Hacker News new | past | comments | ask | show | jobs | submit login
What Movies Get Wrong About Time Travel (slate.com)
60 points by edw519 on Aug 14, 2009 | hide | past | favorite | 65 comments



For a real mindfuck, check out Primer: http://en.wikipedia.org/wiki/Primer_%28film%29


I highly recommend you stop recommending Primer when time travel is brought up. The movie is better viewed not knowing anything about it. It could be a character drama before the weird shit starts going on. By the way, this is an awesome timeline http://neuwanstein.fw.hu/primer_timeline.html


I agree. I went into the movie knowing nothing and it was the biggest mindfuck ever.


Too bad any synopsis you read pretty much blows the surprise.


awesome, on another note i have a 30" computer monitor and i still had to scroll left


37" and still needed to scroll. Who's next?


pixels matter, bro. 30" mac display is 2560x1600


I'm one of those few people who completely hated Primer. I found it to be masquerading poor narrative as intellectual mystery. It's possible to present an intricate, complex logic puzzle as story without being totally incoherent, but Primer doesn't even bother.

For a movie that matches Primer in everything it does right, but none of what it does wrong, check out Timecrimes (Los Cronocrimenes). It is far superior.


Fully agreed!

Well, I didn't hate Primer, but genuinely thought Timecrimes was way superior. Plus the latter scared the shite outta me :-)


I agree! I Loved that film. Especially given the fact that it was made for so little. Huge fan. If you haven't seen it and are willing to put the intellectual effort in, buy it and watch it.


This is the only DVD I've bought in a store in the last 5 years. It's really an amazing film.


Truely confusing :-D but excellent movie!! :-P Very cleverly done.


Primer is an excellent film, especially when you consider that it was made for a total of only $7,000. That's not a typo: only seven thousand dollars!

Hollywood could really learn a lot from this movie.


Oh for goodness sake. Time Travel _does not exist_. Therefore you can't get anything wrong about it, except including it in the first place.

If you're allowing them to have time travel then you can only quibble about internal inconsistencies, not inconsistencies with the real world.


Geeks like to make "minimum assumptions," and then work everything else out from those. For example, the Minovski (or however you spell it) Particle is the only real change you need in physics before Gundams start looking like a good idea. If the emergent properties of the world, as shown through the work, cannot be deduced from the single "axiomatic" change, then the movie feels "wrong" to most people, whether they can bring it into their conscious mind or not.


You sound awfully sure of yourself.


I think his point is that this is like having a debate about how hyperspace works in Star Wars, or how phasers work in Star Trek. The tech is a plot device best assumed to be far outside the reach of today's science.

Part of that assumption is that the current version of physics we understand is incomplete and at worst wrong. There is little use quoting a source proven wrong within the world of the story.

Besides, I came from the future to write this comment, so the article is prima facie wrong.


There is a class of things it's possible to be very sure of.

For example, I feel very sure there's no such thing as blue weight. The expression "blue weight" is syntactically legal, but that doesn't mean it refers to anything. "Time travel" appears to be in the same category. Syntactically legal, but you find that if you try to express precisely what you mean by it, you can't. It's effectively a type error, like asking for the square root of a stream.

Incoherence as a concept is a different sort of impossible from mere technological "impossibility," like e.g. a pill you could take that would let you survive in outer space without a suit. You can say what that means, even if you can't do it. I have never heard a coherent explanation of what "time travel" could even refer to.


We're travelling through time right now, and not even at a fixed velocity given relativity. That makes the question whether you can control your velocity, not whether you can travel.


So we can change the velocity at which we travel through time? To what, 2 seconds per second?


Yes, you can travel two seconds forward in time for what appears to be one second to you. It depends on reference frames and velocities.


If you could travel back in time (not saying anything about forward), would we not have seen evidence of it by now? If you can travel back to a different "time stream" of sorts, then I'd argue its not really the same "time", but rather another one that happens to mirror the one you intended to travel back to (except that you're there when you shouldn't be and any changes you make).


RTFA


No, the article is retarded. It compares fiction with fact and then goes on about why the fiction is wrong - its fiction! Its not meant to be right. Also, we don't KNOW that what he said is actually how time travel would work, were it possible. So its stupid.


> 2) You can't visit any time before your time machine was built.

I think this is a very strange assertion to make. I find it more likely that one of the following is true:

1) Time travel is not possible

2) We'll wipe ourselves our before we discover it


Can somebody clear this up for me?

Relativity states that time slows down as you approach the speed of light. Given that fact, lets say I hop in a spaceship and zip around the earth at near light speed for several years (Yes, I know there are problems with this in terms of increased mass, energy needed, etc.) Ok, so after several years I slow down and land. For me, it seems as if I've traveled into the future becasue the people on earth have experienced more time than I have.

How is that not time travel, at least in a functional sense? I get that by strict definition it may not be, but the result would be the same. It seems like in this sense time travel is not only possible but proven.


In fact, that is time travel. In the same way that you travel into the future every day, since what was the present when you woke up in the morning is now the past. In this restricted sense, yes, time travel is possible and proven.

But when people refer to time travel, they generally mean arbitrary travel to nonsequential points in time. Going from 22:00 today to 17:00 yesterday to 10:00 today.

The difference here is between floating down a river with a strong current and being able to slow down or speed up your progress a bit (the restricted sense of "time travel" that you mentioned), and taking a ship out onto the ocean and going where you want (the more typical use of the term "time travel").


This is indeed what will happen, you will arrive at Earth finding the planet and the people much older than yourself. (It can be by years or by millions of years, depending on speed and duration of travel.)

If you wish to call that time travel, you may. I guess people just reserve the term "time travel" for travel backwards in time, since that is so much cooler.


Hey, I'd rather see the future than the past anyway :)



What you've witnessed is merely an appeared acceleration of time, but you didn't "leap" through time, in the traditional time-travel sense.

Technically, we're all traveling through time. In your case, you would simply be moving a little faster.


And this is the main point that makes time travel in movies completely unbelievable to me. According to the theory or relativity, there are no absolute coordinates. And even if their were, is earth the center of the universe that stands still? So how could you jump for one time to another and expect to land in the same relative coordinates to the earth. As the author of the articles says, there would need to be end points at both ends of the leap. Like two time machines. Or some naturally occurring end points. The only other option is to move through time and space at the same time(much like we do naturally), and not instantaneously jump through time.


If there are no absolute coordinates then there is an arbitrary choice of coordinate system and you are welcome to choose your jumping off point as fixed.

When I jump I push the entire universe down by about 30cm.

Makes the maths a reall bugger to work out though.


You don't travel through time. Time is not a physical thing that you can move through. Time is not a river, or a place on a map.

What is time on an atomic level? It's the speed at which interaction occur, transfer of energy mainly.

When you move at relativistic speeds those interactions occur "slower" (which is a self referential statement, but nothing I can do about that).

But you do not travel though time - you don't travel at all. The various forces that affect you work slower. It's like slowing down the motor of an assembly line robot. You are not moving through time just that things that you are made of do their thing less often.

Since you are moving slower, the rest of the word seems to move faster.

Which is why time travel is impossible - there is no place to travel to.


I fail to understand how time can slow. So way beyond me as an HS grad.

If two people program 60 lines of code per hour, one does so in a spaceship traveling nearly the speed of light and one does so on earth... after 1 year/365 days/8760 hrs they make a report on their progress and they don't have the same results? If a watch ticks every second, the faster you go the slower it ticks? Why? How?

Will people be buying a 2100 model Bugatti, hiring a driver, and working in the passenger seat to be more productive?


Time is relative to the coder in question. Neither programmer will be any more productive.

So, each person experiences 1 second ticking as normal, but if you were to look at both watches together, one would be much, much slower than the other.

This has actually been proven by syncing atomic clocks and putting one on the space shuttle. The one on the shuttle comes back showing a time a few milliseconds less than the one on the ground.


> Will people be buying a 2100 model Bugatti, hiring a driver, and working in the passenger seat to be more productive?

They'll be very slightly less productive in that situation because time has slowed down for them. For everyone sitting at home, time moves 1 second every 1 second. For passenger in Bugatti, a stopwatch says 0.99999* second for every 1 second felt by everyone else. So what feels 1 second to the passenger is 1/0.99999 = 1.000001 to everyone else. Hence he's actually slower.



So you got person A on earth writing code and person B going around REALLY fast (very close to the speed of light lets say) writing the same code.

When person A finishes (say a week later) and comes back, he'll find that person B finished "a long time ago." That is, person B finished, and then waited till person A came back (say for a year, depending on how fast person B was going).

What happened? Time went at different speeds for each person. They each took a week from their own point of view!

Which one is right? Both of them. Thats what relativity says. There is not one thing called "time" but that its "relative" (hence the name).

So, if your going to choose to hire one of them... you'd hire person B (assuming your business is on Earth).


You got it backwards... person A's week is much faster than the dude in the spaceship. (Hence person B "travels into the future")


"Time travel" is real, but the only thing you can do is slow down relative to others. The way you describe is a valid method, though some might quibble that it's the accelerations involved that slow down time for you, rather than simply the movement at a high velocity.

You can also "time travel" by moving to an area of higher gravity (per general relativity). Time travels more slowly for you there, so you can let your original location get ahead of you, then go back there to effectively travel to "the future".

Of course, gravitational time dilation is a tiny effect, so you'd have to get near something unhealthy like a black hole to notice much difference.


This is the time travel I like to think about: Suppose we actually are simulated ancestors of an advanced civilization. Then time travel can take all sorts of interesting forms.

For example, suppose our simulation is run sequentially. Then someone traveling back in time would travel back to an earlier state in the simulation that is already out of "RAM" and stored. This agent is "dead", the simulator does not bother to re-simulate the past just for them, they just disappear.

Suppose instead the simulator is more flexible; then there's no reason they couldn't "fork" the timeline and let the gal do whatever she felt like and change the past going forward on their own. Since this would take vast computational resources, it would only be worth doing if you changed something big so maybe time travel wouldn't work unless your plans were grand enough.

This also resolves the grandfather paradox; your grandpa is dead, you are never born in this timeline, but your simulated agent is just fine since it was copied from a different code base. You live out your life, reproduce, share technology, do whatever you want to do and life goes forward normally. In fact, it'd be likely they wouldn't bother simulating the whole world but just pull a Truman Show kind of thing and only propagate the changes, and copy from old data elsewhere.

I'm sure there are even more interesting kinds of time-travel you could come up with given a simulated universe. It actually could be fun to write some sci fi short stories with this in the background, unexplained.


> I'm sure there are even more interesting kinds of time-travel you could come up with given a simulated universe. It actually could be fun to write some sci fi short stories with this in the background, unexplained.

I'm actually working with this exact idea, more or less, for a [roughly proposed] series of books, the first of which I'm procrastinating upon as I type this. :) Time travel, though, is a rather small part of what is possible with my metaverse model. I assume that the simulations are infinitely stacked, and each one has a 0:1 time-passage correspondence, so that for a "parent" simulation, the "child" simulation runs instantaneously, no matter the in-sim execution time, thus making Hypercomputation and such possible. All sorts of fun-with-philosophy falls out of that:

The protagonists are Star Trek-style "everything is science" types, who discover that magic exists (for "magic," read "simulation runtime bugs.") They go on "Hard Science-based Fantasy" adventures using said magic for ten or twelve books, before figuring out that it was (very complicated) science all along. They then care for their own sim for a while, hacking it to fix some kinks (such as the inevitable existence, given probability, of a malevolent entity who has the root password.) Bored of that, they proceed to "break out" of their simulation, repeatedly, visiting "higher planes" that each have radically different physics and sometimes even logic, but that all share the property that Turing machines can be constructed within them. Eventually, they get to the "root" universe (really just a chroot jail) and meet God—that is to say, an AI who was left to run for an infinite time on an analogue computer and thus became omniscient merely by evaluating all possible states of all possible universes (that it could detect from within the chroot.)

And that's only the first half of the sequence. 's called "Infinity's Tale", by the way. :)


Now that's singularity fiction; I think speculative fiction really does well when there's a case for a truth or a worldview underneath, and like you said there are lots of "fun-with-philosophy" implications here. Good luck shipping!


The point of this article isn't so much to prove some theories incorrect or impossible. As time travel doesn't really exist we cannot say anything about it and say its a fact. Its about logic. If you are making a movie about time travel, than do it the way it makes sense.

There are to many movies this days that employ some kind of time traveling mechanics which make no sense at all. Instead of using time travel as a good plot device, it is more and more used as a deus ex. Writers aren't able to think of better ways to resolve the plot. Don't you just hate it when that happens. Oh, no, the main character died, its all over, the ship is destroyed and the world has ended. But no, wait! What's that? Oh, its OK. We managed to go back in time and prevent the tragedy, and now we can all go on with our lives forgeting that we just found a way to travel through time.

Now that i think about it, there is not a single movie including time traveling that hasn't got some kind of a plot hole. But that's ok, as you said, its a movie and it doesn't have to be all real. But it has to have some sense and meaning. Using time travel as a main plot device where it is not necessary or even sane, is a serious crime.

The actual act of time traveling isn't that important, or the science behind it. There are many theories and no one can say one is the best. A far far more important aspect is how you handle the story and the world after the actual act. Its easy to think of a functional super cool futuristic time machine, but it is hard to think of what to do with it once it is actually there.


1) movies are works of fiction.


I would put a spoiler alert on this.


I prefer the theory of the parallel universes over the self-correction of the time.

Most movies go that way too, since otherwise, the plot would be fatalistic (eg: '12 monkeys'). So they aren't really wrong.


I'm too tired to explain it here, but I like the explanation of time travel in the book Cowl: http://en.wikipedia.org/wiki/Cowl_(novel)


Sounds like a good book... thoughts on the story?


I personally really liked the story. It had some interesting ideas and the author seems (IMHO) quite good. I'd reccomend it.


IMO the simplest resolution of things like the grandfather paradox are obtained by including quantum mechanics in the equation.

Usually in quantum mechanics we find that waves are quantized because they must be periodic over very tiny reaches of space, and the quantization comes about because only certain frequencies of wave can meet themselves over these distances.

But if we have time loops, it's pretty much the same thing, except the consistency conditions can be experienced on arbitrarily large scales - the quantum state where I kill my own grandfather is not a self-consistent loop, and won't exist as a solution to the wave equations. In essence, the "wave" that describes me killing my grandfather will interfere with itself over the time loop, and will eventually (hopefully? I don't have a proof...) converge on a wave that does reinforce itself, and ends up consistent. This may require a pretty serious divergence from what we'd usually assume about classical physics, even at the macroscopic scale.

So quantization is the real reason we shouldn't worry about the grandfather paradox; any solution to the equations of QM will have to be self consistent over a closed timelike loop, even if the results are ridiculous as far as classical mechanics is concerned.


Dismissing Everitt branches by saying "there is no evidence," is totally off base. There's a mountain of math that practically demands that it's true.


Theory != Evidence

"Evidence" is by definition empirical; it's generally described as 'the result of empirical observation' or something substantially similar.


It's more of a gradient, not a strict inequality; cf. argument from parsimony, inference vs. deduction.


Isn't there a fair bit of quantum physics theory that's based in this?


I'm really surprised that guy hasn't seen Primer.


He's arguing for a non-forking timeline, which is nice, but I like soft science fiction with stories rather than people spending much of the story monkeying around with technology so that it's consistent with what we currently know about science.

I'll quote something I wrote elsewhere: The two major time travel literary views are forking & non-forking timelines:

* The game Achron has forking with merges via non-instantaneous time waves, sorta like Back to the Future.

* Lost is not forking. I think.

* The original Terminator's non-forking, but the sequels & series have forking timelines.

* Forking: Primer, The Butterfly Effect.

* Non-forking: 12 Monkeys, Slaughterhouse Five, Bill & Ted, All You Zombies.


The book on time travel describing model which made most sense to me for years: http://en.wikipedia.org/wiki/The_Technicolor_Time_Machine

This novel presents a clear use of the restricted action resolution of the Grandfather paradox. The actions of the characters can not have changed the past because the past is what their actions brought about.


It makes sense on a social level, but not on a physical one. No matter what you do, you change things - you ate, you breathed.


Is it just me, or does the concept of "near speed of light" not make any sense? I was under the impression that light travels at a constant speed no matter how you are moving. And also, since there are no absolute coordinates, the concept of moving only make sense in relation to other things. Isn't it impossible to move "near speed of light"? Isn't light always going to move C faster then you?


> Isn't light always going to move C faster then you?

Yes.

> Isn't it impossible to move "near speed of light"?

From your perspective, you'll always be slower than light by C. Your speed relative to another observer, however, can be arbitrarily close to C.

In classical mechanics, your speed relative to the other observer could add to C, resulting in "your" light, from her point of view, traveling faster than C. To avoid that discrepancy, special relativity postulates that her (or your, depending on the POV) movement through time is slowed until you both agree that light is moving at precisely C.


"Near the speed of light" carries with it the implicit disclaimer of "relative to a stationary observer" (that is why the theory has the word "relativity" in its name :)

Light travels at C (through a vacuum) regardless of your frame of reference, so it used as the comparator for speed/velocity among different frames of reference.


The speed of light in a vacuum being [constantly and universally] constant is an axiom (ie assumption, belief) of special relativity so it applies to inertial frames of reference.

So, strictly it is not "regardless of your frame of reference". Accelerating frames, eg rotating or accelerating (eg under gravity where General relativity applies) require special treatments.


I always thought the moniker "time travel" was not specific enough. I'm traveling through time right now. I can't help but travel through time. Maybe a more appropriate term would be "time navigator."




Consider applying for YC's Spring batch! Applications are open till Feb 11.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: