The stories of J. J. Abrams’s Lost (2004–2010) and Star Trek (2009) take us back in time. What makes these time-travel narratives stand out is how they abide by the logical prohibition against changing the past and display an understanding of the logical problems involved in doing so. However, Lost and Star Trek differ in how they handle this prohibition. While the characters of Lost travel into their own past but are unable to change anything, the characters of Star Trek are able to change things but do not travel into their own past. In neither case do the characters travel back into their own past and make changes.
But why can’t the past be changed? Why would logic prohibit changing the past, but not time travel, parallel universes, or alternate timelines? After all, none of these ideas seems more plausible than any other. Furthermore, how could it be logically possible to travel back into one’s own past if it wasn’t possible to change it? These are the questions that form the basis of this chapter, and they are also the kinds of questions that one would encounter while investigating the philosophy of time.1 To better understand the unique logical problems involved with changing one’s own past, I will begin by explaining the philosophy behind the prohibition before moving on to describe how this shapes the narratives of both Abrams’s Lost and his Star Trek.
It isn’t possible to travel through space faster than the speed of light. To the best of our present knowledge, physical law prohibits doing so. However, there is another sense in which faster-than-light travel is possible. Things could have been different, or, more specifically, physical law could have been different such that our universe placed no upper bound on speed. This is why it is something of a mystery as to why our universe has the particular physical laws it does rather than some other set; that is, given that our physical laws could have been different, why are they the way they are?
What this illustrates is that there are different kinds of possibility and impossibility. Since the physical laws investigated by science prohibit faster-than-light travel, it can be said to be physically impossible to do so; that is, impossible given the physical laws that actually govern our universe. However, there does not appear to be anything in logic that rules out the possibility that the universe could have been governed by a different set of physical laws. Given this, faster-than-light travel can still be said to be logically possible; that is, possible given the laws of logic.
It also isn’t possible to draw a figure that is both a square and a circle. However, there is nothing in physical law that prevents me from doing so. Instead, it is the laws of logic that make this impossible. Which is to say, it is not logically possible to draw a figure that is at once both four-sided and yet also only one-sided. Furthermore, unlike the physical law, the laws of logic could not have been different than they actually are.2 In fact, this might be the distinctive feature of a law of logic that distinguishes it from a physical law. Consequently, there is no way things could have been different so that I could draw a circle-square.3
Where this is all relevant to our examination of time travel is concerning whether we are trying to answer the question “Is time travel physically possible?” or whether we are trying to answer the question “Is time travel logically possible?” If time travel isn’t logically possible, then it isn’t physically possible either. However, time travel could be logically possible and yet physically impossible. Which is to say, time travel might be allowed by the laws of logic but prohibited by the physical laws that actually govern our universe—as faster-than-light travel happens to be.
As an answer to the question “Is time travel physically possible?” the physical laws that govern our universe seem to be surprisingly amenable to time travel. This could turn out to be mistaken, but for now it is an open question among scientists whether anything in physical law rules out time travel, and so far, nothing appears to. But physical possibility is the domain of science and the scientist. Since this is a work concerning philosophy and its place in the work of Abrams, I will be concerned only with what philosophers have to say about the logical possibility of time travel and changing the past. Furthermore, when people express a skepticism about the possibility of time travel they usually do so because they think of it as logically impossible. The question, then, that will set my agenda in this work will be “Is time travel logically possible?” rather than “Is time travel physically possible?”4
A paradox is a set of propositions that independently appear to be true but that together appear to give rise to a contradiction. Which is to say, while they each seem to be true by themselves, it seems that they cannot all be true together. There are two ways of resolving a paradox. Either we can reject one of the propositions that give rise to the contradiction, or we can show how (despite appearances) no contradiction arises even if all the propositions were true.
The most famous reason for thinking time travel might not be logically possible comes in the form of a paradox. The so-called grandfather paradox can be thought of as the collection of the following independently plausible propositions (assuming time travel is possible), which together give rise to a contradiction:
(P1) It is possible for me to go back in time.
(P2) If it is possible for me to go back in time, then it isn’t possible that I never came to exist.
(P3) If it is possible for me to go back in time, then it is possible that I never came to exist.
The first proposition encapsulates what we are taking for granted, that time travel is possible. The second proposition points out the apparently incontrovertible fact that for me to be capable of going back in time, then I must have come to exist at some point in time if I am to do so. And the third proposition encapsulates the fact that if I could go back in time, then I would seem to be capable of doing things like killing my paternal grandfather before my father was born (i.e., things that would ensure that I never came to exist). Given that all three of these propositions cannot be true together, we are faced with a paradox. To resolve the paradox we must either reject one of the propositions as false or we must (cleverly) find a way in which they could all be true without giving rise to a contradiction. Since there doesn’t seem to be any way around the contradiction these propositions give rise to and given that we are trying to investigate the possibility of time travel, we should see if we can reject either proposition P2 or P3 first and abandon proposition P1 only if it is our last resort for resolving the logical paradox. If there is no other way to resolve the grandfather paradox than to reject P1, then we will know that it is not logically possible to go back in time.
Proposition P2 seems the hardest to justify rejecting. How could I go back in time if I never came to exist in the first place? We might try making sense of the idea by saying something like, “I came to exist in the past before I went back in time and changed the past,” but this notion doesn’t stand up to close scrutiny. Notice how such a response relies on the notion of “before.” However, what is it for me to have come to exist before I go back in time and change the past other than that I came to exist in the past? If I were to change the past so that I will no longer come to exist at any point in the past leading up to when I travel back in time, then there will be no time in which I came to exist before I traveled back in time. It does make sense to say that I could travel back into or create an alternate timeline where I kill my own grandfather and thereby never come to exist, but then my original past would still exist as it was and in that sense I will not have actually changed my own past.5
Holding onto proposition P1 until our only option is to abandon it, we might try rejecting proposition P3 to resolve the grandfather paradox. That is, we might reject the idea that if I could go back in time, I would then be able to change the past in ways that would ensure I never came to exist. However, if it is possible for me to go back in time, then what would prevent me from doing things like killing my paternal grandfather before my father was born? We might try to say that logic would prevent me, but this runs contrary to the way the laws of logic work. For example, suppose that I build a time machine and with that time machine I were to go back in time with an arsenal of weapons and the goal of killing my grandfather. Will the laws of logic make my weapons misfire? Make me slip on a banana peel? Make my bullets bounce off of my grandfather as if he were Superman? The problem is that the laws of logic do not work like physical laws. Logic doesn’t prevent me from drawing a figure that is both a square and a circle by repeatedly breaking my pencil or causing the paper I am trying to draw the figure on to burst into flames. Rather, what the laws of logic show is that there is just no such figure for me to draw. Logic is not some force in the universe guiding banana peels beneath time travelers’ feet in order to avoid contradictions, and so if logic supposedly prevents me from killing my grandfather then there would need to be some less “active” way for it to do so.
The result is that while logic doesn’t appear to give us grounds for rejecting either proposition P2 or proposition P3, logic does demand that we reject one of the three propositions that give rise to the contradiction and so we appear to be forced into rejecting proposition P1. That is, forced to say that it is not logically possible for me to go back in time. Or at least this is how things appeared until some recent work in philosophy breathed new life into the idea that instead of rejecting proposition P1, the correct resolution of the grandfather paradox is the rejection of proposition P3.
The problem with rejecting proposition P3, that is, allowing time travel but disallowing time travelers from doing the sorts of things that would generate a contradiction, is that logic isn’t in the business of causing guns to jam, banana peels to appear, or bullets to bounce off flesh. What is needed in order to reject proposition P3 is some way in which the laws of logic entail that one of these coincidences will occur without having to say the laws of logic are acting like a physical force pushing objects around in space to actively cause these coincidences.
Philosophers began to realize how logic might entail these coincidences with the publication of David Lewis’s article “The Paradoxes of Time Travel” in 1976. Since then, Lewis’s ideas in that paper have been further developed and defended by philosophers like Nicholas J. J. Smith and Theodore Sider and a consensus has been rapidly forming among philosophers that the correct resolution of the grandfather paradox involves rejecting proposition P3 rather than proposition P1.6 In the following passage, Sider offers a helpful way of understanding the line of thinking that changed philosophers’ minds about time travel by drawing our attention to a more mundane example of how logic might entail a coincidence:
Suppose I tried to throw a heavy stone at a fragile window. Since I have good aim and a strong arm, the window would break. I might, I suppose, slip on a banana peel, or hit a bird passing by with the rock, or have my throw deflected by a great gust of wind, or have a sudden failure of aim despite many years of training in stone-throwing. But at the very least, it surely is not the case that one of these strange coincidences would happen. The [proposition:] If I were to try to throw the stone at the window, I would slip on a banana peel or hit a passing bird or . . . is false.7
That is to say, it might occur that some coincidence will prevent me from breaking the window, but it would be false that to say that such coincidences would occur. This is also the problem we faced in trying to reject proposition P3 to resolve the grandfather paradox. It might occur that some coincidence would prevent a time traveler from killing his or her paternal grandfather before his or her father is born, but there does not seem to be any reason we can say such coincidences would occur—at least without mistakenly thinking of the laws of logic as a force in the world ensuring that they would. But then Sider considers a different proposition: “If I were to try to throw the stone at the window but the window did not subsequently break, then I would slip on a banana peel or hit a passing bird or . . .” Unlike the first proposition, this one is true. Sider explains that the difference lies in the fact that this second proposition’s antecedent (i.e., what appears after the “if” but before the “then”) already contains circumstances that could only come about through an unlikely coincidence: “Here I have built my failure into the antecedent; the [proposition] concerns what would happen had I tried and failed. Here, I think, our sense is that the [proposition] is now true. Given the background facts, the only way for me to fail to hit the window would be for some strange coincidence to occur.”8 Which is to say, it is only because we already know that the window will not break, that we can know that a coincidence not only might prevent me from breaking it but that a coincidence would prevent me from breaking it. Furthermore, if we already know that the window doesn’t break, then we need not think of logic as actively causing a coincidence to prevent me from breaking it. Given the unlikely setup and result contained in the antecedent, logic entails that an unlikely coincidence will occur but without having to cause that coincidence. And, in fact, the same thing can be said about time travelers attempting to kill their paternal grandfather before their father is born.
Suppose I travel back to a time before my father was born and attempt to kill my grandfather. Supposing that I have good aim, a reliable weapon, and so on, some unlikely coincidence would have to occur for me to fail at my task. Yet we already know that I do fail. Since I exist, we already know that my grandfather does live long enough to conceive my father and, consequently, we also already know that I fail in my attempt to murder him. Which is to say, we know that the following proposition is true: “If a time traveler were to try to kill his paternal grandfather before his father was conceived, he would slip on a banana peel or have a sudden change of heart or . . .” Such coincidences might seem so unlikely that we cannot say for certain that they would happen without having to also say that logic would actively cause them to occur, but in fact we can know that they would occur without seeing logic as playing such an active role since we also already know that an unlikely result occurs (that my grandfather survives) given the setup (my capability and motivation to kill him). It is only if we ignore the fact that we know that my paternal grandfather survives at least long enough for my father to be conceived, that it would be mysterious how we could know a coincidence would occur to prevent me from killing him. However, given that we do know that my paternal grandfather survives at least until my father is conceived, our knowledge that a coincidence would occur to prevent me from killing him before that time is an unremarkable logical entailment.
Of course, this line of reasoning rules out more than time travelers simply killing their own grandfathers, it also rules out their changing the past in any way. Suppose that instead of setting out to kill my grandfather, I were to travel back in time with the aim of assassinating Adolf Hitler in 1930. Given that we already know that Hitler lived until 1945, we also already know that I will fail at assassinating him in 1930. Or, to put it another way, we know that I have already failed to kill him. While if I had access to a time machine and were myself a skilled assassin it might require some unlikely coincidence to prevent me from killing Hitler in 1930, given that we already know that Hitler lived until 1945 we also already know that some coincidence prevented me from assassinating him in 1930. If I had been successful in assassinating Hitler in 1930 our past wouldn’t change, it would have always been different. There is only one past and it cannot be changed. Either Hitler died at the hands of a time traveler in 1930 or he lived until 1945. If a time traveler wanted to go back and kill Hitler in 1930, then his only hope is the fact that recorded history might be wrong about the date of Hitler’s death. If recorded history is correct, then the time traveler already knows that some coincidence prevented him from being successful once he arrived in his past.
The result is that while it is logically possible to travel back in time and affect the past, it isn’t possible to change it. If I were to travel back in time and contribute to the events of the past, then it will have always been the case that I affected the past in the way I did. Which is not to say I cannot travel to some parallel universe or alternate timeline that exactly resembles a point in my own past and cause the past of that universe or timeline to unfold differently than my own. However, then I would not be changing my own past but merely contributing to the past of another universe or timeline. If I travel back to my own past, then logic entails that what I will do is already contained in that past.9
If the philosophy of time travel allows that we can travel to our own past and affect it but not change it, then storytellers have three formats they may follow in telling a time-travel narrative:
(F1) allow backwards time travel and allow characters to change the past in violation of logical constraints,
(F2) allow backwards time travel but disallow characters from changing the past, or
(F3) allow so-called backwards time travel to a parallel universe or alternate timeline that is similar to the characters’ own past and allow those characters to affect the past of that universe such that it comes to differ from the past of their own universe or timeline.
Usually time-travel stories are instances of format F1. Famous examples include the Terminator and Back to the Future series of films. One thing that is noteworthy about both Abrams’s Lost and his Star Trek is that while they do not follow the same time-travel-story format, both avoid the philosophically problematic format F1. While this fact alone would be enough to show some appreciation of the complexities of time travel, Lost goes beyond merely avoiding format F1 in making the question of which format Lost follows a major plot point of the show’s last two seasons and—in a remarkable way—the cliffhanger ending of Lost’s fifth season.
To illustrate how the philosophy of time travel becomes such an important question in Lost, let’s consider a few examples of how the characters of Lost raise the question. Here is Daniel Faraday in the first episode of the fifth season (“Because You Left”), explaining that while it is possible to travel back into one’s own past it still isn’t possible to change it—suggesting that Lost is an instance format F2: “Time—it’s like a street, all right? We can move forward on that street, we can move in reverse, but we cannot ever create a new street. If we try to do anything different, we will fail every time. Whatever happened, happened.” Then, in the eleventh episode of the fifth season (“Whatever Happened, Happened”), Miles attempts to explain to Hurley, Kate, and Jack why he knows Sayid’s attempt to kill young Ben in the past will fail:
HUGO “HURLEY” REYES: [Hurley looks at his own hand mystified]
MILES STRAUME: What the hell are you doing, tubby?
HURLEY: Checking to see if I’m disappearing.
MILES: What?
HURLEY: Back to the future, man. We came back in time to the island and changed stuff. So if little Ben dies, he’ll never grow up to be big Ben who’s the one that made us come back here in the first place. It means we can’t be here and therefore, dude—we don’t exist.
MILES: You’re an idiot.
HURLEY: Am I?
MILES: Yeah, it doesn’t work like that. You can’t change anything. Your maniac Iraqi buddy shot [Ben] Linus, that’s what always happened. It’s just we never experienced how it all turns out.
HURLEY: That’s really confusing.
MILES: Yeah, well, get used to it. But the good news is that Linus didn’t die, so that means the kid can’t either. He’ll be fine.
KATE AUSTEN: Didn’t look like he’s gonna be fine. What if you’re wrong?
MILES: Well, if I’m wrong then I guess we all stop existing and—none of it matters anyway then does it?
HURLEY: Let me get this straight. All this already happened.
MILES: Yes.
HURLEY: So—this conversation we’re having right now, we’ve already had it.
MILES: Yes!
HURLEY: Then what am I gonna say next?
MILES: I don’t know.
HURLEY: Ha! Then your theory is wrong.
MILES: For the thousandth time, you dingbat. The conversation already happened, but not for you and me. For you and me it’s happening right now.
HURLEY: Okay, answer me this. If all of this already happened to me, then why don’t I remember any of it?
MILES: Because once Ben turned that wheel, time isn’t a straight line for us anymore. Our experiences in the past and in the future occurred before these experiences right now.
HURLEY: Say that again.
MILES: [Miles tries to hand Hurley his gun] Shoot me. Please, please.
HURLEY: Ah ha! I can’t shoot you, because if you die in 1977, then you’ll never come back to the island on the freighter 30 years from now.
MILES: I can die! Because I’ve already come to the island on the freighter. Any of us can die! Because this is our present.
HURLEY: You said that Ben couldn’t die because he has to grow and become the leader of the Others.
MILES: Because this is his past.
HURLEY: But when we had captured Ben and Sayid, like, tortured him then why wouldn’t he remember getting shot by that same guy when he was a kid?
MILES: Huh. I hadn’t thought of that.
While for most of the fifth season the audience is given explanations for why the past cannot change like the two above from Faraday and Miles, the season is also littered with hints that this might be mistaken. For example, at the end of the above dialogue between Miles and Hurley, Hurley points out that if the past cannot change then older Ben should have remembered time-traveling Sayid shooting him as a child. This suggests that Miles might be wrong about the inability to change the past. Even Faraday comes to later question whether he is right about the prohibition on changing the past when, in the fourteenth episode of the fifth season (“The Variable”), Faraday postulates a (logically absurd) exception to the prohibition:
DANIEL FARADAY: But—we can change that. I studied relativistic physics my entire life. One thing emerged over and over—can’t change the past. Can’t do it. Whatever happened, happened. All right? But then I finally realized—I had been spending so much time focused on the constants, I forgot about the variables. Do you know what the variables in these equations are, Jack?
JACK SHEPHARD: No.
FARADAY: Us. We’re the variables. People. We think. We reason. We make choices. We have free will. We can change our destiny.
The problem with Faraday’s reasoning here is that the logical prohibition on changing the past isn’t a constraint on free will. The past cannot be changed, but if a time traveler has gone back and affected it, then the past happened the way it did partly because of that time traveler’s own free choices. The past already includes whatever free choices the time traveler makes, but they are still his free choices.
While this argument of Faraday’s is a philosophically interesting bit to include in the show itself, what is perhaps more impressive is what Faraday’s second-guessing of the logical prohibition on changing the past contributes to the story. While the audience has been told time and again by characters like Faraday that the past cannot change, there are also hints likes these that the characters might be wrong about this. What this sets up is the season finale of the fifth season (“The Incident”), in which the characters attempt to detonate a nuclear bomb in hopes of changing their past. The episode ends with a cliffhanger question: “Did the bomb go off? Can you—and did they—change the past?”
Season 6 of Lost begins without clear answers. There now appear to be two universes or timelines, one where the bomb has gone off and one where it hasn’t. The characters in the apparent universe or timeline where the bomb hasn’t detonated are left thinking that you can’t change the past after all, while the characters in the other apparent universe or timeline have never time traveled in the first place. The impression the story leaves on its audience is that Lost has unexpectedly turned out to combine formats F2 and F3 in an interesting and original way. While the characters had traveled back into their own past and were unable to change anything, because they tried to detonate the nuclear bomb in their past, when a parallel universe or alternate timeline branches off from that point it is a universe/timeline where the bomb has successfully gone off. What is interesting about this move is that while the inhabitants of the parallel universe or alternate timeline never travel back in time and detonate a bomb in their past, their universe/timeline shares a past with the original universe/timeline that includes time travelers from the future of the original universe/timeline detonating a nuclear bomb in their shared past. The characters of the original universe/timeline remain stuck with the hellish past they’ve been trying to change, but in attempting to change it they have unknowingly created a better parallel universe or alternate timeline for their counterparts to live in.
As season 6 progresses, one of the characters in Lost, Desmond Hume, comes to believe that this better parallel universe or alternate timeline exists and that it is possible for the inhabitants of the original universe/timeline to escape over to it. The audience is led to believe Desmond is correct about the idea of traveling to the parallel universe or alternate timeline due to the flashbacks the characters in the better parallel universe or alternate timeline begin to have of the original universe/timeline. This plot culminates in the series finale (“The End”), leading up to which Jack and Desmond argue over whether the better parallel universe or alternate timeline exists and whether they might be able to escape to it:
DESMOND HUME: This doesn’t matter you know.
JACK SHEPARD: Excuse me.
DESMOND: Him destroying the island, you destroying him. It doesn’t matter. You’re going to lower me into that light and I’m going to go somewhere else. A place where we could be with the ones that we love. And we’ll never have to think of this damn island again. And you know the best part Jack?
JACK: What?
DESMOND: You’re in this place. You know we sat next to each other on Oceanic 815. It never crashed. We spoke to each other. You seemed happy. You know maybe I can find a way to bring you there too.
JACK: Desmond, I tried that once. There are no short cuts, no do-overs. What happened, happened. Trust me, I know. All of this matters.
While at this point, the audience has been led to believe Desmond is right, Jack is proven correct in the end. What the other, seemingly better parallel universe or alternate timeline turns out to be is the afterlife. Lost reveals in its final episode that it has always been an instance of time-travel story format F2, that even while you can travel back into your own past, you won’t be able to change it. Although Jack does find a positive interpretation to give this logical prohibition in the end. In his exchange with Desmond, Jack argues that if the past could be changed then what we do in the present wouldn’t matter. Unless your past remains the same, then your actions will end up being erased by later changes in the timeline and in this sense it won’t matter what you do. Jack has given up on trying to change his unfortunate past, but he has now come to think that in having an unchangeable past his actions have value in a way they wouldn’t if the past could change. The question Jack now leaves the audience to wrestle with is not the logical possibility of changing the past, but how a changeable past would affect the value of our actions.
In an important way, the question of value Jack wrestles with in the finale of Lost is also one that plays a role in the way time travel is treated in Star Trek. If Abrams’s film about the young crew of the Enterprise had followed the lead of Lost and treated the past as unchangeable, then the writers of the new series of films would be committed to ensuring that their film is consistent with the many Star Trek shows and films that are set later in the franchise. This would mean, for instance, that the audience would know from the start that none of the central crew will die since they already know that they survive to be consistent with the shows and films in the franchise where their older selves appear. However, if Abrams’s Star Trek took format F1 and allowed the past to be changed, then it would wipe out the future of every other show and film in the franchise. Despite being fictional, for the events portrayed in those shows and films to be fictionally eliminated from the timeline would rob them of their value. Watching the new films inaugurated with Abrams’s Star Trek, we would be engaged in a story that results in every other Star Trek show and film no longer mattering because they will no longer have occurred.
The result is that Abrams’s Star Trek adopts format F3. Specifically, it is set in a parallel universe at a point similar to the past of the Star Trek universe we are already familiar with from other shows and films. In order to connect the story of this universe with the stories of the other Star Trek shows and films and in order to allow it to still vary from the universe of those shows and films, Abrams has several characters from that universe “time-travel” to his parallel universe and affect it in such a way that its past will now unfold differently than the past of their original universe.10
As we can see, time travel plays a different role in the story of Star Trek than Lost. In Lost questions over whether or not you can change the past, free will, and fate are central to the story. In Star Trek time travel serves as a practical storytelling tool, allowing Abrams to tell an origin story of the crew of the Enterprise that allows the future to unfold differently than the rest of the Star Trek franchise but without undermining the value of those shows and films by telling a story in which they are made to no longer exist.11
Still, there is a storytelling problem that time-travel narratives following format F3 risk falling into. Specifically, when characters don’t travel back into their own past and change things there, the question arises why they would care to change how the past unfolds in a parallel universe or alternate timeline? Star Trek avoids this problem by having its character Nero travel to the parallel universe in pursuit of the Spock of the original universe. Nero isn’t just out to kill Spock in as many parallel universes he can. Rather, the story’s antagonist is motivated by seeking revenge on the original Spock who he believes to have wronged him. That might still leave open the question why Nero takes his revenge on Spock by destroying the counterpart to Vulcan in this parallel universe, but we might allow for the purposes of the story that the original Spock is himself concerned with the welfare of his counterparts, even in other universes. It isn’t original Spock’s actual home world being destroyed, but it is the one inhabited by his species and family in this universe. Furthermore, madmen bent on the destruction of whole planets should not really be expected to be consistent in the motivation and thinking behind their acts.
In the end it is not a mark against a work of fiction if it does not abide by the constraints of logic. Both the Terminator and Back to the Future series of films do so with great success. However, a time-travel narrative that does follow one of the logically consistent formats is rare and therefore offers a writer the opportunity of a more original story. While Star Trek follows the more well-known logically consistent format for time-travel narratives, the film is noteworthy in that it consciously adopts the only format that both avoids devaluing the rest of the Star Trek franchise and frees the story it tells from being constrained by the future portrayed by the rest of the Star Trek franchise. Lost stands out not only because it adopts the most uncommon time-travel narrative format, but also because it finds a way to make the philosophical questions surrounding time travel and changing the past central components of the story it tells.12
Thank you to Em and Morgan Stinson as well as Cody Cox for their input throughout the process of drafting this chapter.
1. See, for example, Robin Le Poidevin and Murray MacBeth, eds., The Philosophy of Time (Oxford: Oxford University Press, 1993); Kip Thorne, Black Holes and Time Warps: Einstein’s Outrageous Legacy (London: W. W. Norton, 1994); John Earman, Bangs, Crunches, Whimpers, and Shrieks: Singularities and Acausalities in Relativistic Spacetimes (New York: Oxford University Press, 1995).
2. Which is not the same thing as saying that we couldn’t be wrong about the laws of logic. In this sense, it is possible that I could draw a circle-square. This wouldn’t mean that it is possible for the laws of logic to be different from what they in fact are, but rather that it is possible for the laws of logic to be different from what I believe them to be. Philosophers mark this distinction by saying that while it is epistemically possible for the laws of logic to be different (i.e., they might differ from what I believe them to be), it is not metaphysically possible for the laws of logic to be different (i.e., differ from what they in fact are).
3. Now, certainly, the words “circle” and “square” could have had different meanings than they do. We might be able to refer to those properties with different words or refer to different properties with those same words so that I would be capable of drawing a figure that we would call both a square and a circle, but what would remain logically impossible is for one figure to at the same time have the properties of both of the figures that using current English we refer to with the words “circle” and “square.” Which is just to say, the claim that drawing a circle-square is logically impossible is not merely a linguistic claim about the words “circle” and “square” but rather a claim about the properties we refer to with those words.
4. Philosophers also deal with another kind of possibility called metaphysical possibility, which is different than the sort of metaphysical possibility that I contrasted earlier with epistemic possibility. This second sort of metaphysical possibility concerns what would be possible given the correct metaphysical theory. For example, given a metaphysical theory of time called presentism, the past no longer exists, and if this metaphysical theory were correct then it would be hard to see how it could be metaphysically possible to travel back into the past. However, presentism is a minority view among philosophers because of worries about how any propositions concerning the past could be true on this view. If the past no longer exists, then it does not appear that statements like “I existed five minutes ago” could be true given that absolutely nothing existed five minutes ago. Certainly everyone would agree that the past does not presently exist, but presentism is the rather extreme metaphysical position that the past does not even exist in the past. It might also be worth pointing out that there are also many more mundane forms of possibility that philosophers as a profession are not concerned with. For example, what is and isn’t financially possible for me. Given my current income, it might be financially impossible for me to afford my own private jet but still financially possible for me to afford a trip on a commercial airline. However, what I will be concerned with in this work is logical possibility.
5. The idea is essentially that time might have two or more dimensions (rather than the one we normally suppose it to have) and that while I would not be able to travel back to the exact time on the two-dimensional temporal plane and kill my own grandfather then, I would still be able to travel back to an adjacent time and kill my grandfather. In the following passage, philosopher David Lewis explains why this would still not constitute traveling back into one’s own past and killing one’s own grandfather: “On closer inspection, however, this account seems not to give us time travel as we know it from the stories. When the traveler revisits the days of his childhood, will his playmates be there to meet him? No; he has not reached the part of the plane of time where they are. He is no longer separated from them along one of the two dimensions of time, but he is still separated from them along the other.” Which is to say, the playmates the time traveler visits will not be the ones from his own past but rather their counterparts in a past existing adjacent to his own. A similar problem frustrates the attempt to kill one’s own grandfather rather than his counterpart in a timeline running parallel to one’s own. David Lewis, “The Paradoxes of Time Travel,” American Philosophical Quarterly 13 (1976): 145–52.
6. Lewis, “Paradoxes of Time Travel,” 145–52; Nicholas Smith, “Bananas Enough for Time Travel?” British Journal of Philosophy of Science 48 (1997): 363–89; Nicholas Smith, “The Problems of Backward Time Travel,” Endeavour 22, no. 4 (1998): 156–58; Theodore Sider, “Time Travel, Coincidences and Counterfactuals,” Philosophical Studies 110 (2002): 115–38.
7. Sider, “Time Travel, Coincidences and Counterfactuals,” 122.
8. Ibid.
9. Perhaps a better way of making this point is stating that while it is logically possible to go back and change the past, logic entails that as a matter of fact no one will. Which is to say, there is a logically possible way the world could have been where a time traveler would successfully kill Hitler in 1930, but since our past does not contain a time traveler assassinating Hitler in 1930 we know that the actual world won’t ever include any time travelers who go back in time and successfully kill Hitler in 1930. In this sense it can be said that it’s logically possible for me to go back in time and kill Hitler in 1930, only we already know that any attempt on my part happens to fail.
10. In Abrams’s Star Trek film time is not taken to be multidimensional with many timelines running parallel to our own, but rather the film assumes the many-worlds interpretation of quantum mechanics, where parallel universes branch off from our own during certain quantum events. In the film itself this is hinted at when young Spock is asked the question, “What is the central assumption of Quantum Cosmology?” and he answers, “Everything that can happen does happen in equal and parallel universes.” One of the most peculiar consequences of quantum mechanics is that it appears to say that things can enter a state of simultaneously possessing two conflicting properties. This consequence of quantum mechanics has entered popular culture through the example of Schrodinger’s cat being simultaneously both alive and dead. The interpretation of quantum mechanics that says there exist many other universes parallel to our own is an attempt to do away with this strange result by instead postulating that the cat is alive in one universe while dead in another.
11. One might be concerned, however, that by following format F3 for the Star Trek film there is a conflict with the many other franchise shows and films that follow format F1. However, while format F1 and format F2 could clearly not coexist, there is nothing preventing a story from involving both format F3 and either format F1 or F2. Which is to say that, while the first two formats deal with traveling back into one’s own past, the third concerns traveling to a past-like parallel universe or alternate timeline and there is no reason to suppose that you couldn’t sometimes have a story that travels back into the character’s own past and follows format F1 while sometimes having a story that travels to a parallel universe or alternate timeline and follows format F3.
12. For more on why the worlds described by make-believe stories are capable of containing logical contradictions even while actual and possible worlds cannot, see Kendall Walton, Mimesis as Make-Believe (Cambridge, MA: Harvard University Press, 1990).