The rules of time travel?

In a somewhat whimsical podcast episode, Sean Carroll explores the physics and “rules” of time travel. Probably the first two thirds explore the physics. Carroll notes that if time travel under general relativity is at all possible, it would more likely involve a spaceship attempting to navigate some kind of closed timelike curve than stepping into a teleporter, phone booth, or something else along those lines.

The discussion is an interesting one, but it can be summed up that general relativity seems to dangle the possibility of time travel in front of us, but there always seems to be something that gets in the way. Carroll’s final summation is that it doesn’t appear possible, at least as things stand with our current understanding of physics.

He doesn’t really look at FTL (faster than light) proposals, such as the Alcubierre warp drive, although he does note that traveling faster than the speed of light is always moving backward in time in someone’s frame of reference. As I noted in a post a while back, having FTL without time travel requires adding special assumptions to remove the possibility.

This seems to accord with what I consider the Fermi paradox of time travel: if it’s possible, where are all the tourists from the future? That said, Carroll does point out that this observation only pertains to travel to arbitrary points in the past. If wormholes, or something similar, can be used to travel to the past, then we shouldn’t expect to see future visitors until the first such wormhole is built.

Anyway, the last 50 minutes or so of the discussion focus on what the rules of time travel might be. Carroll notes that he was a scientific advisor on Avengers: Endgame, a movie I criticized for its handling of time travel. But Carroll argues that he thought it got it right. I went back and watched some of the key moments in the movie, and can see a narrative for how it might be consistent. Interesting. Carroll notes that the role of the science advisor on movies and TV shows is basically to find a plausible scientific account for what happens in the story, not to sit around and tell the scriptwriters what they can’t do.

Anyway, time travel in most science fiction seems to take place within three broad paradigms.

  1. Whatever happened, happened
  2. The past can be changed, with consequences
  3. Changing the past creates new timelines

Under option 1, you can’t really change the past. If you try to do so, something always intervenes to stop you, or it inevitably works out that your actions were always part of what happened. Carroll’s example (with appropriate ethical caveats) is trying to back in time and kill baby-Hitler. Under this paradigm, anything you might try do only becomes part of Hitler’s biography. Maybe you end up traumatizing baby-Hitler in a way that sets him down his destructive path.

This paradigm shows up a lot in fiction. Notable examples are the first Terminator movie, or the original 12 Monkeys. It also tends to show up in fiction which has time travel in it, but the time travel dynamics, in and of themselves, aren’t the focus. So Doctor Who usually operates under this paradigm. (Of course, like all long running franchises, it isn’t consistent, flouting the rule whenever convenient to the plot.)

The biggest issue with this option is that it becomes hard to maintain if someone really wants to make a change. Suppose we form a league determined to kill baby-Hitler. Undeterred by initial failures, we keep sending expeditions to the past. All these expeditions must fail, in perhaps increasingly improbable ways. And they all must merge together in the historical timeline, all while not being noticed by any historians.

But whenever sci-fi does focus on the time travel element, it’s usually under the second option: the past can be changed, also changing the future. But this brings up the infamous grandfather paradox. If I go back in time and kill my grandfather before he met my grandmother, then I was never born. But if so, then who went back in time and killed my grandfather? Using the baby-Hitler example, I go back in time and kill baby-Hitler, but then I grow up never having known anything about Hitler, so what motivates me to go back in time and kill him?

Most movies and shows ignore the grandfather paradox and simply allow changes to exist despite the consequences. All the Terminator sequels fall into this category. But now we appear to have a situation where effects (the changes to the past) can continue to exist orphaned from their cause (someone deciding to go back in time to change something). This appears to break causality, violating a central understanding of how reality is supposed to work.

A way to perhaps avoid that is the third option: any change to the past creates a new timeline. If I go back in time and kill baby-Hitler, I change history, but only in a new timeline. The old timeline still exists. So killing Hitler doesn’t really save anyone, since World War II still happens in the other timeline. Causal orphans still appear to exist in any one timeline, but across all timelines causality is preserved.

In most fiction, it’s not possible to travel between timelines. This has the interesting effect of making Option 3 look like Option 2 for any traveler. It raises the question of how anyone could even know they exist in an Option 3 universe. At least unless the time traveling technology also allows travel between timelines. Although we might imagine a traveler going back to before a divergence and traveling forward again, but then, which timeline do they end up traveling on?

Option 3 also bears some resemblance to the many-worlds interpretation (MWI) of quantum mechanics, but the resemblance might be somewhat superficial. Certainly if the MWI is true, events diverge in other worlds. However, under the MWI, any travel to the past might pretty much make it impossible to return to your original world, even if you were careful not to change anything, since traveling forward again, the nature of quantum interactions means there’s no way to guarantee you would go down the exact same branches. Some minority portion of the different versions of you might make it back to a world that at least resembled your original, but most would be forever exiled from the world you left.

The multiple timelines thing does offer another possible explanation for the Fermi paradox of time. Maybe we’re just in the timeline that hasn’t yet been polluted with time travel. Although the probability of us just happening to be in that particular timeline seems infinitesimally remote. Still, we might imagine there is at least one timeline like that, and the people in it would consider it extremely unlikely they’re in it.

That said, a far simpler, and therefore more probable explanation remains: time travel isn’t possible. Still, it’s fun to think through what it would mean if it were.

What do you think? Is there an explanation for the absence of tourists from the future that I overlooked?

31 thoughts on “The rules of time travel?

    1. Eric,
      That was my meaning, although I took it to be implicit in what I wrote. Certainly we travel to the future all the time. In fact, we appear to have little choice in the matter. Although we could accelerate that travel if we could find enough energy to experience time dilation.

      Liked by 2 people

      1. Okay Mike, I just think it’s notable enough to explicitly state that in a natural sense fully half of all time travel scenarios are theoretically possible, or the forward kind, whereas the backward kind is utterly ridiculous. It’s like what Anonymole implies below. Each electron and all else for all the cosmos would need to reverse back to the their exact state one second ago in order for it to exist again that way and thus be experienced again by us. I think it’s safe to say that such speculation is well beyond causal dynamics of this world. For a physical determinist like myself, here things are coherent — it’s a single world (contra Sean Carroll) that moves forward by means of causal dynamics. Thus both the past and future would be fixed in an ontological sense, though freewill would exist epistemically given our vast ignorance.

        Liked by 2 people

  1. Option 1 – “Whatever happened, happened” – has a lot more flexibility than it first seems. I like the movie Deja Vu starring Denzel Washington on this score. After receiving clues that are vital to solve the case he’s working on, Denzel goes back in time and then realizes he needs to leave those clues for himself. Situations that look hopeless turn out to look a little better. Whatever happened happened, but whatever you thought you observed, didn’t necessarily happen exactly as you read it.

    This has the interesting effect of making Option 3 look like Option 2 for any traveler. It raises the question of how anyone could even know they exist in an Option 3 universe.

    Because … logic? Less snarkily, Option 2 requires a new definition of “time” that makes it not be a coordinate, which may be a sufficient reason to coin a new word. Then we have to reinterpret everything we assigned to times, in terms of this new concept. I have no idea how to do that, much less how it could be more elegant a description/explanation than “multiple timelines”. If I were a character in the relevant sci-fi stories, “multiple timelines” would be the mildest conceptual shift I’d need to make sense of my observations.

    Liked by 3 people

    1. Definitely option 1 might allow a time traveler to examine an event in far more detail then they ever could with only one pass through it. Deja Vu, if I recall correctly, involves investigation of a terrorist event, but that capability might enable historians to know far more about historical events in general than we do right now.

      I agree that if we found ourselves in Option 2, Option 3 would be a strong theoretical possibility. But finding conclusive evidence for it might be impossible. Again, unless we have some technology for traveling, or at least detecting the other timelines.


      1. Let’s go back to logic for a second. “The past can be changed” – what would that even mean? Take an example: Once yesterday was cold in Denver, but now yesterday was hot in Denver. Wait, what? Do you mean part of yesterday was cold, and another part hot? No, I mean once, yesterday at noon in the geographic center of Denver it was 45 F, but now, yesterday at noon it was 85 F. Oh, you mean previously, you were using Spock’s inertial reference frame to identify “yesterday noon”, but now you are using Kirk’s reference frame? No, I’m doing all this from Denver’s inertial reference frame. Oh I get it, you mean within your memory, there is an impression of being in Denver hearing yesterday’s date on the radio and feeling cold, followed by an impression of getting in the Tardis, followed by an impression of seeding some clouds to affect the weather, followed by an impression of being in Denver and feeling hot… No! I mean it was really hot, and also cold, at the same exact place and time, at different times! Same time, different times – I’m confused!


        1. The weather example had me confused for a bit, or maybe I missed your point. If not, another way to say it is once the archduke of Austria-Hungary was assassinated on July 28, 1914, but now his assassination was avoided on July 28, 1914. Yeah, it is weird to talk that way, and only people from the original timeline would have any idea of what it meant.


          1. I’m not clear how that gets you out of saying that there were two July 28 1914’s. Which means two (at least partial) timelines. Which means Option 2 has disappeared/morphed into Option 3.


          2. I understand the logic you’re following, and it makes sense to me. But then I also understand the logic for the MWI. But in both cases, we’re dealing with logical deductions, based on how we expect physics to work, not empirically tested facts. I think the debate in the interpretations of quantum mechanics shows that people disagree about which physical principles are indispensable.

            Liked by 1 person

          3. Indeed, which physical principles are indispensable. I view this as a crucial insight, and it’s why I didn’t want to just say “Logic! So there!” But indispensability is a continuum, and simplicity is also a player when we are deciding how to describe something.

            If we are going to say both “the Archduke was assassinated at [space, time point]” and “the Archduke was not assassinated at [space, time point]” then either we view these statements as incomplete, or we have just contradicted ourselves. It seems you are suggesting that to keep Option 2 alive, the incomplete specification is [space, time]. So: perhaps he was assassinated at [space, time, metatime1] and not at [space, time, metatime2].

            The problem is coming up with a way to do that which is NOT elegantly summarized as: there are multiple time lines. I’m not claiming to have proven you cannot come up with such a way – it just seems like a really tall order.

            Liked by 1 person

  2. Few if any time travel scenarios / memes take into account the fact that not only would we be traveling in time, but in space. Just in the last second [……….] you and I have traveled millions of miles—relative to some fixed spot in the universe—from where we were back when that second started. The Galaxy Song from Monty Python’s The Meaning of Life is apropos here. In order to go back in time, we would also have to resituate *ALL* of the cosmos’ particles back to where they were—way back one second ago. That possibility, I suspect is the truly impossible part; entropy flows one way.

    Additionally, we *can* time travel, but only in the forward direction. My mind travels forward 8 hours every time I wake up in the morning. BAM! I’ve jumped forward in time. Imagine pulling a Ripvanwinkle for a year or a thousand? One of my favorite sci-fi themes is the statis-bubble: freeze yourself in perfect, rigid, indestructible stasis for X amount of time, and wake up when the timer dings done. Given the opportunity, I’d entomb myself in one today. Of course, my enemies need only toss me into the sun and wait for me to wake up… whenever.

    Time travel is such an anthropocentric concept: I don’t like the way things are so I’m going to go back in time and change them (for the better), yeah, good luck.

    Last thought in association with time travel: the ripple effect. I’ve often wondered about the smallest, the shortest, most inconsequential event in history and had it not happened, how different the world would have been. For instance, had grain-sized meteor A not hit granule-sized meteor B, which went on to bump thimble, golf ball, softball, basketball, pool table, swimming pool, city block-sized asteroids C,D,E,F,G,H,I… Then the last would have never hit planet killing KT asteroid J causing it to miss the Earth 66 million years ago. The smallest, most minuscule motion or delay would have changed the universe. His sperm was delayed by a muscle spasm and never reached her waiting egg thereby avoiding the birth of Hitler.

    Humans have such capacity for pontification…

    Liked by 4 people

    1. Good point on space travel. I always wondered how Wells’ time traveler didn’t find himself in deep space. It’s possible to concoct an explanation on why he didn’t, but I doubt Wells actually gave it any thought. And if time is emergent from entropy increase, it does make the idea of just navigating along that dimension seem pretty hopeless.

      I don’t know if I would entomb myself in stasis, at least not for an extended period of time. We don’t know what the future might hold, so it could be suicide. But I might do it in such a way that I wake up regularly to ascertain the situation and decide whether I want to enter the world or go back into stasis.

      On the ripple effect, definitely. It seems like any travel to the past, no matter how inconsequential, could cascade over time into changes. Although it’s also possible to have a mix of us always having been to the time we traveled to and therefore just part of the existing mosaic, unless we go out of our way to change things. (I think that’s the way to make sense of Avengers: Endgame. Maybe.)

      Liked by 1 person

    2. “For want of a nail, the kingdom was lost.” Ever see the movie, Sliding Doors? It explores the idea of those moments that matter.

      The problem of location bothered me about time travel since I was a kid. (I was into astronomy and knew about the Earth’s motion.) I always thought ghosts had a similar problem — they can go through walls, but don’t sink through the floor. Why not? What tethers them to the Earth at all?

      Liked by 3 people

      1. Ghosts! Great point. Either they’re affected by gravity, or their spiritual tendrils keep them locally bound. Which would make more sense as it’s not people who are haunted, but houses, insane asylums and abandoned school basements.

        Liked by 1 person

        1. That’s the thing, one can justify ghosts more easily than time travel. They’re bound to places, plus ghosts are intentional — they will themselves to stay level on floors (there was a really bad movie starring someone who turned out to be a bad person, but it’s the only story I know of where a ghost had trouble staying on the floor — “Pudding Pops” kept sinking through it until he learned to will himself).

          OTOH, ghosts suffer from the “how does anything invisible see or hear?” problem. So do shape-shifters. I always wondered how Odo heard or saw anything when he was being a glass on a tray. (I always wondered what happened to his weight, too. SF loves shape-shifting, too, but it’s usually some pretty serious bunkum.)

          Liked by 1 person

        2. In Oxford, the ghost of Francis Windebank is allegedly only seen from the knees up, due to the change in ground level since his execution in the time of the English Civil War. And on the subject of location, there exists a theory, if you can call it that, called Stone Tape Theory, which explains ghosts as imprints left on physical objects, hence explaining why they are located in specific places.

          Liked by 3 people

          1. Spirits being attached to things… We’ll have to assume, then that there must be some unknown physical aspect to them, else they’d have been flung into the cosmos.
            (That’s always been my personal giggle point regarding anti-gravity. Disable gravity for yourself and you’d get spun off into space in a heartbeat. Or slammed into the ceiling, if indoors.)

            Liked by 1 person

  3. Absolutely. You don’t even need grain-sized meteor A. Just put a minor gravitational difference into the solar system a few million years before the KT asteroid – say, the mass of one human time-traveler who appears on Earth, hangs around for an hour, then goes away. That ought to do it. Chaos rules. It’s not like the KT asteroid had some homeostasis that would bring it back to dinosaur-killing position if disturbed.

    Liked by 1 person

    1. I’d always hoped that some studio would create a movie along this theme. Half a second’s hesitation at a traffic signal, saves or kills some kid riding a bike. Or a sneeze on a metro train forces someone to lookup and not see the pickpocket. So many scenarios to explore, infinite, I suppose.

      Liked by 1 person

      1. There’s an old Ray Bradbury short story: A Sound of Thunder, about how the death of a butterfly millions of years ago led to dramatic changes in the future. I think there was a film adaptation. But definitely it doesn’t occur as much as it could in mind expanding sci-fi.

        Liked by 2 people

  4. I assume you’re talking about the rules of time travel in fiction. In that case, one way to have option 2 (the past can be changed, with consequences) and avoid paradoxes and causality-breaking is to invoke the idea of some kind of personal time that the character follows, which is not the same as universal time.

    Suppose I decide to go back in time and kill my younger self. At the beginning of the story, we can agree that this did not happen, because I am alive. But now I jump back in time one year, kill my younger self, and return to the starting point. Now I am dead. I died one year ago. That’s what history records. And yet I am alive.

    The paradox is explained by following my own personal timeline, which makes perfect sense to me – I travelled back in time, killed someone, then returned to the present. The past I remember is the past I experienced, but it is different to the past that history records. The past has changed. It “was” A, “now” it “was” B.

    Liked by 1 person

    1. I’d say fiction, or very speculative physics. Take your pick.

      I can see what you mean with the personal identity timeline. For you personally, there would be no paradox. But the paradox would arise for the rest of the universe, where an extra you would appear out of nowhere and kill the you of a year ago, then disappear for a year, then reappear, talking about another reality where you were a continuation of the killed you.

      The question is whether that other reality at that point exists anywhere other than in your head. If not, then we’re in Option 2, and it seems like current you has succeeded in erasing a part of reality. But that still seems problematic to me. All the causal effects of that other reality, except you, disappear? It feels like there are some conservation issues here. Although depending on how we conceive of the other timelines(s), there may be conservations issues there as well.

      Liked by 1 person

        1. That actually may be the biggest conceptual problem with the idea of time travel. If we imagine it becoming a commonplace technology, then every traveler ends up existing in their own personal history, with only brief ephemeral intersections between other travelers or people. I think that’s true regardless if it ends up being Option 2 or Option 3.

          If both you and I are history altering time travelers, and we meet for lunch, we might have contradictory memories of what reality has been when we exchange stories, many of which would only be reconcilable through our personal histories. The next time we meet, we might be literally different people.

          Liked by 1 person

  5. I remember reading the two novels Blackout and All Clear by Connie Willis. Historians are sent back to WWII from the Oxford University History Dep’t in 2060, but the rules of time travel were interesting to me. They were close to Option 1 (whatever happens happens), if I recall, but time was like an organism in a sense. It would prevent historians from visiting points in time that would be problematic, and so the historians thought it was impossible, due to time’s own intercessions in their traveling, to actually change anything. But I think–my memory is very fuzzy as it’s been a while since I read these–they start to get the sense some change is possible. At any rate, several characters get trapped in WWII…

    The interesting thing about Option 1 is: how do you define “what happens?” In the big picture, Hitler becomes the leader of Germany and embarks on all the horrible things he embarked on. But there’s so many levels things could be described so there is this weird coarse-graining in this concept: like smaller or lower level events don’t matter so long as some over-arching narrative is preserved. It makes for interesting opportunities for the writers I think, in terms of where to draw that line.


    Liked by 1 person

    1. I haven’t read those books, but they won the Nebula Award for Best Novel. And by all accounts, Willis is an excellent author. It sounds like she’s tuning into something like Stephen Hawking’s Chronology Protection Conjecture, the idea that that the laws of physics conspire to prevent changes to the timeline. Although Hawking most often presents it as physics making sure time travel is impossible.

      Alastair Reynolds had an interesting take on that in one of his novels. Basically if you create a wormhole to a location in the universe, you can no longer interact with that location except through the wormhole. It essentially prevents causality from being violated.

      On “what happens”, I do see something like that in some fiction, where minor changes can happen, but things always seem to converge back to the overall narrative. In one story, I think it depended on how well known the event was. Widely known events had a causal weight of some kind that made them very difficult to change, but the fate of, say, one non-famous individual during WWII was a different matter.

      Of course, those “fixed points” themselves have a causal light cone. They’re dependent on all kinds of micro-events in the deep past. To anonymole’s point, the actions of Hitler depend on all the actions of individual nobody’s deep into human history. The Star Trek episode The City on the Edge of Forever explored the idea that saving an unknown woman cascaded into changing world history in a way that erased the Federation.

      Liked by 1 person

  6. I did start listening to Carroll’s podcast, but first let me say that I consider the man to be so smooth, so talented, and ultimately so manipulative, that what he says needs to be considered with extra caution. Here’s an issue that I noticed.

    He says that special relativity doesn’t let us imagine past time travel, but general relativity does given greater potential for extremely high speeds. (I suppose he meant “is conceptually possible under general relativity”, since we can obviously imagine things that are both real as well as impossible either way.) At about 1:06:00 he provides a scenario under general relativity that in 100 billion years everything becomes the same as it is now, and so a closed loop exists which naturalism mandates must repeat every 100 billion years.

    I’m not sure why he thinks mere special relativity would preclude that scenario since going faster than light isn’t mentioned. And are we to believe that the Big Bang reverses in this period through gravity or something to get everything in the universe then back to where it is today? Though incredibly improbable, in a conceptual sense I do think I grasp the idea of everything becoming the same as it was 100 billion years previously.

    Then he compresses this timeline down to 10 years and asks if you could thus meet yourself from 10 years ago. Um… if all of reality changes to become exactly the same as it was 10 years ago, then you wouldn’t meet your former self, that is unless you did then, which of course you didn’t. He rolls with this however, I suppose to imply that faster than light travel could be possible under general relativity, not that anything about going extremely fast is actually mentioned in the scenario itself.

    Note that earlier he stated that he didn’t believe that time travel to the past was possible. I found that a bit like credibility bait before this unfounded looping business. One thing than “many worlds” and “eternalism” have going for them, is that they’re great for sci-fi and Hollywood scripts!


    1. Eric, I’ve noticed you seem to take a similar attitude toward just about any public thinker you disagree with. It can’t be that people pay attention to them because of their ideas. It must be that they’re smooth operators, con men of some kind. You have similar takes for Dennett, Chalmers, Graziano, or just about any other pubic intellectual you don’t see eye to eye with. Has it ever occurred to you that you just disagree with them? And that a more productive approach might be to address those ideas? I don’t think the personal attacks persuade anyone who doesn’t already agree with you.

      Anyway, Carroll does explore a lot of hypothetical conceptual space in this podcast. He does conclude in the end that travel to the past probably isn’t possible. But rather than just state that, he eliminates each of the possibilities and takes you through the reasoning to that conclusion. It involves some explanatory stages, stages that resemble the ones I mentioned in the epiphany post, ones that are trivial to strawman.

      Liked by 2 people

      1. I did listen a bit further Mike, and noticed that my assessment here might be considered a bit strawmanish. For this scenario and some others he did present a, “Well, probably not….” assessment a bit later. And I guess he meant a fixed rather than expanding universe that changes back into itself for whatever the reason.

        I do consider Dennett inordinately smooth, though not Chalmers. He’s quite the opposite and I find often difficult to listen to. He’s a dualist, but an honest one. I forget about Graziano at the moment. Isn’t he the guy with the puppet? Anyway sorry for being a downer. I suppose I shouldn’t have sent that one, or at least not publicly.

        Liked by 1 person

Your thoughts?

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.