: the event must be first in the future, then in the present, and then again in the past. And having added these temporal indices (“first”, “then”, “then again”) to our description of the change, we have reverted to an eternalist idiom. We have said that the event’s status as a future event occurs first — that is, earlier than its status as present and past; that its status as present occurs later than its status as future; and that its status as past occurs even later than the other two. Thus, in order to complete our description of how time passes, we have been forced to describe it once again in terms of those temporal relations which never change.

There is a temptation to say, at this point, that what moves is not the future or past but the present, or rather the property of being the present, which belongs successively to different sets of events. But if we try to describe how the property of being present passes from one set of events to the next, we will end up saying that it belongs first to one set, then to another, and then again to a third, as they occur in succession. We will thereupon have said no more than this: that the property of being present belongs earlier to some events than to others, always belonging to events simultaneously with their occurrence. So in what sense can the present be said to move? Every event is present sooner or later, and that fact never changes.

One fairly desperate attempt to solve the problem is a theory known as presentism. According to presentism, only the present exists; past and future are merely tenses modifying facts about the present.18

Presentism is best explained by an analogy between time and modality. Consider the fact that John Kerry might have won the 2004 presidential election. We could restate this fact by saying that a Kerry victory occurs in a merely possible history, alternative to the one that actually unfolded in 2004; but perhaps we wouldn’t then be speaking with metaphysical strictness. Perhaps we should acknowledge only one event — Kerry’s loss, which actually occurred — plus the subjunctively statable fact, also true of actuality, that Kerry might have won instead. According to this view, called actualism, there is no Kerry victory that occurs in a realm of mere possibility: actual events are the only events there are.19

Presentism goes one step further, refusing to acknowledge even an event of Kerry’s losing the election. For when we describe Kerry’s loss as occurring in the past, the presentist claims that we are speaking just as loosely as we would in describing his victory as occurring in some alternative possible history. The only events there are, according to the presentist, are the ones occurring now in actuality. Just as Kerry’s possibly having won is a fact about actuality, statable in the subjunctive, so his previously having lost is a fact about the present, statable in the past tense. That he might have won, and that he did lose, are subjunctive and past-tense facts about the actual present, which is all there is for facts to be about. There is no Kerry victory occurring in a realm of possibility; and there is not even a Kerry loss occurring in a realm of the past.

The presentist claims that his view enables us to represent the passage of time. The occurrence of an event entails the fact that it will have occurred, and hence that it will later be a matter of past-tense fact. (More precisely, the event’s occurrence entails the future-tense fact that there will be a past-tense fact of its having occurred.) This entailment is said to represent the passage of the event from the present into the past. The occurrence of an event is also incompatible with the fact that it wasn’t going to occur, and compatible with the fact that it was going to occur. Hence its present occurrence entails that it was previously a subject of future-tense facts, an entailment that is said to represent its passage from the future into the present. Finally, the occurrence of an event is compatible with its being the case neither that the event was going to occur nor that it wasn’t going to, while nevertheless entailing that the event definitely will have occurred. That is, while there previously may have been no fact of the matter whether the event would occur, there will later be a determinate fact of its having occurred — a constellation of facts that is said to represent how an open future gets closed up into a fixed past.20

The presentist also claims that his view enables us to solve our problem about the concept of endurance. Just as there is no John Kerry existing in an alternative possible history in which he won the election, according to presentism, so there is no John Kerry existing in a past in which he lost: all there is of John Kerry is the present John Kerry. This person has the past-tense properties of having existed in 2004 and having lost the election of that year, just as he has the subjunctive property that he might have won; but the presentist insists that these properties belong to Kerry’s actual present self, which is all of him that exists. Hence, the presentist can deny that John Kerry perdures, by denying that he has any temporal parts. According to presentism, Kerry’s existence is confined to the present.

One drawback of presentism is that it requires the present to bear sufficient features to render true not only present-tense facts but all past-tense facts as well: the present must, as it were, bear witness to all of history.21 A more serious problem, for my purposes, is that presentism doesn’t really solve the problems of endurance and the passage of time. What presentism describes is — not a changing prospect in which events approach from the future, arrive in the present, and recede into the past — but a single, static structure of past-prospective and future-perfect facts, all true of the present. Tensed facts about the present entail other tensed facts about the present, but nothing moves. Similarly, presentism describes objects as being wholly present at every moment of their existence, but only because it describes them as existing at only one moment, the present; and so it describes them as enduring in only a trivial sense. According to presentism, objects have past- and future-tensed properties, but the objects themselves exist only in the present, and so they don’t persist at all, much less endure.

Surely, we should hope for a more intuitively satisfying solution to the problems of endurance and temporal passage. I think that the solution is to recognize that both phenomena are illusions, and that these illusions are interdependent. I have already suggested how the illusion of an enduring self might arise from the structure of first-personal memory and anticipation. I will now suggest that the illusion of an enduring self gives rise to another illusion: that of movement with respect to time.

Our difficulty in characterizing such movement was that, when we tried to identify something toward which a future event draws nearer or from which a past event recedes, we focused our attention on other events. Yet each event depends for its identity on when it occurs: it could not be closer to a future event, or further from a past event, without occupying a different temporal position and hence being a different event. This conception of the problem suggests the solution. Whatever the future draws nearer to, or the past recedes from, must be something that can exist at different positions in time with its identity intact. And we have already found such a thing — or the illusion of one, at least — in the form of the enduring self.

Suppose that I endure in the admittedly incoherent sense that is suggested by experiential memory and anticipation. In that case, I exist in my entirety at successive moments in time, thereby moving in my entirety with respect to events. As I move through time, future events draw nearer to me and past events recede. Time truly passes, in the sense that it passes me.

If I merely perdure, however, then I do not move with respect to time. I extend through time with newer and newer temporal parts, but all of my parts remain stationary. A perduring self can be compared to a process, such as the performance of a symphony. The performance doesn’t move with respect to time; it merely extends newer and newer temporal parts to fill each successive moment. The last note of the performance is of course closer to midnight than the first, but we wouldn’t say that midnight and the performance move closer together. Midnight is separated from the performance by a timelessly fixed but extremely vague interval, which can be made precise only with respect to particular parts of the performance — the first note, the second note, the third note — each of which is separated from midnight by an interval that is also timelessly fixed. Similarly, we wouldn’t say that the ceiling and I get closer together from my feet to my head. The ceiling stands above me at a fixed but vague distance, which can be made precise only with respect to particular parts of me — feet, waist, head — each of which is separated from it by a fixed distance.

But if I am an enduring thing, then midnight and I get closer together, and not just in the sense that I extend temporal parts closer to it than my earlier parts. I don’t just extend from a 9:00 p.m. stage to a 10:00 p.m. stage that is closer to midnight, as I extend from my feet to a head that is closer to the ceiling; I exist in my entirety within the stroke of 9:00, and I exist again within the stroke of 10:00 — the selfsame entity twice, existing once further from midnight and then all over again, closer. Midnight occupies two different distances from my fully constituted self. From my perspective, then, midnight draws nearer.

If this enduring “me” is an illusion, however, then so is the passage of time. And ceasing to think of myself as an enduring subject should result in my ceasing to experience the passage of time. Coming to think of myself as perduring should result in my coming to experience different temporal parts of myself at different moments, but no enduring self past which those moments can flow.

Suppose that I could learn to experience my successive moments of consciousness — now and now and now — as successive notes in a performance with no enduring listener, no self-identical subject for whom these moments would be now and then and then again. In remembering a scene that I experienced in the past, I would distinguish between the “I” who remembers it and the “I” who experienced it; in anticipating a scene that I would experience in the future, I would distinguish between the anticipating “I” and the experiencing “I” as well. Hence, my present self would be cognizant of being distinct from the past subjects from whom it receives memories and the future subjects for whom it stores up anticipations. It would therefore have no conception of a single subject to which events could bear different relations over time, nothing to which they could draw near or from which they could recede. It would think of itself, and each of the subjects with whom it communicates by memory and anticipation, as seeing its own present moment, with none of them seeing a succession of moments as present.

The result would be that time would no longer seem to pass, because my experience would no longer include a subject of its passage — just successive momentary subjects, each timelessly entrenched in its own temporal perspective. I would think of myself as filling time rather than passing through it or having it pass me by — as existing in time the way a rooted plant exists in space, growing extensions to occupy it without moving in relation to it. Having shed the illusion of an enduring self, I would have lost any sense of time as passing at all.

One small bit of evidence in support of this speculation is that when I lose awareness of myself, by “losing myself” in engrossing activities, I also tend to lose awareness of time’s passing.22 With my attention fully devoted to playing a sport, reading a book, writing a paragraph, I am drawn out of myself and, as it seems, out of the passage of time as well. Conversely, when I have nothing to occupy my attention — that is, when I am bored — my attention returns to myself, and the passage of time becomes painfully salient. Self-awareness and time-awareness thus seem to go hand-in-hand.

Clearly, I am nowhere near to “losing myself” in this way on a lasting basis, despite being convinced, by the arguments of Locke and Parfit, that I am in fact a perduring rather than an enduring self. Truly assimilating the implications of those arguments would entail radical changes in my experience, changes of the sort that no argument can produce. No wonder the Buddhists believe that dispelling the illusion of an enduring self requires an arduous regimen of meditation.

As we have seen, Parfit blames our belief in an enduring self for emotions that might well be the essence of our existential suffering: grief over time past and anxiety at the prospect of death. Parfit suggests that these emotions get their sting from our proprietary interest in our one and only life — that glass tunnel in which we imagine ourselves to be enclosed, when we believe that we have enduring selves. Parfit claims to derive consolation from shedding this belief because he no longer views his relation to the person lost in the past, or to the person who will die in the future, as a relation of identity. The consolation comes when he escapes from seeming imprisoned in an enduring self.

Yet I don’t see why bearing a less robust relation to his own past and future is any consolation to Parfit. Why should a sense of partial alienation from past and future selves leave him feeling relieved rather than bereft? It’s not as if he has come to realize that this isn’t his “only life”; he has merely come to realize that it isn’t even his in the sense that he previously thought. This realization provides only the cold comfort of having nothing to lose.

When Parfit describes the drawbacks of believing in an enduring self, he speaks not only about the loneliness of proprietorship in a single life — being imprisoned in a glass tunnel — but also about the emotions attendant upon time’s passage. He complains of the sense that he is “moving faster and faster” through the tunnel, toward the “darkness” at its end, and of the sense that “so much of one’s only life should have gone by”. Surely, the remedy for these anxieties and regrets is not to get out of the tunnel and live “in the open air”; the remedy is to stop moving.

The remedy for Parfit’s distress, in other words, is to become an eternalist. Consider: 23

[W]hen a person dies he only appears to die. He is still very much alive in the past, so it is very silly for people to cry at his funeral. All moments, past, present, and future, always have existed, always will exist. … It is just an illusion … that one moment follows another one, like beads on a string, and that once a moment is gone is it gone forever.

The speaker here is Billy Pilgrim, relating what he learned on the planet Tralfamadore, where he was once on display as an intergalactic zoological specimen:

When a Tralfamadorian sees a corpse, all he thinks is that the dead person is in bad condition in that particular moment, but that the same person is just fine in plenty of other moments. Now, when I myself hear that somebody is dead, I simply shrug and say what the Tralfamadorians say about dead people, which is “So it goes.”

The Tralfamadorians are eternalists about time, and they have managed to derive great comfort from this philosophy.

Note, however, that whereas Parfit has overcome the illusion of an enduring self but not the illusion of time’s passing, the Tralfamadorians have done the reverse: they have overcome the illusion of time’s passing, but they still speak as if they believe in an enduring self.24 This incomplete disillusionment is just as unsatisfactory, to my way of thinking, as Parfit’s. Parfit and the Tralfamadorians have divided between them what is a larger truth: the enduring self and the passage of time are interdependent illusions. The Tralfamadorian half of the truth is more consoling than Parfit’s, to my mind; but taken by itself, the Tralfamadorian half of the truth is unstable.

The Tralfamadorians speak as if they occupy moments in time with their entire selves, not just temporal parts. Regarding themselves as enduring objects, they manage to deny the passage of time only by asserting that they can stand outside of time and range across it at will:

The Tralfamadorians can look at the different moments just the way we can look at a stretch of the Rocky Mountains, for instance. They can see how permanent all the moments are, and they can look at any moment that interests them.

Billy Pilgrim never fully attains the Tralfamadorian view of time, but he does lose the normal human view: 25

Billy Pilgrim has come unstuck in time.

Billy has gone to sleep a senile widower and awakened on his wedding day. He has walked through a door in 1955 and come out another one in 1941. He has seen his birth and death many times, he says, and pays random visits to all the events in between.

He says.

Billy is spastic in time, has no control over where he is going next, and the trips aren’t necessarily fun. He is in a constant state of stage fright, he says, because he never knows what part of his life he is going to have to act in next.

How do the Tralfamadorians manage to visit different moments in time, betaking their complete selves from one moment to another? This process would require a higher temporal order of “first” and “later” within which the desultory visits could occur, and across which the Tralfamadorians would retain their identities. A Tralfamadorian’s visits to random moments in ordinary time would themselves have to occur at well-ordered moments in a meta-time, which would constitute a temporal stream washing over the Tralfamadorians as relentlessly as ordinary time washes over us. Similarly, Billy Pilgrim is washed by a stream of meta-moments ordering his visits to random moments of ordinary time.

In short, “coming unstuck in time” is not as easy as it sounds. Billy Pilgrim may jump around in one temporal order, but he moves through another in sequence. Escaping the passage of time would require the dissolution of his enduring self. In order to come completely unstuck in time, Billy himself would have to come unglued.

Although the tale of Billy Pilgrim gives a partial and imperfect portrait of life without the illusion of temporal passage, it seems correct in portraying that life as lacking many of our ordinary worries about mortality. Even so, not all such worries would disappear along with the passage of time.

Billy describes the Tralfamadorians as unconcerned about being dead. But of course Epicurus long ago taught us that being dead is nothing — literally — and hence that it is nothing to worry about. The anxiety that makes sense, at least for those of us who live with temporal passage, is anxiety about the inexorable approach of death, about time’s running out. This anxiety would be allayed if time no longer seemed to pass. And once time no longer seemed to pass, the mere fact of our mortality would no longer seem regrettable. When time seems to be running out, we wish for immortality, which would amount to having infinite time left on the clock. But in an eternalist world, immortality would amount instead to a kind of temporal ubiquity — existing at every future moment. Having an infinite amount time left seems desirable if time is running out; but if time is standing still, then filling an infinite amount of it might well seem unattractive.

Still, those of us who die young could continue to lament the truncated extent of our lives: having too short a life would still be grounds for unhappiness. What would be groundless is unhappiness about mortality itself — the unhappiness that affects everyone, no matter how long-lived, at the sound of death’s approaching tread.

Would liberation from the passage of time free us from other kinds of suffering? It certainly wouldn’t spare us from physical pain or other unpleasant experiences. But it just might prevent pain and unpleasantness from being transformed into suffering.

We can undergo pain or unpleasantness without suffering under it: suffering is a particular way of experiencing pain or unpleasantness — specifically, of not coping with it.26 And I suspect, though I cannot argue here, that the way of not coping that’s constitutive of suffering results from the perception of time as passing. What undoes us, when we suffer with pain, is panic at the thought that it will never abate, that no end is in sight. Patients can learn to bear pain by “accepting” or “being with” it, focusing on the pain of the moment, without thinking about what’s next.27 It’s not the pain they’re in that makes them suffer but the prospect of its endlessly going on.

Perhaps, then, liberation from the passage of time would entail liberation from suffering altogether, though not, of course, from pain. There would be bad moments and good moments, but no panic about the coming moments, and hence no suffering.

The Tralfamadorians express the consolations of their perspective by saying, “So it goes.” Come to think of it, though, the point of this motto is less than obvious. After all, the Tralfamadorians inhabit a perspective in which “it” doesn’t “go” at all, since they do not experience time as passing. Why do they say “So it goes”? Why don’t they say “So it is”?

Maybe the Tralfamadorian motto has been translated in a manner suitable to us, who simply cannot escape from the illusion of time’s passing. “So it goes” means “so it goes for you.” They are recommending the attitude that is appropriate for creatures who can’t help but experience time as passing. Buddhism must offer similar advice, exported not from one planet to another but from the meditative state to the state of ordinary consciousness. What is the appropriate attitude to have in ordinary life, where the self unavoidably seems to endure and time unavoidably seems to pass, given that both appearances are illusions?

I think that the exportable lessons here must include something about the way we cope with the passage of time. We can’t stop the self from seeming to endure, or stop time from seeming to pass, but we can cope with these phenomena better, given the knowledge that they are merely phenomenal.

Ordinarily I cope rather badly with temporal passage and personal endurance. I don’t exactly live in state of Pilgrim-esque stage fright, continually unsure when I might find myself at my fifth birthday party or my seventy-fifth. In some respects, I feel like a Tralfamadorian, because I can choose which parts of my life to visit, in memory and anticipation. Yet I have a disconcerting tendency to live different parts of my life all at once — to relive the past and pre-live the future even while I’m trying to live in the present. And even as I relive my past in a memory, it is at the same time slipping away from me, as there comes bearing down on me a future that I am pre-living in anticipation.

It’s as if too many parts of my life are on the table at once, and yet somehow they are continually being served up and snatched away like dishes in a restaurant whose waitstaff is too impatient to let me eat. And this whole grief- and anxiety-provoking conception of my life has been adopted out of panic over the passage of time, which requires me to anticipate the future precisely because it’s bearing down on me, and to remember the past precisely because it’s slipping away.

Once I know that the self doesn’t endure, and time doesn’t pass, then even when under the illusion to the contrary, I can better follow the Buddhist injunction to be fully aware of the present moment. The realization that I am of the moment — that is, a momentary part of a temporally extended self — can remind me to be in the moment, which draws my attention away from time’s passage, even if it doesn’t succeed it stopping time from seeming to pass. Insofar as I can be in the moment, I can perhaps gain some respite from the grief and anxiety of that overwhelmed diner, on whom loaded plates are bearing down even as uneaten dishes are being borne away. Each moment can be devoted to savoring the dish of the moment.

1 This chapter was presented as the first Amherst Lecture in Philosophy on March 9, 2006 and published at It was reprinted in Introduction to Philosophy: Classical and Contemporary Readings, ed. John Perry, Michael Bratman, and John Martin Fischer (Oxford: Oxford University Press, 2010), pp. 371-382. I am grateful to Amherst College; its philosophy department; and the department’s then-chair,Alex George, for organizing the lecture series and inviting me to inaugurate it. The chapter was also presented to the graduate students in philosophy at New York University and to the philosophy departments of Wake Forest University, the Graduate Center at The City University of New York, Dartmouth College, Georgetown University, the University of Melbourne, Monash University, and the Research School of Social Sciences of the Australian National University. The chapter was written during my term as a visiting professor in the philosophy department at the University of North Carolina at Chapel Hill. My visit was funded by a grant from the Mellon Foundation toSusan Wolf. Some of the ideas were developed in a reading group on the metaphysics of time, led by ThomasHofweber. My thanks go to Susan, Thomas, the Mellon Foundation, and the UNC department for a very stimulating semester. For comments on earlier versions, I am grateful to John Bigelow, Jay Garfield, ThomasHofweber, Joel Kupperman, Peter Ludlow, and Daniel J. Velleman.

2 Maurice Merleau-Ponty, Phenomenology of Perception, trans. Colin Smith (London: Routledge and Kegan Paul, 1962), pp. 411-412.

3 See Christine M. Korsgaard, The Sources of Normativity, ed. Onora O’Neill (Cambridge: Cambridge University Press, 1996); idem., “Self-Constitution: Action, Identity, and Integrity”, The Locke Lectures, 2002; Daniel C. Dennett, “The Origins of Selves”, Cogito 3 (1989): 163-173; idem., “The Reality of Selves”, in Consciousness Explained (Boston: Little, Brown and Company, 1991), pp. 412-430; idem., “The Self as the Center of Narrative Gravity”, in Self and Consciousness: Multiple Perspectives, eds. Frank S. Kessel, Pamela M. Cole, and Dale L. Johnson (Hillsdale, NJ: Erlbaum Associates, 1992), pp. 103-115.

4 One might think that Parfit’s arguments militate not just against the self’s endurance but also against its persistence in any sense, including perdurance. (For the difference between endurance and perdurance, see below.) But as David Lewis showed, Parfit’s arguments do not necessarily militate against perduring selves. (See Lewis, “Survival and Identity”, in The Identities of Persons, ed. Amélie Oksenberg Rorty [Berkeley, CA: University of California Press, 1976], pp. 17-40, reprinted in Philosophical Papers, vol. I [Oxford: Oxford University Press, 1983], pp. 55-77.)

5 Derek Parfit, Reasons and Persons (Oxford: Oxford University Press, 1984), p. 281.

6 Parfit, “Personal Identity”, The Philosophical Review 80 (1971): 27.

7 See Parfit, Reasons and Persons, pp. 273, 280, 502-503.

8 Ibid., p. 280.

9 See Sally Haslanger, “Persistence Through Time”, in The Oxford Handbook of Metaphysics, ed. Michael J. Loux andDean W. Zimmerman (Oxford: Oxford University Press, 2005), pp. 315-354.

10 The following objections to the traditional conception of endurance are developed more fully in ThomasHofweber and J. David Velleman, “How to Endure”, The Philosophical Quarterly 61 (2011): 37-57. These objections would not apply under the theory of time known as presentism. I discuss presentism briefly below.

11 This paragraph and the four that follow summarize a lengthy argument presented in my “Self to Self”, The Philosophical Review 105 (1996): 39-76; reprinted in Self to Self: Selected Essays (Cambridge: Cambridge University Press, 2005), pp. 170-202. See also “The Identity Problem”, part 1 of chapter 6 in this volume.

12 This point was made by Bernard Williams in “Imagination and the Self”, in Problems of the Self: Philosophical Papers 1956–1972 (Cambridge: Cambridge University Press, 1973), pp. 26-45. I discuss Williams’s paper in “Self to Self”.

13 The second ‘I’ functions as what Héctor-Neri Castañeda called a quasi-indicator — a pronoun in indirect discourse that takes the place of what was a first-personal pronoun in direct discourse. For an explanation of quasi-indicators (clearer than Castañeda’s), see John Perry, “Belief and Acceptance”, in The Problem of the Essential Indexical and Other Essays (New York: Oxford University Press, 1993), pp. 53-68.

14 But isn’t it a contingent truth-condition of my memory that the remembered experience has been undergone by me rather than someone else? And if so, how can the second ‘I’ in “I remember that I was the birthday boy” refer merely to the subject of the remembered experience, who necessarily did undergo it, if anyone did? The answer is that the memory refers to the subject of the remembered experience indexically, pointing to him at the perspectival point of origin in the remembered experience, by pointing to him at the corresponding point in my memory-image, which purports to be a copy derived from that experience. If the image is indeed a copy derived from an experience, as it purports to be, then indexical reference to the “me” of that experience succeeds, and his being the birthday boy is what I veridically remember; if the image is not copied from an experience, then its indexical reference to the “me” of that experience fails — it refers to no one at all — and the memory is illusory. In order for the memory to be veridical, then, the remembered experience must have been undergone by me in the sense that its subject must be accessible to indexical reference as “me”.

Of course, your image of being my five-year-old self also refers to the birthday boy as “me”, but not in the same, genuinely indexical way. In conjuring up this image, you had to stipulate that its point of origin is occupied by the five-year-old David Velleman, thus referring to him by name before you could go on to think of him as “me”. In remembering the experience, I can refer to him as “me” directly, without any stipulation about whom the pronoun refers to, relying on the causal history of my image to secure my reference to the original subject. That is the sense in which I have first-personal access to him whereas you do not. (For further discussion of this issue, see “Self to Self” and “The Identity Problem”.)

15 I find indirect evidence for these claims about autobiographical memory in the experience of reading truly gifted autobiographical novelists, such as Laura Ingalls Wilder (Little House on the Prairie) or Elspeth Huxley (The Flame Trees of Thika). These authors were able to depict past experience as it was registered by the childish minds of their younger selves. Reading their work, I am struck by the contrast with my own childhood memories, in which the psychological distance between the mind that stored a memory and the mind that retrieves it is foreshortened, so that past experience seems to have been registered by my current, adult consciousness — the remembering “I”, who has been superimposed on the “I” remembered.

16 In Everything You Always Wanted to Know About Sex* But Were Afraid to Ask, dir. Woody Allen, Rollins-Joffe Productions, United Artists, 1972.

17 J. Ellis McTaggart, “The Unreality of Time”, Mind 17 (1908): 457-474.

18 In the following paragraphs I have drawn on John Bigelow, “The Passage of Time” (MS).

19 So-called modal realists, such as David Lewis, believe that there are events and things inhabiting such a realm, but the intuitions of most philosophers run to the contrary.

20 That there was previously no fact of the matter whether the event would occur, and that there will later be a determinate fact of its having occurred, are of course past- and future-tense facts about the present, according to presentism. The same goes for all of the entailments discussed in this paragraph.

21 For this objection, see Simon Keller, “Presentism and Truthmaking”, in Oxford Studies in Metaphysics, vol. I, ed.Dean W. Zimmerman (Oxford: Oxford University Press, 2004), pp. 83-106 (cited by Bigelow).

22 See Mihaly Csikszentmihalyi, Flow: The Psychology of Optimal Experience (New York: Harper & Row, 1990). According to Csikszentmihalyi, losing awareness of self and losing awareness of time are two of the characteristic features of “flow” experiences. I discuss these experiences further in “What Good is a Will?”, in Action in Context, ed. Anton Leist (Berlin: Walter de Gruyter, 2007); and “The Way of the Wanton”, in Practical Identity and Narrative Agency, ed. Kim Atkins and Catriona MacKenzie (London: Routledge, 2008), pp. 169-192; both reprinted in The Possibility of Practical Reason, second edition (Ann Arbor, MI: Maize Books, 2015).

23 Kurt Vonnegut, Jr., Slaughterhouse-Five or, The Children’s Crusade: A Duty-Dance with Death (New York: Dell Publishing, 1969), p. 23.

24 But: “Tralfamadorians don’t see human beings as two-legged creatures, either. They see them as great millepedes — ‘with babies’ legs at one end and old people’s legs at the other,’ says Billy Pilgrim” (ibid., p. 75). This suggests that Tralfamadorians see people as perduring space-time worms rather than enduring objects. Nevertheless, their first-personal descriptions of their own experiences sound like those of an enduring self.

25 Ibid., p. 20.

26 I discuss this conception of suffering in “Beyond Price”, chapter 4 in this volume; and “The Gift of Life”, part II of chapter 6.

27 Here I am merely gesturing at a large and controversial research program. For just one example, see Lance M.McCracken and Christopher Eccleston, “Coping or Acceptance: What to Do About Chronic Pain?”, Pain 105 (2003): 197-204; and Lance M. McCracken, James W. Carson, Christopher Eccleston, and Francis J. Keefe, “Acceptance and Change in the Context of Chronic Pain”, Pain 109 (2004): 4-7. One of the methods discussed in the latter article is “Mindfulness-Based Stress Reduction”, which is described as “moment-to-moment observation and acceptance of the continually changing reality of the present” (5). For some of the methodological problems in this area, see Christopher Eccleston, “The Attentional Control of Pain: Methodological and Theoretical Concerns”, Pain 63 (1995): 3-10.

Originally published at





This article and its reviews are distributed under the terms of the Creative Commons Attribution 4.0 International License, which permits unrestricted use, distribution, and redistribution in any medium, provided that the original author and source are credited.