1

LOST IN LOST’S TIMES

Richard Davies

 

 

Lost and Losties have a pretty bad reputation: they seem to get too much fun out of telling and talking about stories that everyone else finds just irritating. Even the Onion treats us like a bunch of fanatics. Is this fair? I want to argue that it isn’t. Even if there are serious problems with some of the plot devices that Lost makes use of, these needn’t spoil the enjoyment of anyone who finds the series fascinating.

Losing the Plot

After airing only a few episodes of the third season of Lost in late 2007, the Italian TV channel Rai Due canceled the show. Apparently, ratings were falling because viewers were having difficulty following the plot. Rai Due eventually resumed broadcasting, but only after airing The Lost Survivor Guide, which recounts the key moments of the first two seasons and gives a bit of background on the making of the series.

Even though I was an enthusiastic Lostie from the start, I was grateful for the Guide, if only because it reassured me that I wasn’t the only one having trouble keeping track of who was who and who had done what.

Just how complicated can a plot become before people get turned off? From the outset, Lost presented a challenge by splicing flashbacks into the action so that it was up to viewers to work out the narrative sequence. In the fourth and fifth seasons, things got much more complicated with the introduction of flash-forwards and time travel. These are two types of narrative twists that cause special problems for keeping track of a plot and that also open a can of philosophical worms about time itself.

Constants and Variables

To set the scene about plot complication, I want to call on some very influential thoughts first put forward by the ancient Greek philosopher Aristotle (384–322 B.C.E.).

In his Poetics, Aristotle discussed tragedy, a form of theater written for civic and religious celebrations, in which the best plays were awarded prizes. Because ancient Greek tragedy was designed to gain the approval of the judges and the public, it followed certain formulas (think the Oscars, rather than Cannes or Sundance). Aristotle’s analysis of these formulas can provide us with pointers for assessing the difficulty with Lost.

Most tragedies are based on well-known historical or mythic events. For instance, Ajax by Euripides (480 B.C.E.–406 B.C.E.) concerns a great hero of the Trojan War who commits suicide in a fit of shame and self-disgust when he does not receive the reward he thinks he deserves.

Using this example, Aristotle argued for two principles. First, every tragedy should deal with a single episode in the life of its main character. The audience should follow a clear causal chain from start to finish. Let’s call this “the principle of closure.” In line with this principle, Euripides’ play begins with Ajax’s coveted reward being given to someone else and ends with his death.

Second, there should be some unity to the action, which is to say that merely accidental or unrelated events should be excluded. Let’s call this “the principle of relevance.” In line with this principle, Euripides’ play does not recount Ajax’s boyhood, regardless of how interesting this topic might be.

Does Lost follow Aristotle’s principles of closure and of relevance? At the outset of the series, Oceanic flight 815 crashes, providing a clear starting point for the succeeding chain of events. We are introduced to the survivors, who all share the same predicament. Although the flashbacks begin right away, they are all carefully designed to shed light on the island narrative.

Complications, however, arrive with the Others. Although at first they function merely as antagonists for our survivors, they soon take on lives of their own. For example, through the character of Juliet, we follow a causal chain that begins before the crash of Oceanic flight 815 and ends before the resolution of the survivors’ predicament. Aristotle would not give up on Lost so easily, though.

In addition to single tragedies, Aristotle discussed longer poetic compositions, known as epics, such as the Iliad and the Odyssey of Homer. These are big stories, the former dealing with the Trojan War and the latter with the ten-year journey home of one of its heroes. In epics, the narrative structure is much more complex than that of the standard tragedy. Yet Aristotle notes that even here, the story concentrates on a sequence of interconnected phases of action.

Thus, the Odyssey effectively begins, in Book One, not by focusing on its hero, Odysseus, who has not yet returned from the war, but on his son Telemachus, who is told to go and track down his father. The two don’t actually meet until Book Fifteen (out of twenty-four). In the meantime, they are wandering around the Mediterranean and often find themselves recounting their travels to others, thus supplying the hearer/reader with backstories. For example, during his journey (and before the time of the events recounted in Book One), Odysseus outwitted the one-eyed monster known as Cyclops, but we find out about this only much later, in Book Nine, when Odysseus narrates his trick to the Phaeacian king. In this way, even though many events are presented out of their chronological order, we don’t have too much trouble constructing a coherent time line.

It seems that Lost is not so much a tragedy as an epic. Any given episode of Lost features a single individual who stands at the center of attention and who is the primary subject of the flashbacks and the flash-forwards. Although many episodes finish with cliff-hangers, the principles of closure and relevance are still at work over the longer run.

So Juliet’s causal chain can become part of the story as long as the audience cares about her connection to the survivors of Oceanic flight 815. If her mud fight with Kate wasn’t enough to make us care, then her relationship with Sawyer was.

A blur of unrelated incidents that is spread out over too long a time and that involves too many characters will not hold our attention. The point seems obvious. On the other hand, a story that is too simple is just boring. The hard part is finding a balance between narratives that are challenging and those that are merely confusing.

We’re All in This Together

Aristotle has a lot of other rules, and perhaps Lost does occasionally break them. But so did Shakespeare, and we can gain more pointers from what critics have said about him.

Taking a cue from a brief passage in Aristotle’s Poetics, some critics have objected that many of Shakespeare’s plays bring together an inappropriate array of characters. For example, in A Midsummer Night’s Dream, nobles interact with “rude mechanicals.” Although there may be more than a little elitism behind this concern, we can take a point about the importance of portraying plausible social relations.

The premise of Lost deliberately throws unlikely people together. For sure, there are differences between those who were previously mixed up in crime (Sawyer, Kate, and Jin) and those who had been “pillars of the community” (Jack, Marshal Mars, and, in a sense that might make Americans uncomfortable, perhaps Sayid). But we’re on the Island of Second Chances, and such distinctions have been erased by the crash of Oceanic flight 815.

Aristotle made the further claim, however, that tragedy properly concerns noble persons (not merely those with noble titles), whereas persons of little worth are the suitable subjects of comedy. After all, why would an audience cry over someone they didn’t care about? And how could they laugh at someone they did?

Clearly, Lost evokes both laughter and tears, but there is an easy out here. We can consider it a tragicomic epic that involves both noble and ignoble characters, or—better still—both noble and ignoble phases in its characters’ lives. We do laugh at those we love in their lesser moments, and we cry for those we don’t love in their best.

The same readers of Aristotle, however, have further objected that Shakespeare’s plays do not observe the so-called unity of genre. What this means is that Shakespeare often alternated scenes of dramatic tension with knockabout farce and facetious wordplay.

And, of course, so does Lost. For example, scenes of Hurley building a golf course are interspersed with scenes of Sayid being taken prisoner (“Solitary”).

Yet who says genre should be unified anyway? Would Aristotle really have approved of a play that was unrelentingly tragic? Unlikely. Surely, even Ajax could provoke a giggle or two, depending on exactly how the actor played the part.

Another Aristotelian rule concerns realism. Thus, someone might object to Shakespeare’s The Tempest on the grounds that it demands that we believe in a magic island where witches and various types of monsters lurk. Likewise, the polar bear and the Smoke Monster of Lost might put viewers off.

But who’s to say that what we’re doing when we are watching these sorts of productions should be described as “believing” anything? For my part, I don’t find Shakespeare’s magic island any less believable than the Dharma Initiative. Yet I’d have to be very literal- (not to say narrow-) minded to let that get in the way of my enjoyment. Indeed, suspending disbelief is an important part of the fun. More on this to come.

The Aristotelian tradition has two things to say about the presentation of the characters in a play. One is that there should not be too many, and the other is that they should be consistent during the course of the action.

The first of these can be applied to Shakespeare’s King Lear, a chaotic business in which lots of men with the names of English counties shout at one another. For sure, telling your Northumberland from your Westmoreland takes a bit of work to begin with, but it is a labor of love! Consider the average soap opera. Although soaps repeatedly introduce “your-mother-is-your-sister-but-your-uncle-doesn’t-know” sorts of complications, they are followed by millions of uncomplaining viewers.

Of the forty-eight survivors of Oceanic flight 815, only relatively few—hardly a quarter of the total, when you think about it—come into any sort of focus. The rest have little more than walk-on parts. Likewise with the Others: most of them do little more than stroll about on the lawns. In this sense, Lost is hardly more abundant in characters than the average TV show.

As to the idea that the persons depicted should be consistent over time, Aristotle seemed to mean by this that each person should correspond to some virtue or vice or other stable character trait. Yet we have to be very careful not to interpret this in a way that contradicts Aristotle’s rule about realism. After all, people don’t stay the same; they change, as does Shakespeare’s Henry V, when he goes from listless prince to brave king. Aristotle may simply have meant that the decisions a character makes at any given stage should be psychologically plausible. In any case, if, again, our point of reference is the epic (or the soap), lapse of time and variation in influences can make significant differences to temperament.

We may consider a couple of cases where the stability-of-character criticism might be applied to Lost. Perhaps the least problematic is that of Kate. Once we grasp why she led the tear-away life she did before being arrested, we can understand why, on the island, she behaves, as Jack testifies at her trial, as someone who cares for others (“Eggtown”). It’s not Kate who’s changed but her circumstances. Perhaps something of the same can be said about Sawyer.

Slightly more demanding is the case of Locke. In terms of psychology, his rugged individualism remains pretty constant. What does, of course, change is his physical state. He was in a wheelchair on boarding Oceanic flight 815 and gains the use of his legs once on the island. It’s only when we first see him in flashback (“Walkabout”) that we begin to have ominous thoughts about the healing powers of the island. If anything, this transformation—not to mention the later one when he returns to the island in a coffin (“There’s No Place Like Home: Part 3”)—is a challenge to what we are prepared to believe. But, as I said before and we shall see again, strict believability is not really the point: once we grant Locke’s situation, his responses to it are what catch our interest.

The case of Ben is altogether more puzzling. As we try to find some principle or project that drives his various behaviors and attitudes, we suppose there must be something. he’s up to, but it is hard to tell what. At some level, much of his motivation derives from his vendetta against Charles Widmore. Yet the various positions and expedients he adopts seem to fall into the category of the predictably unpredictable. Ben makes me think of Shakespeare’s character Iago: someone whose actions, for good or ill, seem underdetermined. As with Iago, what makes Ben interesting is that it is hard to guess what he’ll say or do next.

Two other rules laid down by the Aristotelian tradition deal with limits on space and time. Concerning space, Aristotle suggested that the action of a play should take place in a single location. This follows from the physical configuration of theaters from Ancient Greece down to at least the time of Shakespeare: the substantial lack of props meant that it was hard to signal clearly that the action had moved from, say, the royal court to a tavern or a graveyard. But with the modern means to make obvious the difference between a scene set on the island and one set in an L.A. psychiatric institution (even when they are both actually filmed in Hawaii), this sort of criticism is a bit hollow if leveled at Lost.

A more aesthetic consideration in favor of the unity of place derives from the idea of the unity of plot. Yet also in this case, we may say that the island provides the spatial focus for everything else that goes on, and the backstories set elsewhere help us understand the problems of the individuals we find there. Even though they are spread out in space from Iraq to Australia, from Britain to the United States, these background episodes are funneled through the check-in at Sydney Airport. And on the island itself, we come to identify certain sites, such as the camp on the beach, the Dharma bunkers, and the Others’ compound, as being places where the action is most decisive.

I submit that Lost is in the clear with regard to space and the other Aristotelian rules so far considered. Although Lost may sometimes push up against the limits of what viewers can handle by way of coordinated action and coherent character, it is not in flagrant breach of the Aristotelian standard of evaluation. Neither Aristotle himself nor Shakespeare and his admirers should object to the complexity of Lost, whatever some readers of Aristotle may say.

What about time, though? This question deserves careful attention.

“We Have to Go Back”

According to Aristotle, a tragedy should recount the action of not much more than a day. Although a television series of 120 episodes need not be this limited, a single episode that observes this rule helps the viewer keep track of things.

In its first three seasons, Lost uses flashbacks much more than most TV shows do. This doesn’t cause real headaches, because the survivors come to life more if we know about Jack’s “Daddy Issues,” Kate’s criminal deeds, and Hurley’s lottery win. Yet the final scene of the last episode of season 3 (“Through the Looking Glass”) introduces a very unusual sort of complication.

We’ve been watching scenes of Jack bearded and drink-and-drug-sodden but still capable of saving people from car wrecks. All the while, we’ve been assuming, perhaps somewhat uneasily, that they are flashbacks. What a shock, then, when this Jack meets Kate out near LAX and says, “We have to go back.”

Up to this point, all of the off-island business we have seen is at least consistent with being earlier than 2004. Suddenly, just as things seem to be coming to a close (we know that this is the last episode in the season, and we’re a bit afraid that there won’t be a fourth), we are shown a meeting that, at the moment of first viewing, admits of two interpretations.

In one interpretation, Jack and Kate knew each other before boarding Oceanic flight 815—but this won’t hold water. The sequence of their relationship—meeting after the crash, getting to know each other, and falling in love—couldn’t have been a pretense. So we have to revise our assumption that what we are seeing is a flashback.

In the other interpretation, even if we have become accustomed to flashbacks as the narrative mode of Lost, we are pushed to understand “We have to go back” as a return to the island, meaning Jack and Kate have already left the island. Meaning we are at a date later than the narration of the preceding three seasons. After all, the on-island action into which this scene is inserted has a freighter arriving on the island. So we are ready to believe that the survivors are about to be saved.

As soon as I got over the shock, my first thought was, Well, at least we can look forward to a fourth season!

Then a second thought kicked in: Now that we have seen the “We have to go back” meeting, everything that happens on the island and whatever means Jack and Kate find to get off the island cannot not have their meeting as its outcome. The narrated time up to this point has counted as the past and the present. We know the past through flashbacks to off-island incidents, and we take the on-island narrative as the narrative present. Suddenly, though, just as Hurley and Desmond see Naomi parachute in before she actually does so, we can “see the future,” and the future contains—already contains—Jack meeting Kate out near LAX.

I want to look a bit harder at what it can mean for the future relative to the freighter’s approach to the island already to “contain” the meeting between Jack and Kate. There is a separate and very difficult question about what it might mean to “see the future.” Yet I want to get clear why it might be puzzling to think that there is anything there to be seen.

The Course of the Future

To get a grip on why there’s a problem here, it is a good idea to make a couple of distinctions. (This is a standard philosopher’s trick to delay having to give an answer.)

First, we must distinguish a little bit more carefully between the narrative time of the characters’ lives and the viewer’s time in watching Lost on TV or DVD (assuming that the viewer respects the sequence of the seasons and the episodes). In one sense, the narrative time begins on September 22, 2004, and the events can be ordered as a sequence of presents from that point on. In another sense, the times of the flashbacks are earlier than that date and make up the past relative to what is happening in on-island time. In the sequence that the viewer sees, narrative times earlier than September 22 are spliced into times later than that date. This, if you like, is a description of what a flashback is: the past of the narration is shown as present to the viewer. In terms of this distinction, we can say that a flash-forward is showing the future of the narration as present to the viewer.

Second, we must distinguish two ways of understanding time itself. According to one way, the whole history of the world is, in some sense, already fixed or determined or written or scripted, and the relations of before and after, and of earlier and later, among events do not themselves change. In the other way of thinking, as time passes and the date of the present becomes successively later and later, events come into being as they are produced by what went before them. The English philosopher John McTaggart (1866–1925) first called attention to these two different ways of thinking about time. Philosophers have come to call the first position eternalism and the second presentism.

Because it is not immediately obvious what difference the distinction between eternalism and presentism might make, it may help to give a little bit more detail about these two views.

Eternalism is the view that a sentence such as “Oceanic flight 815 has crashed” is, in a certain sense, incomplete as it stands. To say what makes a sentence like this true, we have to separate two elements. The first is the element that describes a kind of event. Thus, in the eternal sentence “Oceanic flight 815 crashes,” the verb “crashes” does not refer to a particular time, in just the way that the “is” in “two and two is four” does not refer to a particular time. So the second element is a relativization to a time or a date such as “on September 22, 2004.” In this view, then, “Oceanic flight 815 crashes on September 22, 2004,” can express the self-same truth whether someone says it in 1977 or in 2010. For eternalists, only sentences that spell out a date can express a genuine or complete truth about an event in time.

Presentism, on the other hand, takes it that there is nothing difficult about tense and no analysis is needed of “Oceanic flight 815 has crashed.” According to presentists, eternalism puts the cart before the horse in thinking that we have to use a system of time or date coordinates when we talk about what is happening “now.” Many presentists (including myself) think of the story of the world as becoming ever fuller and more complete as time passes: the future doesn’t (yet) exist, but what is happening and has happened are genuine facts in their own right.

McTaggart himself thought that because eternalism cannot give an adequate account of change over time and because presentism cannot give a satisfactory analysis of when the present is, time is not really real but rather an all-pervasive illusion. Most of his readers, however, have not wanted to accept this conclusion. Eternalists bear the burden of showing that their account of change is, after all, adequate, while presentists have to explain why there is no need to say when the present is (other than by saying what the time is now).

What difference does the difference between eternalism and presentism make toward understanding what a flash-forward is? For eternalists, there is no problem. The arrival of the freighter occurs long before the “We have to go back” meeting. The fact that we initially thought that it was a flashback and knew nothing of what happened in between is irrelevant. The distance in time between the two events is a fixed quantity, just like the distance in space between Sydney and L.A.

For eternalists, then, TV can use props and locations to show first a scene in Sydney and then a scene in L.A., or vice versa. There is nothing puzzling about this as long as we have some markers of the difference, such as the Sydney Opera House. Likewise, TV can use props and locations to show first a scene in 2004 and then a scene in 2007, or vice versa. And there is nothing puzzling about this as long as we have some markers of the difference, such as the state of Jack’s beard.

Most presentists, however, do see a problem. This can be expressed in terms of the viewer’s entering into the narrative present of the on-island affairs at the moment when Jack is calling the freighter at the end of season 3. From that point of view, there are lots of things that Jack has to deal with—Locke, Rousseau, and Ben are all causing trouble, and he has to do something about each person. That is to say, what he decides and does will make a difference to the outcome. From his perspective, the future is not fixed, because the way things will turn out depends on his actions. So whether and how he is to get off the island is not “there” yet.

Of course, we’re aware (perhaps somewhat distractedly) that Lost is scripted in advance, and there is nothing we, as viewers, can do to change the course of what has already been decided in building 23 of the ABC lot in Burbank. Yet when we are following Jack’s actions, that fact of fixity has to be put on hold. If we don’t put that fact on hold, we lose empathy and suspense: the sense that what is going on onscreen is present to us.

With the flash-forward, we have to adopt two sorts of attitudes at the same time. On the one hand, there is the attitude of seeing Jack call the freighter at the end of season 3, where his actions will make a difference to what happens next. And on the other, there is the attitude of seeing his actions in season 3 (that is, 2004) from the point of view of someone who knows about the meeting out near LAX, which occurs (or, if we prefer, recurs) in season 5 (that is, 2007), and so who knows Jack has already gotten off the island.

Eternalists will say that when all is said and done, Jack’s making his call at the end of season 3 is just as much part of the plot as his meeting Kate out near LAX in season 5. In this sense, eternalists take the position of observers standing outside the narrative, which includes the two events on the same footing. Presentists, by contrast, think that only one of these times can, at any given time, be the present (at that time). In this sense, presentists privilege the position of agents within a plot and can adopt only one position at a time for the purposes of seeing the plot through.

What’s more, presentists say that if—and presentists say that it is a very big “if”—there is, at the time of the call to the freighter, a fact about the meeting out near LAX, then there is nothing that Jack can do or fail to do between those two times that will make a difference to whether the meeting occurs. Eternalists are committed to saying that there is such a fact because if the sentence “Jack meets Kate in 2007” is ever true, it is always true. So it is also true in 2004. For this reason, presentists say that eternalism implies (or indeed is identical with) a view known as fatalism.

The Shape of Things to Come

In Lost, there is a great deal of talk about fate and destiny. A lot of it comes from Locke, who has a habit of appealing to fate when he is trying to get people to make up their minds in a certain way. For example, he convinces everyone to return to the island by appealing to what their destiny is (“The Life and Death of Jeremy Bentham”).

This is a bit perplexing, because it seems to play on something like the double take of the flash-forward. On the one hand, if it really is Kate’s destiny to return to the island, then she’ll return there whatever she decides. And if it is not her destiny to return to the island, then she’ll not return there whatever she decides. Appealing to what her destiny is can’t really help her to decide. On the other hand, if there is no such thing as Kate’s destiny and she is free to decide, she shouldn’t be influenced by what anyone, including Locke, says is her destiny.

The most general sort of trouble with talk about fate or destiny is that it appears to be in conflict with what we take ourselves to be doing when we make a decision or perform an action. When we do these things, we generally assume that we are making something true that wasn’t previously true and that wouldn’t have been true if we hadn’t done what we are doing. In a short-cut phrase, we believe that we have free and effective choice.

Yet fatalism denies that any choice is either free or effective, because fatalism is the view that everything that is ever true was always true. So either fatalism is false, or there is no such thing as what we generally understand ourselves to be doing when we exercise free and effective choice. What’s more, if fatalism is implied by (or is identical with) eternalism, then, if eternalism is the fundamental truth about time, there is no such thing as what we generally suppose ourselves to be doing when we make a decision or perform an action.

For myself, it seems wildly implausible to think that nothing I have ever decided or done has ever made anything true that wasn’t previously true and wouldn’t have been true if I hadn’t decided or done what I did. For instance, if I now decide to advise my gentle reader to reread the previous sentence, then that decision and my acting in accordance with it by typing these words is just what is making, at the time of typing, the present sentence the sentence it is, which it wouldn’t have been if I hadn’t so decided and typed in an exercise of free and effective choice. In this sense, I cannot bring myself to believe that fatalism is a true doctrine. That being so, and given the intimate relation between fatalism and eternalism, I cannot bring myself to believe that eternalism is the fundamental truth about time.

I admit that eternalism might be the fundamental truth about time, even though I cannot bring myself to believe it. And I think I can see why, whether it is true or not, eternalism might be attractive and believable to many people. One very strong attraction of eternalism lies in the effort to see the succession of the events that make up the whole history of the universe from a point of view outside that succession. Whatever else it is, the external point of view is more complete and objective than that of any of the partial and subjective positions from within the sequence of events.

The ideal of completeness and objectivity is not only noble, it is what scientific endeavor is all about. It is also the perspective that many philosophers and theologians have attributed to God, but we have to pass on that one for now. The trouble is that as of the time of my gentle reader’s reading this sentence, the history of the universe is not yet complete. Unless you, gentle reader, are so radically unlucky as to spend the final moment of the Existence of Anything poring over this page of The Ultimate Lost and Philosophy, there is still some future to be filled in. Hence the external point of view is not yet there to be occupied, and no complete and objective story can yet be told.

In a certain sense, eternalism fixes us within a plot, whether it was scripted in building 23 of the ABC lot in Burbank or elsewhere, about whose later phases we just happen not to have enough flash-forwards. Presentism, on the other hand, will say that there isn’t yet anything to have a flash-forward on. In either case, there is no question of whether we can “change” the future. For the eternalist, the future is just there waiting for us to experience it. For the presentist, we do what we can to make things happen that wouldn’t otherwise have happened, by deciding and acting, by exercising free and effective choice.

Whatever Happened, Happened

If deciding and acting don’t change the future, what about changing the past? Although the difference between eternalism and presentism divides philosophers into two heavily armed camps, the question of whether the past can be changed is, relatively speaking, a side issue. Almost all philosophers are in substantial agreement: No, sir, it cannot. Aristotle regarded the past as necessary, and St. Augustine (354–430 C.E.) thought that not even God can change the past. Surprisingly for philosophy, almost everyone who has thought about the matter has followed suit.

Only a few philosophers—plus the odd (both in the sense of “rare” and in that of “cranky”) theoretical physicist—have tried to find some sense in the idea of bringing about what didn’t happen. They are backed up by a grand tradition of science fiction tales, to which we now add Lost, beginning with the first episode of season 5 (“Because You Left”). The idea of changing the past exerts a great fascination, perhaps because it fulfills a pretty deep and widely felt wish. If it were possible to change the past, the pains of regret and remorse could sometimes be relieved. Almost everyone can think of a bit of the past he or she would like to be able to change, to do what was left undone (regret) or to undo what was done (remorse). So, nearly almost everyone would like time travel to be possible.

Interestingly, both eternalists and presentists deny that the past can be changed, and for very similar reasons. Eternalists will say that given a certain (complete and objective) history of the universe, which is made up of all of the truths there are, adding something else that is inconsistent with one of those truths will produce a contradiction. Presentists will say that given the history of the world so far, if we say of a certain past time that something both did and didn’t happen, then we have a contradiction.

Why should contradictions bother us? After all, didn’t Walt Whitman say in “Song of Myself,” “Do I contradict myself? / Very well then I contradict myself / (I am large, I contain multitudes)”?

One thing about contradictions that bothers logicians—and most conscientious philosophers have a touch of the logician in them—is the fact that a contradiction is never true. Let’s take, as an example of the basic form of a contradiction, the sentence “Hurley was in the Dharma Initiative in 1977, and Hurley was not in the Dharma Initiative in 1977.” This sentence is made up of the affirmation that Hurley was in the Dharma Initiative in 1977 along with its negation. If the affirmation is true, then the negation is false, and if the affirmation is false, then the negation is true. Our sentence is made up of these two parts by way of an “and.” A sentence in which “and” holds the parts together is true only if both of the parts are true, and if even one is not true, the sentence as a whole is false. Either the affirmation that Hurley was in the Dharma Initiative in 1977 or its negation must be false, granted that the other is true. That being so, the sentence as a whole is false.

Whether we take the eternalist view, from the complete and objective standpoint of the entire history of the universe, or the presentist view from a moment in, say, 2007, then either Hurley was in the Dharma Initiative in 1977 or he wasn’t. Hence, there was nothing that Hurley could do in 2007 to make it true that he was in the Dharma Initiative in 1977 if he hadn’t been in the Dharma Initiative in 1977, and there was nothing that he could do in 2007 to prevent his being in the Dharma Initiative in 1977 if he had been in it in 1977. If there is a past, then it cannot be changed.

The other thing that bothers logicians about contradictions is that they do indeed, as Whitman said, contain multitudes, but rather more multitudes than Whitman himself could contain. We’re not and Whitman was not, after all, quite as large as Whitman thought he was.

The argument for this is swift, decisive, and absolutely general. It is swift in the sense that it can be presented in five easy steps. It is decisive in the sense that it depends only on the meanings of the really basic words and, not, and or. And it is absolutely general in the sense that any sentences whatever can be substituted for the example I offer of the situation in which Hurley finds himself in 2007, just before the moment at which his next experience will be that of being in the Dharma Initiative in 1977.¹

Granted that a sentence in which and holds the parts together is true only if both of the parts are true (step 1), from “Hurley was in the Dharma Initiative in 1977, and Hurley was not in the Dharma Initiative in 1977” we can deduce (step 2) either of the parts, such as “Hurley was in the Dharma Initiative in 1977.” Now, from any affirmation, we can deduce (step 3) a sentence in which the parts are held together by “or,” because such a sentence will be true so long as at least one of its parts is true. Thus, from ”Hurley was in the Dharma Initiative in 1977” we can deduce, with due respect to Douglas Adams, “Hurley was in the Dharma Initiative in 1977 or the Answer to the Question of the Meaning of Life, the Universe, and Everything is 42.” Now, using same principle by which we deduced at step 2 “Hurley was in the Dharma Initiative in 1977” from our starting assumption of a contradiction “Hurley was in the Dharma Initiative in 1977, and Hurley was not in the Dharma Initiative in 1977,” we can deduce (step 4) “Hurley was not in the Dharma Initiative in 1977.” Now, if an or sentence is true and one of its parts is false, then it must be the other part that’s true. So, taking the product of step 3, ”Hurley was in the Dharma Initiative in 1977 or the Answer to the Question of the Meaning of Life, the Universe, and Everything is 42” along with the product of step 4 “Hurley was not in the Dharma Initiative in 1977,” we can deduce (step 5) that the Answer to the Question of the Meaning of Life, the Universe, and Everything is 42. Some people find this last move a bit hard to follow, but if we think about a sentence like “My wallet is on the table, or my wallet is in my pocket,” we can see that as soon as I know that my wallet is not in my pocket, I can deduce that it is on the table.

The point about the argument just outlined is that from a contradiction, anything and everything follows. So, to assert a contradiction is to assert anything and everything, including all of the possible Answers to the Question of the Meaning of Life, the Universe, and Everything and, among them, also the true one. The trouble is that given all of these multitudes contained in a contradiction, we still don’t know which is the true Answer to the Question of the Meaning of Life, the Universe, and Everything (or, indeed, where my wallet is).

If time travel is possible, we can change the past—but changing the past leads to contradiction. A contradiction is never true and implies anything whatsoever. So time travel never happens. When we try to imagine time travel, we have to accept at least one contradiction and, consequently, all of the things that follow from it—namely, every possibility whatsoever.

The Total Experience

To return, in conclusion, to our “poetic” considerations about plots, we have already heard that Aristotle advised against introducing elements that run up against what is believable. If what we have said about changing the past is right, then we can see why the introduction of time travel as a key plot element is likely to have the effect of making the plot unbelievable.

Once time travel is introduced, it becomes harder to know which plot we are being asked to follow because it becomes uncertain what we are being asked to include and what we are being asked to exclude from the plot. This is because we don’t know where we are being pointed by a contradictory plot element: Whitman’s multitudes are too multitudinous for us to be sure how to slim them down to a single storyline.

Yet I have also admitted that the idea of time travel doesn’t bother most people the way that it bothers the philosophers who sit around theorizing about time as eternalists or presentists and who object to time travel as a generator of contradictions. That is to say, it isn’t time travel as such that causes problems for most people who watch science fiction or Lost. My guess is that most people—including logicians of either basic orientation in the philosophy of time—aren’t too bothered by time travel as a plot element in science fiction or in Lost. And I guess that they aren’t too bothered by it because, in one way or another, they shield it from its multitudinous consequences. What I mean by being “shielded” here is a cousin of the way that when we are following Jack’s call to the freighter, we put on hold the fact of his later meeting Kate out near LAX.

One way we can think of shielding is this: We’ve been following Hurley from his lottery win through the downing of Oceanic flight 815 and the psychiatric institution back in L.A. to his return to the island on Ajira flight 316. So we’ve built up a sequence of how things have seemed to him at the various stages. Each of these experiences can be given a date. What is more important for Hurley, these experiences form a sequence for him. If, then, after an experience that can be dated to 2007 there comes an experience that can be dated to 1977, why should Hurley be bothered? The 1977 experience is just the one that, for him, comes after the 2007 experience. That is, if we put ourselves in the shoes of the time traveler, there is a sequence that doesn’t too obviously lead to a contradiction because it is just one thing after another.

We generate the shield in order to be able to follow how things successively seem to Hurley, and we put on hold the idea that there would have to be two mutually contradictory 1977s for Hurley to “go to” from 2007, one in which he is in the Dharma Initiative and one in which he is not.

We privilege the sequence that starts on September 22, 2004, because that’s the one we know about, and we put on hold the 1977 (in which Hurley perhaps wasn’t even born) that led up to Hurley’s boarding Oceanic flight 815. In privileging this sequence, we allow that there is a 1977 accessible from 2007 that has Hurley in the Dharma Initiative, even though that was not part of the past when he boarded Oceanic flight 815. We allow this because we are asked to by the people writing in building 23 of the ABC lot in Burbank. Given that they have done a good job in helping us shield the contradictions that arise from time travel, why not go with the flow?

What’s more, even if we are not sure why we sometimes run into trouble when we try to follow a plot with time travel in it, the attempt is a challenge. We enjoy the challenge, just as we enjoy the other twists and turns of the plot.

Aristotle, in the only explicit reference he made to the so-called unity of time, was right to be a bit vague in saying that the action of a tragedy should correspond to “a single passage of the Sun, or just a little more.”² He was vague, because it is not clear exactly where the upper limit to plot complexity lies. And he specified only one day’s events because he thought that was about as much as a viewer could take in at one sitting. Yet this leaves open the possibility—exploited by Shakespeare and others—that a longer span can be presented to and followed by an audience that is willing to take the right time-lapse cues and to interpret a change of props as a change of place, and so on.

Likewise, I want to suggest that even if I have philosophical worries about the genuine possibility of time travel, they don’t bother me much when I am watching Lost and shielding a plot that has time travel as one of its elements. I want to leave it an open question where limits might lie in the massive use of time travel, as in Lost’s seasons 4 and 5, as a plot device that calls on viewers simultaneously to shield many different experienced time sequences so as to keep the multitudes at bay. I suppose it depends on how much help viewers get from building 23, how nimble they are at keeping the sequences distinct, and how willing they are to do the work. Although I may not always succeed, I’m willing to give it a try because it’s all part of the fun.

NOTES

1 Even if readers don’t like footnotes and don’t like the symbols that logicians notoriously hide behind, this argument is so important and so general that I want to offer a schematic version of it, in which p stands for any proposition whatever and q stands for any proposition whatever:

1. p and not-p (assumption of a contradiction)

2. p (from 1, by the meaning of “and”)

3. p or q (from 2, by the meaning of “or”)

4. not-p (from 1, by the meaning of “and”) Therefore

5. q (from 3 and 4, by the meanings of “or” and “not”)

2 Aristotle, Poetics, 1449b13.