Comments on the Fray

First of all, I’d like to thank everybody for all of the interest in my article in yesterday’s slate. It was apparently the most viewed story of the day, and even briefly made it on to the front page of digg.

There were a lot of really interesting comments in the fray, as well as quite a few in emails sent directly to me, and I’ll try to respond to everyone directly. For the moment, though, I wanted to hit some of the big points that people brought up.

  1. Time travel’s not real.

    Of course it isn’t. I personally don’t think it likely that humans will ever travel through time. But the important point (to me, at least) is that based on what we know about how the universe works, it’s not actually impossible to have time machines or “closed timelike curves.” Rich Gott (one of my friends and inspirations) along with his former student, Li-Xin Li even showed how one could solve Einstein’s equations to make a closed timelike curve as a source for the big bang.So no, time travel isn’t real, but it could be. This is where science fiction (to me, at least) can be so much fun. Take something that you know to be true, and follow it to its furthest extremes. If what you have is still consistent with how the universe works, it’s interesting to think about it, even if realizing that picture is a very long way off.

    While I’ll deal with a couple of specifics later on, I’ll say this. We know that general relativity provides a very good model for gravity, space and time on scales larger than the Planck scale. We know that quantum mechanics passes every test at the subatomic scale (and predicts macroscopic behavior to boot). Ideally great science fiction should be consistent with our two great triumphs of 20th century physics.

    A couple of people brought up 19th century sci-fi (Wells, in particular) as being inconsistent with my rules. But I wouldn’t say that’s fair. I’d judge them by what was known at the time, not by what we know now.

  2. Primer

    Oh my goodness! How many of you brought up Primer? I can’t even count.For the record, I have seen it, but only after Dan Engber, the science editor of Slate, recommended it during the writing of my article. It’s a very interesting movie, but I didn’t discuss it because I wanted to focus (with a few well-known exceptions: Terminator, BTTF) on TV and movies over the last year or two, and Primer came out in 2004.

    The Primer time machine is ingenious, no doubt. I love the way it works: you get in at the end time, and exit at the beginning, and all the while, you’re inside. It guarantees my rule #2. During the first description of time travel, it even looked like they were going to follow the “single self-consistent history” model that I advocate. However, it became very clear early on that each time trip changes the timeline. That became the central focus of the movie.

    While it doesn’t follow my rules, it is one of the most meticulously thought out visions of time travel that I’ve ever seen.

  3. Parallel Universes and Many Worlds

    You folks really love your parallel universes, don’t you? Jeff and I take a much looser stance on them in our book, but clearly give preference to the “single universe model.” In the slate article, I wanted to give full expression to what I think are the most well-established laws in physics, and so I discount parallel universes outright. This made you angry (far angrier, oddly enough, than my stance on free will versus determinism). In particular, people object to the following:

    • The “double slit” (and other quantum mechanics) experiment can be explained by having parallel universes.True. But this is an interpretation of quantum mechanics. “Interpretation” in this case, means that the existence or non-existence of parallel universe isn’t directly testable, and is just another way of thinking about the underlying equations. It is an alternate to the standard “Copenhagen Interpretation.” This was the nature of my comment that the “Many Worlds Interpretation” couldn’t be either proven or disproven observationally.However, and this is the important point, Everett’s picture doesn’t give us any method of visiting these parallel universes (even if they exist). As far as any physical experiment ever conducted is concerned, parallel universes need not exist. What’s more, quantum theory says nothing about the flow of time. It basically assumes that time is unidirectional.

      On the other hand, general relativity does provide for self-consistent models of time travel. Gott and Thorne have each come up with them. Yes, quantum mechanical arguments come into play (especially in Thorne’s model), but those QM issues are all local. They are primarily concerned with keeping the wormholes open, not with the curvature of space itself.

    • What about brane theory, M theory, string theory? Don’t they predict multiverses?I’m not going to badmouth string theory. The jury is still out and it really could go either way. However, one of the constraints on a successful theory is that they must satisfy general relativity on large scales. As you may know, one of the cool ideas that has been floated (no pun intended) by people like Paul Steinhardt is our universe is a 3-dimensional brane (a structure allowed by M theory) floating around in a higher dimensional “bulk.” This means (among other things) that there are other “universes.”Even without string theory, the standard inflationary model predicts that we can get many (infinitely many, perhaps) universes in the “multiverse.” Aren’t these (you might ask) parallel universes that could be visited via time travel? No. First, so far as we know, we can’t actually visit them, and second, these other universes aren’t the parallel universes predicted by Everett. They are likely to have nothing at all in common with our own, let alone an almost identical version of you who just happens to have a goatee.
    • How dare I badmouth BTTF?

      You’re right. It’s a great movie. So are a lot of others with a BTTF time travel model. That said, there is something uniquely satisfying in seeing a time narrative click into place. That’s why I was playing up TTTW (and other movies that satisfy my rules).

  4. Conservation of Matter

    This seemed to be the most popular topic in the fray, and it’s a toughie. But I have a couple of points to make about it.

    • Some of you were concerned about having duplicates of your atoms existing simultaneously in the same point in time. However, all electrons in the universe are identical to one another. All of the protons are identical to one another, and so on. The point is that there’s no problem having two of you existing simultaneously because there’s nothing special about “your” atoms.
    • Conservation of mass isn’t necessary, but from everything we’ve seen about physics, conservation of energy is, and energy and mass can be converted into one another. The trick is that it’s hard to describe the energy “in the entire universe.” Instead, all we can say is that locally, the energy current is conserved. In every model I’ve seen, the energy continuity equations still seem to hold.
    • Energy conservation in time machines reminds me a bit of Hawking radiation. In Hawking radiation, vacuum fluctuations outside of a black hole produce particles and anti-particles ultimately resulting in one of those particles escaping (and subsequently being observed as radiation), while the other falls into the black hole. Where did the energy of radiation (ultimately) come from? Immediately, it seems like it was “borrowed” from the universe. However, you can’t borrow energy (using uncertainty) for very long. Ultimately, that debt is repaid by taking mass from the black hole, which slowly evaporates.Similarly, I could imagine (though this is admittedly quite speculative) that some of the mass a wormhole time machine (say) is evaporated in the process of a traversal.
  5. Time travel=Space travel

    Finally it was noted by a few of you that travel through time is useless unless you travel through space as well. Since the earth (and sun and Milky Way) are moving through space at ~600 km/s, you will be nowhere near where you started when you land. True enough. But since we can travel through space (and since, according to special relativity, we don’t really get too worked up about things moving at constant speed anyway), it doesn’t seem unreasonable that a time machine could be built to take that motion through space into account as well. The time part is the toughie (practically speaking).

Thanks again for all of your great comments!


This entry was posted in Uncategorized and tagged , , , . Bookmark the permalink.

18 Responses to Comments on the Fray

  1. Paul says:


    Thanks for a great article. I’m surprised you didn’t mention “Twelve Monkeys,” the movie starring Bruce Willis and Brad Pitt which I feel portrayed time travel perfectly. “Termintator” was flawed in the sense Skynet actually acted as if changing the past was possible whereas in “Twelve Monkey,” the scientists who invent time travel know the past cannot be changed.

    I don’t know if you remember the tagline, “The future is history.” Most people thought it referred to the plot about the release of a devastating genetically engineered virus that destroyed humanity’s “future,” but that tagline was really about the immutability of both the past and the future because with time travel, there is no difference between the future, past, or present.

    The movie does a great job stringing you along in several misleading directions, one of which is the false hope that past events can be changed, i.e. the virus won’t be released and billions of deaths can be averted. Bruce Willis’ character acts out of this conviction, misunderstanding his mission and ultimately believing he can completely escape fate. But it turns out the very act of doing so simply reveals events as they originally occurred. “Twelve Monkeys” shows that in a universe with time travel, free will is an illusion and destiny is supreme because the future will *always* be history.

  2. K-sky says:

    And fans of Twelve Monkeys who haven’t watched its source film, Chris Marker’s short La Jetée, must.

  3. mandi d. says:

    Dr. Gravity, I salute all the good works you have done. By the by, did you ever come up with a proper equation to explain the amazing gravity which sucked my yellow spatula for Patty into the black hole of the Goldberg homestead? I would be interested to evaluate your findings.

  4. John Gregg says:

    I found the original article facile. The anti-parallel universe stance is not a scientific position, since as you say, it violates no scientific principle. It is untestable, but not inconsistent with established facts or generally accepted theories. It is thus perfectly fair game to speculate about parallel universes in what is by definition speculative fiction. Your aversion to parallel universes is not a scientific opinion, just an opinion – sort of a conservative sqeamishness on your part, with no more authority than my taste, or my mother-in-law’s taste. It is a purely aesthetic matter, really.

    In a single block universe, you just can’t travel backwards in time. How would the physics work out, exactly? Obviously BTTF was silly (I liked the movie a lot, BTW), with the photo actually fading out from top to bottom. Try explaining that in terms of reflectivity of the photographic paper, etc. But all block universe interpretations are prey to the same problems. If I had an elite, highly trained commando army and I sent them back one by one to kill Hitler, by what physical process would they all, each one, be stopped? I was thoroughly unconvinced by the hand-waving about billiard balls. What if the army didn’t have to kill Hitler or anything so dramatic, but only had to make some change that I could detect from the future? Recall that (barring black holes) information can never be destroyed, it is conceivable that one day we might get very, very good at detecting the vapor trails of the distant past, so even one of my time travelling commandos popping into existence in a deserted forest, then returning to the future might be detectable. What if I had a video camera trained, say, on an ATM in a bank foyer running all night, and a week later decided to send one of my commando guys to that same foyer to smile and wave at the camera? How exactly would the physics work out so that he would be prevented?

    Parallel universes are the only coherent way to talk about time travel. No fading photos, no worrying about the combinatorics of human conception (zillions of sperm cells – your particular conception is so astronomically improbable that it is far from enough to make your parents meet – you better not change anything about either of their lives in any way). As soon as you wink into existence in Berlin in 1933, the universe forks. On one fork, you didn’t appear, and that fork evolves into the universe that produced time-travelling you. The other fork is a brand new universe with you in it from the future. You get to kill Hitler, prevent your parents from marrying, blow up the world, whatever. I think its actually pretty intriguing. You can never get back to your own world, and it blurs the distinction between reality and something like Star Trek’s holodeck. What moral obligations do you have in a world that is essentially a playground you created?

    I’ll have to check out Primer.

  5. dleone says:

    I’m surprised — well, no, not really — no one has mentined Bill and Ted’s Excellent Adventure. Its entire postion seems to be that time travelling is absurd. The duo make virtual magic happen without any effort, reminding themsleves to come back in time later on and set it all up. In fact, their future selves rescue them from failure several times without their knowledge. I rather feel as if the film’s makers were poking fun at the entire concept.

    But I was really thinking of the curcuits of time which appear to incorporate physical distance as well. The circuits would have to have some elasticity to deal with the earth moving around the sun and the solar system and galaxy’s motion themselves.

  6. Doug says:

    When I consume science fiction, I usually allow for the term fiction to be applied to the science aspect of the novel. If a sci-fi story is not going to use the science that we know to be true, I like to see two things:

    1) There is some explanation of how the made-up science functions in the story. This doesn’t have to be in depth, or even that logically convincing for me to enjoy it. Sometimes it’s absent, but I appreciate it when it’s there.

    2) Whatever system is created is used consistently throughout the story. The science doesn’t have to work or be compatible with science as we know it, but it should work consistently throughout the story. It’s not cool if in the first chapter time travel works differently from in the last chapter.

    I think at some point this approach becomes necessary for most epic sci-fi stories to function. If we only allow for what is actually possible given our current knowledge of science (even if it is extrapolated a bit), we wouldn’t have cool things like hyperspace travel, light sabers, the Force, or spice that enables the Kwisatz Haderach.

    There’ll always be a place for hard science fiction, but I think we should give creative types the leeway to create sci-fi where the very laws of science are free to be altered and manipulated for the sake of telling a good story.

  7. dave says:

    Doug – I agree. If the “fi” is good enough, the “sci” can get a pass, but think about movies like “the core,” or “Armageddon” in which the science is so awful that you can’t ignore it. That said, my main focus of the article was why I liked when movies got this particular aspect of the science (a single self-consistent history) correct. This brings me to:

    Gregg – I respectfully disagree. My like or dislike of parallel worlds isn’t just a preference, and I wouldn’t say it should simply be agreed to based on authority. The reason I don’t like it and I _do_ like general relativity as a means of time travel is that:

    A) GR has passed every observational test thrown at it, and

    B) It’s possible to create stable and practical time machines (at least in principle) using very simple GR arguments. To wit, the Gott time machine that I mention in my article, is (mathematically) simple enough that I could derive it for an undergraduate level GR class. I’ve excerpted the schematic from our book at:

    C) While it’s fun to talk about parallel universes (even for me) there’s no comparable argument that’s consistent with either relativity or quantum mechanics (even, it should be noted, by advocates of the many world interpretation) which could similarly tell us how to travel through time or visit a parallel universe.

    – Dave

  8. John Gregg says:

    I don’t understand the Gott machine schematic. I assume that it still falls prey to the block universe problem. What about my commando army killing Hitler one by one? What about the video camera in the bank foyer? All time machine scenarios that posit a single universe immediately run into silliness like Marty McFly’s photo fading out. Lots of hand-of-God intervention preventing any of the commandos from killing Hitler.

    As you say, it is fun to talk about parallel universes. That’s the point, isn’t it? Of science fiction, anyway? Why make a rule against a premise that you admit is fun, and crucially does not contradict any known scientific laws? All SF involves some smoke and mirrors and goofiness, but parallel universes is the least silly or obviously contradictory way of talking about time travel that I can see. You are right that there is no solid theory that says absolutely that we could fork off a parallel universe in the way that I suggested, but as long as it is not ruled out by scientific theory, let alone common sense, why can’t I write a cool show, book, or movie using it? Other than your personal preference, that is. The hand of God stopping the commandos from killing Hitler or from waving at the camera in the bank foyer, on the other hand, seem to me to violate common sense pretty outrageously.


  9. K-sky says:

    John Gregg! Dave conceded that parallel universes are fun. He’s not saying you can’t write a show or book about them. The problem with parallel universes is that, using what we know of science, there isn’t that much you can say about them. They’re like God, or love. They’re not falsifiable. Write whatever you want. No cabal of physicists will stop you from making the next Back to the Future.

    Anyone who has spent any time at all studying relativity knows that it is often hard to square with common sense. Why are you so surprised that an analysis of time travel which makes use of theoretical physics requires a degree of sense that is not “common”? That, to me, is the fun.

  10. Alex says:

    As a (former) physicist I read your article in Slate with interest. I think it did a great job of explaining certain physical concepts, but it would have been better if you hadn’t danced around the idea of causality. Once you assume causality, the conclusion is pretty clear: Only the boring kind of time travel, closed time-like curves, are possible, and they do the expected damage to free will during their lifetime.

    But causality is an assumption. So far it has worked for us pretty well, but on the other hand we’ve never explored the energy densities necessary for a GR wormhole. I think it’s reasonable to point out that we have a lot of evidence for causality, and that a lot of the fun time travel ideas violate it, but to (at least implicitly) claim that physics has demonstrated causality is a bit much for me.

    Finally, just a note on something else you said here: Energy conservation depends on time-invariance (or a salvage of that), and by creating a GR time-traveling wormhole you’re doing enough local damage to time-invariance that I would actually be surprised if there was any meaningful way to save energy conservation.

    But anyway, there are few enough physicists willing and able to write well for a general audience. (actually I could have left out the last clause.) Thanks for the article, and I look forward to your book.

  11. Alex says:

    Oh, one more thing to keep in mind: In string theory there’s some question about whether closed time-like curves can form. They might be prevented by the string or brane modes wrapping the putative curves. Take that information for what it’s worth…

  12. dave says:


    You make several fine points. A couple of quick observations.

    – I don’t so much assume causality as self-consistency. GR assumes both the former and the latter, while QM assumes only the latter.

    – Excellent point about energy conservation! Another way of thinking about it is: Noether’s theorem is based on the idea that invariances give rise to conserved quantities. Time invariance (in particular) is the generator of energy conservation.

    – I intentionally steered clear of string theory in the original article. There’s too much of “such and such MAY be allowed in string theory” (none of which, it should be said, suggest Everett’s many worlds can be visited). I will say that many people (including Hawking, as I recall) thought that closed time-like curves weren’t allowed in ordinary GR either, and that turned out to be incorrect.

    – Dave

  13. scout29c says:

    Your put off of the concept of conservation of energy and matter is a bit too flippant. Did this not point to the neutrino long before we found it? Is this not the only concept that illuminates dark matter? Is it not a basic concept in both the space-time relativity camp and those in the quantum group? It does not say we cannot travel in time. It just says that if we do, energy and matter will be conserved – and some would add momentum, but let’s keep it simple.

    One possible theory: atoms and/or energy are exchanged between two different times. Or maybe, you are right and the same mundane, identical atom can exist at two different places simultaneously, but the conservations equation is satisfied elsewhere.

    Another fundamental I would add to your list is the constraints we currently face in the three dimensions we do travel, not even considering time. Your left hand cannot be your right hand. If you are moving in one direction, you cannot be moving in the opposite direction. Moving up while the elevator is going down only means the rate of descent has decreased. (True, both of these should include that time bitch, simultaneous, but simple solutions to complicated concept should be kept…simple.) The point is we are constrained in how we travel in the three dimensions we have mastery. The same should be true of the time dimension. This could explain the “grandfather paradox” or that pool ball thingy.

    Another mind game you might play in trying to come to grips with time travel is to imagine how an individual from a two dimension world would experience our three dimension world. How would they comprehend that third dimension? Insights here could translate to insights about this fourth dimension in which we can only travel in one direction by experiencing it.

  14. John Gregg says:


    The second assertion in your comment isn’t true. You said that Dave never said I can’t write a show or book about parallel universes. But in fact, that’s pretty much exactly what he said. He said “…time-travel narratives ought to still abide by a few fundamental ground rules. Here’s my list of the most important principles of time travel, real or imagined.” Ground rules, principles – sounds to me like he is saying that as far as he is concerned, I can’t write about parallel universes. Of course I can write anything I want regardless of Dave’s wishes, but the core of my complaint is he basically said, “There has been an awful lot of silliness in time travel fiction over the years, and as a scientist, here is a list of some of the more outrageous silliness that we ought to rule out now and forever.” But his dislike of parallel universes, as far as I can tell, in no way derives from any scientific authority. It’s just a personal preference on his part, with no more weight than, as I said in my original post, my mother-in-law’s opinion on the same subject.

    As you point out, parallel universes are pretty open ended as far as the science is concerned. That’s why they make such good fiction, like love and God (about which there has been some quite good fiction written). This open endedness is a strike for, not against, parallel universes as a premise for speculative fiction. And when I speak of violating common sense, I’m not talking about doing so in the deeply mysterious way that, say, quantum mechanics or relativity violate common sense. I don’t like fiction to violate common sense in the silly sense of Marty McFly’s photo fading out the way it did, or just getting his parents together again guaranteed that he, himself, would be born. I find hand-of-God inhibitions against killing Hitler similarly silly. Parallel universes are not silly in this way. They do not violate any known scientific theories. They present all kinds of wonderful wrinkles, ripe for exploration in fiction. Not only do I not feel obligated to obey a “ground rule” against writing about them, I don’t see why anyone would write such a ground rule in the first place, or why anyone would obey it.


  15. Chris Richards says:

    I really enjoyed the article on Slate. I was also wondering if you’d bring up Donnie Darko and/or Primer, both of which were not in the article but are here. One other example that came to mind: Episode S04E01 of the show Futurama. The cast travels back in time and the main character:
    – meets his grandfather
    – accidentally kills him
    – consoles his grieving grandmother
    – figures he must have been mistaken since he hasn’t disappeared
    – and … becomes his own grandfather

    The episode does break the traveling without an exit rule, but does it also break the altering of history rule? Naturally, the show was written to be funny, not accurate, but still an interesting concept.

  16. k-sky says:

    scout29c: imagine how an individual from a two dimension world would experience our three dimension world.

    I’ve always understood the canonical example of this to be that a sphere passing through a plane appears on that plane as a point expanding to a circle and then contracting back to a point before disappearing. Therefore, the “four-dimensional” equivalent of a sphere would appear, passing through a three-dimensional space, as a dot expanding to a sphere and then contracting again to a dot.

    Is there a way to expand on this in terms of time travel, especially with regard to its paradoxes?

    There’s a fairly obvious sentimental analogy to the way humans pass through four dimensions (from tiny to big to tiny once again) but I’m not sure that’s very helpful.

  17. Meeg says:

    Kudos on the thought provoking article.

    I still have one question. You seem to suggest that unified theory/string theory/M theory (you rightly point out that these are basically speculative hypotheses with little-to-no physical evidence supporting them) still need to reckon with Einstein’s theory of general relativity. But, aren’t these theories to some extant a departure from Einstein’s model of the universe?

    Specifically, I thought that unified theorists were searching for a particle-based explanation for gravity which might replace Einstein’s curved spacetime explanation.

    I’d be curious to hear your thoughts on how much these theories diverge from Einstein and how much of general relativity they retain.

  18. dave says:

    Meeg –

    Theories of quantum gravity will, indeed, need to include a graviton in much the same way that quantum electromagnetism uses a photon. But in exactly the same way, 19th century electromagnetism (Maxwell’s theory) passed virtually every observational test at the time. When quantum mechanics was introduced, it was a way of explaining phenomena on very small scales. It _still_ had to square with Maxwell’s theory, however. The reason that we’re able to talk about radio “waves” is that for large scales/many photons, quantum mechanics reproduces the classical results exactly.

    So must it be with quantum gravity. On small scales — the centers of black holes, the instant of the big bang — quantum arguments must come into play. On larger scales, the limit of quantum gravity must still be general relativity.

    (Although I should point out that, strictly speaking, while String Theory is a “Theory of Everything”, it is not a theory of Quantum Gravity.)


Leave a Reply

Your email address will not be published. Required fields are marked *