It is 9:00 am. Tad and Willie are in the lab talking over cups of coffee when Dr. Rufus arrives. It is obvious that none of the three slept well.
Carlene: Good morning, gentlemen. Would one of you mind pouring me a cup? I was up all night thinking of ways to test our time-travel hypothesis.
Tad [pouring Dr. Rufus a cup]: That makes three of us, but maybe you had better luck than we did. I, for one, couldn’t get over the idea that we’re taking time travel seriously.
Carlene: Please try to get used to it and think about ways to test our hypothesis. Where’s the printout from yesterday?
Tad hands Dr. Rufus the cup of coffee and the printout of Trial 17.
Tad: Let me get clear on what the hypothesis is exactly. You’re claiming that—despite appearances—there’s only one particle whose trajectory is plotted here, and that time-travel departure occurs at t=5. We’re considering this hypothesis because it explains quite a lot; in particular, if true, it would explain both the lack of B-field disturbances at t=5 and the origin of what we thought was an anomalous second particle.
Carlene: That’s a fair enough account.
Tad: Okay, but I have a question about the supposed explanation of the second particle’s origin and decay. Sure, according to the hypothesis, its origin is no longer a mystery; we created it when we knocked the uranium into the helium; it’s a psi-lepton. But we still don’t have an explanation of what’s going on with that particle at t=3. That’s what was really bothering us, what was so unexpected and left unexplained by the theory. So, like you’re suggesting, what if that isn’t the origin of the particle at t=3? According to the time-travel hypothesis, it’s the time-reversed decay of the particle. Sure, I’ll buy that, too. But I don’t see that we have an explanation for why it decays.
Carlene: Excellent, Tad. That’s quite perceptive.
Tad: Yesterday we thought the time-travel hypothesis would account for the origin of the anomalous particle. What we’re saying is that it doesn’t exactly originate, at least not at t=3. That’s neat, but if we’re taking this idea seriously, there’s still something that needs to be explained. In your version, Professor, the anomalous particle isn’t coming into existence at t=3; it’s decaying. But why should it do that?
Carlene: It didn’t occur to me until late last night, but I think that the hypothesis predicts the decay, too. Take a look at the duration of the lifespan of the particle in Trial 17. It originates at t=0, behaves normally until t=5 nanoseconds, and then travels backwards in time for two additional nanoseconds; its lifespan until it decays—how long it exists from its perspective—is seven nanoseconds.
Tad: And?
Carlene: And of course you know that a duration of seven nanoseconds is precisely what our theory predicts for the lifetime of a psi-lepton. It’s also how long our normal psi-lepton lasted in Trial 16 and some of our earliest trials.
Tad: Huh, that’s clever, but I still don’t know.
Carlene: None of us knows much of anything at this point, I’m afraid. We’re doing little more than speculating. What we need is more tests.
Willie: Maybe it would help to think about how the time travel we’re considering differs from the backwards causation that you first proposed, Carlene; there’s a big difference. Yes, the trigger seems to have some kind of backwards causal connection with the existence of the anomalous particle at t=3. It would be amazing if we could confirm even that much. But if the anomalous particle is the psi-lepton time-traveling, then the so-called two particles that the detector is recording are really one. That’s crucial to the time-travel hypothesis; it’s what takes the mysterious origin of the anomalous particle out of play. It also has some more fantastic-sounding consequences; for example, it implies that the psi-lepton is at two different places at the same time.
Tad: You’re the philosopher; you figure it out. It doesn’t seem like such a big deal that there’s really only one particle; their sameness is easy to swallow. All I know is that there’s no intrinsic difference, but they’re elementary particles, so we should expect that. And yeah, they’re in different places at the same time. So what? It’s the problem of something in the future causing things in the past that bothers me. Whether they’re the same particle seems trivial by comparison.
Carlene: Let’s drop the metaphysics for a moment and concentrate on what we really know. The path at the bottom part of the printout of our second trial from yesterday, Trial 17, is the theoretically predicted path of the psi-lepton. At all the relevant points in the chamber, it behaves exactly as predicted by my calculations, until the unexplained event occurs.
Tad: You mean the collision?
Carlene: I’m not ready to decide whether it is a collision or whether it is a reversal in time, at least not yet. The point here is that, according to the data, we have a perfectly stable psi-lepton right up until the time of the trigger. The path at the top is where the confusion is coming from; all the data indicate it’s a psi-lepton.
Willie: Even a perfect match wouldn’t establish the time-travel hypothesis, but I’m not sure what would.
Carlene: That’s enough brainstorming for me; we’re not getting any closer to trying a new experiment. What we need are some ways to manipulate our experimental set-up that might give us some more useful data.
Willie gets up and pours himself another cup of coffee.
Willie: Any ideas, Tad?
Tad: Well, like I said before, what was really keeping me up last night was the hypothesis that this trigger thing is causing the anomalous particle’s presence. More than the very idea of the particle time-traveling, it seems weird that the cause of the particle’s appearance happens later than the appearance itself.
Willie: Right, that’s the big problem. You can bet that if we could find experimental evidence of backwards causation, it would send a shockwave through the philosophical world.
Tad: I’m sure. The big question is how to find the evidence.
Willie: Well, maybe there’s some way to bilk the experiment; maybe we can stop the trigger after the second particle appears.
Tad: That might be tough given there are only a couple of nanoseconds between the particle’s appearance and when the trigger kicks in.
Willie: But it might be feasible. It would take some tricky programming to optimize the processor’s resources, and even then it would be a close call, but I might be able to.
Tad: What would it show us if you could?
Willie: It seems that even if we had time to stop the trigger, we wouldn’t be able to since the effect of the trigger exists before the trigger itself happens, at least according to the time-travel hypothesis. If we do stop the trigger, we can rule out time travel and backwards causation as possible explanations for the second particle.
Tad: And if we don’t stop the trigger, at least we’ll have a lot to talk about. You’d probably argue that we would even have some experimental evidence that backwards causation is happening.
Carlene: I think that would be a great way to proceed. If we can stop the trigger from occurring after the second particle appears, we’ll have to consider that a strike against the time-travel hypothesis.
Willie: I’ll get started on the program. We’ll see how it goes, but I should be able to have it ready for trials today.
Carlene: Excellent. Let me know when you’re ready. I’ll be in my office.
Dr. Rufus leaves Tad and Willie in the lab.
Tad: It seems to me that we don’t even need today’s experiment to disprove the possibility of backwards causation.
Willie [typing at the console]: How so?
Tad: Well, we already know we can change whether the second particle appears and whether the trigger occurs by inserting the new code. There’s the practical matter of optimizing the use of the computer’s resources to allow us to prevent the trigger in the time between the appearance of the second particle and the annihilation of both particles, but that’s just a technological problem, and you’re probably about to overcome it. Still, even if you don’t succeed, we know it’s possible to insert the new code some time after the second particle appears. So, assuming that backwards causation is at work, we can prevent the cause of the second particle’s appearance after it’s already appeared. But we shouldn’t be able to do that, thus backwards causation isn’t at work.
Willie: So, correct me if I’m wrong, but your argument goes something like this: If the trigger at t=5 causes the second particle’s appearance at an earlier time, t=3, then it’s possible to prevent the second particle’s appearance at t=3 by switching in the new code at, say, t=4.
Tad: Right, that’s what I said. We should be able to prevent expected effects by preventing expected causes. On Trial 16, the new code was in at t=4 and the second particle never appeared at t=3. It appears that the new code being in prevented the trigger, which apparently prevented the appearance of the particle.
Willie: But there’s more: Setting the technological concerns aside, it’s surely possible to switch in the new code at t=4 after the particle appears at t=3.
Tad: Correct.
Willie: So, if the trigger at t=5 causes the second particle’s appearance at t=3, then it’s possible to prevent the second particle’s appearance at t=3 by using the new code at t=4 and it’s also possible to switch in the new code at t=4.
Tad: Still correct, and there’s the problem. In fact, it’s impossible to prevent the particle from appearing at t=3 after it’s already appeared at t=3. Preventing it would imply it didn’t happen, but it would have already happened.
Willie: And, since it’s impossible to prevent the appearance of the second particle after it’s appeared, it’s not the case that the trigger causes the second particle to appear.
Tad: Precisely. That’s why it is logical to assume that backwards causation can’t be the case and that the psi-lepton isn’t time-traveling. So, what’s the point of the experiment?
Willie: Well, the time-travel hypothesis might be false, but your argument doesn’t show that.
Tad: Why not?
Willie: Your reasoning is invalid. I’ll use another example to show you what I mean. Just think about this: It’s possible for my coffee cup to be full right now, but it’s also possible for it to be empty right now. Therefore it’s possible for it to be both full and empty right now? That doesn’t work.
Tad: Which is it?
Willie: What?
Tad: Your coffee cup.
Willie: Oh, truthfully? It’s empty.
Tad: So much for your aversion to caffeine from yesterday. I’ll get you some more.
Willie hands his cup to Tad, who goes across the lab to get some more coffee. Tad returns, handing Willie a full cup.
Willie: Thanks, Tad.
Tad: Okay, about your coffee example: I have to admit that it is impossible for the cup to be both full and not full right now; that really is illogical. But doesn’t saying that it’s possible that it’s full and not full right now mean the same as saying that it’s possible that it’s full right now and possible that it’s not full right now?
Willie: It may be tempting to interpret it that way; that’s why your argument may have seemed to refute the hypothesis of backwards causation. As you’ve just acknowledged, however, that’s a bad way to reason. It’s obviously possible for the cup to be full right now and possible for it to be empty right now, but it’s just as obvious that it’s impossible for the cup to be both full and empty right now. More generally, I’m saying that, from possibly P and possibly Q, it doesn’t logically follow that possibly both P and Q.
Tad: It just seems so much simpler regarding the coffee. How does all of this apply to the anomalous second particle?
Willie: It does so in just the same way. On the assumption that the trigger at t=5 caused the particle to appear at t=3, it was possible to prevent the particle’s appearance at t=3 by inserting the new code, and it’s also possible to insert the new code at t=4, but it doesn’t follow that it’s possible to prevent the particle’s appearance once the particle appears. Once it appears it can’t be prevented.
Tad: I see. So, you think we need to run the experiment?
Willie: Absolutely, we do. For starters, Dr. Rufus isn’t going to pass up the chance for more data, especially not due to a philosophical argument—yours or mine. More importantly, we really need to see whether the anomalous particle ever exists without the trigger.
The three complete their tasks individually. After lunch, they reconvene to begin a trial with Willie’s revision to the program.
Carlene: Gentlemen, we appear to be ready for today’s experiment. Willie, you’re sure you have the accelerator program set to remove the trigger as soon as it detects the presence of a second particle?
Willie: Yeah, I was able to optimize the use of the processor so that the removal of the trigger can be accomplished in less time than the interval between the appearance of the second particle and the annihilation event.
Carlene: Excellent. Shall we continue?
Tad: I’m starting the accelerator now. (See Figure 2.2)
Dr. Rufus takes the printout of the latest trial and examines it. A wave of puzzlement passes over her face.
Carlene: This is really puzzling. These data look something like the results when we removed the trigger altogether.
Willie: This is strange; the diagnostic log shows that the trigger was indeed removed upon detection of a particle in the chamber.
Tad: Isn’t that just what it was supposed to do?
Willie: Yeah, so why don’t we see evidence of the second particle?
Carlene: Hang on, Willie; you said your program removed the trigger as soon as it detected a particle, right?
Willie: That’s right.
Carlene: At what time did that happen?
Willie [checking the diagnostic log]: The program removed the trigger at t=0.1 plus or minus 0.05 nanoseconds.
Carlene: Hmm, just moments after the datum showing the creation of the psi-lepton.
Willie [looking embarrassed]: Um, I think I know what happened. When I rewrote the program, I set it to switch the trigger off when it detected any particle in the chamber, so when the psi-lepton showed up, the computer removed the trigger. We effectively repeated our Trial 16 from yesterday with no trigger at all. How stupid of me.
Carlene: That would certainly explain these results, but it doesn’t bring us much closer to establishing or falsifying our time-travel hypothesis.
Tad: Do you think you can modify the program so that it won’t remove the trigger until after the anomalous particle appears?
Willie: Yes, I can do it, but it’s going to take a little while; I don’t want to blow it again. I’ll need to make the program sensitive to the total energy in the chamber and set the threshold high enough to prevent the premature removal of the trigger.
Carlene: Here’s what we can do: one of our sponsors is hosting a conference tomorrow. Tad and I are planning to attend, so we won’t perform any trials. There’s also a conference luncheon. Willie, would you care to join us for that? It’ll give us a chance to touch base and be ready for the following day. I can arrange a place for you.
Willie: That sounds good to me.
Tad: It’s fine with me, too.
Carlene: Excellent. It’s been a long day already, so let’s shut everything down. We won’t be running any experiments until at least the day after tomorrow.
The three proceed to shut down the equipment in the lab. It does not take long for Tad to raise a question.
Tad: You know… I’m still having problems with the whole time-travel idea. Despite the evidence and the philosophy, I can’t help feeling that this whole thing is foolish. It just isn’t possible; you can’t change the past; you can’t time travel.
Carlene: I agree that we can’t change the past, but I fail to see how that makes time travel impossible.
Tad: It seems obvious to me. If you time-travel, you change the past; but since it’s not possible to change the past, you can’t time-travel. That seems like straightforward logic to me.
Carlene: And how could a time traveler change the past?
Tad: Well, lots of ways. In fact, I don’t see how one can possibly avoid changing history when traveling to the past. What if I really didn’t like my grandfather? I wish I could have killed him, but he died before I had the chance, so I decide to go back in time to kill him before he even met my grandmother. Now, I only get one shot at going back in time; I want to make sure I can kill him, so I take all the latest weaponry with me.
Willie: Uh oh, this is sounding familiar. Keep going, Tad.
Tad: Since I have the best available arsenal, along with prior knowledge of my grandfather’s whereabouts, it seems obvious that I can kill him.
Carlene: Well, maybe you slip on a banana peel, or something else random like that happens? It would be a little like there being the misstep in Willie’s program.
Willie: Thanks for the reminder.
Carlene: Sorry, Willie.
Tad: Look, it doesn’t even have to be anything as dramatic as killing my grandfather. I could step on a beetle or crush a blade of grass. Even my presence could slightly alter the flow of air. To me it seems impossible that I could time-travel to the past without altering the past. If nothing else, my tiny mass would slightly perturb aspects of the space-time continuum, resulting in some sort of change in the world. Since it’s impossible to change the past, it seems that I can’t possibly time-travel, and neither can our psi-lepton.
Willie: Tad, you’re saying that you could kill your grandfather because you have what it takes to do so, but that you can’t kill him because doing so would change the past. And you claim that this apparent contradiction rules out the possibility of time travel.
Tad: Yeah, absolutely.
Willie: Okay, but let’s take another example. Let’s say that I’m carrying a lot of groceries home, and I come up to the door, and I say, “Would you mind helping? I can’t open the door”. On the one hand, it seems that I said something true. On the other hand, it seems that I very well could have opened the door because I’ve done it hundreds of times.
Tad: How is that at all like my example?
Carlene: I think I see how they’re similar. Willie can’t get through the closed door while carrying a full load of groceries, but he can get through the door when he isn’t so burdened. He also could if he had longer arms, or if he had superpowers. In your example, Tad, we might have said it was possible for you to kill your grandfather if we hadn’t taken into account the fact that he died of some other cause at a later time, that he fathered your father, and all that. But relative to all those things we do take into account about the case, no, you can’t.
Tad: It seems to me that you’re using some weird definition of ‘can’. I just mean plain old ‘can’, as in ‘possible’.
Willie: Dr. Rufus is right. When you say, “X can’t happen”, you normally don’t mean that there’s absolutely no logically possible way that X happens; there are even possibilities where the laws of physics are different. If ‘can’t’ were to rule out all possibilities, we’d almost never use it. How ‘X can’t happen’ actually seems to work is to indicate that there’s no possible way that X occurs while other pertinent facts hold. What these facts are depends on the features of the conversation.
Tad: Okay, so what if ‘can’t’ means there’s no possibility of the event occurring while certain other facts hold? If I travel to the past, I still both can and can’t kill my grandfather.
Willie: I don’t think so. Either you can or you can’t, but which one it is depends on what facts are taken as given. You can kill your grandfather given just the fact that you’re well-armed, but not given the facts about how he actually died, say, of old age. You obviously can’t kill him, given the fact that you didn’t kill him. It doesn’t really matter that you—his grandson—are the one trying to kill him; the same applies to anyone trying to kill your grandfather at a time he wasn’t killed. Your scenario doesn’t give rise to a contradiction. It would be a problem only if you both can and can’t kill him given the same facts, but that’s not the case.
Tad: But when I step out of my time machine and bend that blade of grass, I’ve changed the past!
Carlene: I disagree. You caused the blade of grass to be bent, but that’s not changing the past unless that blade of grass wasn’t bent at that time in the past.
Willie: And it was bent if your story is going to be consistent. Don’t try to tell us that it was as straight as an arrow at that time the first time, but was bent the second time around. That really is a contradictory story, but there’s no reason to think that time travel is like that. If that blade of grass was bent at that time, then it was always bent at that time.
Tad: Well, if we can’t change the past, what were those cyborgs thinking in Terminator—the movie you think is consistent—when they sent a terminator back in time to kill Sarah Connor?
Willie: That’s one thing I didn’t like about the film. There’s no contradiction, but the cyborgs did reason poorly in thinking that they could prevent something they knew to have happened. But like I said yesterday, what do you expect from cyborgs? It’s unfortunate that this provides the basic premise for the rest of the story. It’s a rare time-travel film that acknowledges our inability to change the past, though there is the sci-fi gem, 12 Monkeys.
Tad: What happens?
Willie: A man, James Cole, travels back in time. He’s sent from 2035 to determine the origins of a deadly virus that plagued humanity in 1996, which drove the survivors underground. His goal is to take a sample of the pure virus to the future, to study and hopefully to discover a vaccine.
Tad: But if the vaccine is discovered, then the deaths of 1996 could be prevented after they occur, which leads to a contradiction.
Willie [interrupting]: Hold on, that’s not what he was up to. The scientists knew they couldn’t save the lives of those who were already dead; they wanted only to make a vaccine to prevent further deaths and to allow people to re-inhabit the surface of the Earth.
Tad: Okay, that makes enough sense, but it’s starting to sound like I would have to step on that blade of grass as I step out of the time machine, which is too crazy even to consider.
Willie: Why is it so crazy?
Tad: Presumably there’s nothing to grab my foot and push it down on that blade of grass; nothing would be forcing me to step on it. How can it be true that I would have to step on that blade of grass? If it had to be bent, it’s not even clear how I could have caused it to be bent. It sounds like it would have to be bent no matter what I did.
Willie: The language is tricky here. When we say, “It would have to be bent” or “You would have to step on it”, we’re taking for granted that you did step on it. We’re saying something pretty trivial, actually: given that you stepped on it, you have to step on it. We may as well say that, given the grass gets bent, the grass must get bent. These are really trivial claims.
Tad: But I wouldn’t have to! No one would be forcing me!
Willie: Forget about the time travel for a second. Given that you will wash your coffee cup later, you have to wash it. There’s nothing mysterious about that. It doesn’t imply the existence of manipulative forces; it’s just what you will do. And it’s the same in the time-travel case: given that you stepped on that grass, you will have to step on it.
Tad: But I could decide to hop out of the time machine, rather than step down, and miss that blade of grass completely.
Willie: Well, maybe you could.
Carlene: Wait a moment, Willie. Now it sounds like you are contradicting yourself. Which is it? Could he have hopped over the blade of grass or not?
Willie: Like I said, it’s tricky. Don’t forget about my trouble carrying the groceries. Could I have opened that door or not? It’s a simple question to answer when it’s clear what’s being taken for granted. I surely said something true when I said, “I can’t open the door”, but there are other contexts where it’s not taken for granted that I’m carrying a heavy, awkward load of groceries. Relative to these contexts, it’s true to say, “Willie can open the door”. After all, I would’ve needed nothing more than slightly longer arms, a tad—pardon the pun—more upper-body strength, or to be able to set down the groceries. Using ‘could’ in place of ‘can’ politely suggests that a change of the context is needed, that we should bring into play some more remote possibilities.
Tad: I’ll just politely ignore your pun.
Willie: Look, Carlene, you asked whether Tad could have hopped out of the machine. Sure, he could have, but can he? This is a hard question to answer when it’s not clear what’s being taken for granted. If we suppose that Tad stepped on that grass however many years ago, that he stepped down from and didn’t hop out of the time machine, then obviously he must step down. When he’s faced with exiting the time machine, he can’t hop out. If we don’t suppose that Tad stepped onto the grass, however, if all we suppose is that Tad has the normal allotment of human capabilities, that there’s nothing coercing him to do anything, and we also have no idea how things went, then he can exit the time machine in all kinds of ways. He may even decide the past looks scary and never leave the time machine.
Tad: Now it’s starting to sound like there’s no fact of the matter about what I’m able to do.
Willie: That isn’t what I’m saying at all.
Tad [interrupting]: Hold it, Willie, don’t say a thing. I’m too tired to keep this up. You haven’t convinced me, but my head is starting to hurt. Professor, are we done here? If so, I’m heading home.
Carlene: That sounds like a fine idea. I’ll see you two at lunch tomorrow.
1 To see an animation of any of the Tuesday illustrations online visit www.openbookpublishers.com/isbn/9781783740376#resources.