I'm working on a simulator-type game which I want to be at least internally consistent, and which I'd like to work as close to reality as possible. That being said, it's set in space, which means to avoid player boredom there needs to be some kind of FTL travel.

I'm trying to work in functional analogs to the four general categories of FTL travel (warp, jump, gate, and hyperspace) and I think I've managed to work it down to only two core mechanics – the Alcubierre Drive and a modified version of Alderson Points. Originally I was going to use stabilized wormholes in place of Alderson Points, but after looking into it I found that that makes it easy to build a backwards time machine (which is annoying/impossible to reliably implement in something like an MMORPG), so after failing to figure out how to "break" all of the versions of the time machine as per the Chronology Protection Conjecture, I gave up and started working through other mechanics.

After expanding on the original Alderson Drive concept for story, time, and map reasons, I've come up with the following list of desired/assumed mechanics for the points.

  • Suitable points exist naturally (~1-5 per star system) positioned based on rules which have yet to be determined, but preferably there's always at least one within 0.3 AU of the habitable zone.
  • Points don't "orbit" the host star, and they probably don't exist in and of themselves; they are either stationary relative to the star or they revolve based on the position of other large bodies similar to the Lagrange points. (One or the other, not both, I just haven't decided yet in case physics decides for me.)
  • Points are tied to networks; they can all connect to more than one point, but only within their network. Being in the same network involves always having the same inertial frame of reference, though that may just be a side effect of some underlying reason.
  • Using the points involves "charging up" the points with a device on your ship for an amount of time (i.e. not instantaneous or measured in nanoseconds, but also not measured in hours or days unless the drive is broken or poorly tuned). The destination is selected by how much energy is "dumped in." (This will likely be proportional to the distance between the points, but may need to be related to the ship's mass as well.)
  • While the point is being charged, both the source and the destination with the energy level closest to the current level give some kind of indication that they're being activated (e.g. "spontaneous" EM radiation, visible or otherwise, possibly similar to black body radiation but emitted from a field).
  • When the drive is turned off, provided the input energy precisely matches what is required to connect to a destination, all of the the matter within a certain radius of the source and destination points is instantaneously swapped. (The definition of instantaneous, in this case, is "now" according to the point network's reference frame.)
  • Artificial point networks can be constructed using devices ("stargates"), though this may not actually need to be explained because I could easily say that "the Ancients/Precursors/Forerunners built the devices and we don't know how they work" or something to that effect.

Now here are my questions:

  • Most importantly, can this set of mechanics be used by itself to violate causality – and not just apparently from an extraneous reference frame, but literally be used to either tell yourself not to do something before you were going to do it or, in the more serious case, "take half a critical mass of Plutonium back to meet itself"? If so, what could I change to make it impossible?
  • Can it be used in concert with something else to violate causality? If so, what would be necessary/what should I watch out for?
  • Are there conditions related to causality violations that could dictate where I need to place the points inside of the star systems or the nodes in the point networks inside of different star systems?
  • Is there any possible scientific or pseudo-scientific base I could latch onto to try to explain some/all of these mechanics? (The original version mentions "Equipotential Thermonuclear Flux" but I thought I'd try finding something that sounds a little more real...)
  • What should happen near the "edges" of the effective area of the points? Or should I just say "nothing good happens" and make determining the radius a function of the drive construction so that individual ships shouldn't ever have to care?
  • Is there a way (plausible or not) that I could pull this out into also enabling FTL communication? Because, like I said, my accidental time machine ruled out movable wormholes, and therefore my only lead on how to make an Ansible. If I have to, I could use something like the "space pony express" where ships download mail, pop through, and offload it, and for urgent/secret messages use unmanned probe "carrier pigeons," but I'd like to figure out a way to do it without physically sending matter to the destination (i.e. "subspace radio") if possible.

(Note: I would have asked this on the Physics StackExchange, but I'm pretty sure it crosses the fictional "line" so I decided to ask here first because of their definition of "on-topic.")

  • $\begingroup$ How is that any different from wormholes? I think the point is that we can't move or accelerate them so as to contrive the time machine. And, being a natural phenomona they would only be present where it doesn't cause loops. My own favorite idea is that all FTL transits are in the same reference frame. $\endgroup$
    – JDługosz
    Commented Jan 22, 2015 at 3:14
  • $\begingroup$ Well, the theory was that this would be different from wormholes because whatever causes the transit exists for less than any measurable amount of time; it wouldn't have time to break away from the point's reference frame. $\endgroup$ Commented Jan 22, 2015 at 21:52
  • $\begingroup$ Use wormholes. Every time they are moved to create a time machine then photons start to loop round the path. This is a physical mechanism to collapse the wormhole as soon as it becomes a time machine. $\endgroup$ Commented Sep 4, 2017 at 13:43
  • $\begingroup$ The concept sounds like the one used in the game Aurora: aurorawiki.pentarch.org/index.php?title=New_Game . You might want to do some research there to see how you want your concept to be similar or different. $\endgroup$
    – ShadoCat
    Commented Dec 21, 2018 at 21:56
  • $\begingroup$ Related worldbuilding.stackexchange.com/q/46873/809 $\endgroup$
    – Mołot
    Commented Dec 22, 2018 at 8:16

7 Answers 7


FTL time paradoxes generally use one or both of the following circumstances:

  1. A round trip between two objects that are moving (relative to each other) at a significant fraction of light speed.
  2. A wormhole that allows physical contact between two separated locations simultaneously.

An Alderson style system can avoid trouble if:

  1. The traveling object is (briefly) removed from communication with the universe, then brought back in at the other end.
  2. The endpoints are reasonably stationary and cannot be artificially moved.
  3. You cannot return to your starting point "quickly" (aka before you left).

Here's some math which may or may not make physical sense (IANAP). The Sun and Wolf 424 have a relative motion of about 555 km/sec (0.0018 c), with a relative time dilation of 0.0000017. At 14 light year distance, that's a discrepancy of about 13 minutes. As long as your charge up delay is that long, causality should be safe. That should be the worst case; other nearby stars have less relative motion, allowing for shorter delays.

  • 3
    $\begingroup$ It's not just "reasonably" stationary. They have to entirely share the same inertial frame. If the point do not share the same inertial frame, then you will (eventually) create a causality violation. $\endgroup$
    – Nick2253
    Commented Jan 22, 2015 at 19:49
  • $\begingroup$ Not if they have a corresponding (and negating) difference in gravitational potential. physics.stackexchange.com/questions/62222/… I can't do the calculations to solve this, but it would entail precisely matched start and end points for each jump, so multiple destinations from the same point would be difficult. I suspect it might even require different placements for each direction (i.e. A to B vs B to A would use different relative gravities). $\endgroup$
    – Foo Bar
    Commented Jan 22, 2015 at 21:11
  • 1
    $\begingroup$ Controlling the required charge time is brilliant! I hadn't thought of that. How would this apply to points that are members of different networks (i.e. different reference frames) existing within the same star system? $\endgroup$ Commented Jan 22, 2015 at 21:57
  • $\begingroup$ @AdamHovorka two ideas: if it would get out of allowed parameters, the jump won't work; perhaps the point is missing or it takes infinite energy like a wall, is in the way. More dramatic idea is that the burden on the user: rather than not working, it is very unwise to attempt. Maybe the ship never reappears, or Fate prevents you from trying usually with abrupt and unpleasant events. $\endgroup$
    – JDługosz
    Commented Jan 23, 2015 at 2:33
  • 1
    $\begingroup$ @FooBar The velocity discrepancy of itself provides a barrier to a quick return. A spacecraft decelerating at 1 g will take 15.73 hours to arrive a relative rest. If it turns around to head back into the Alderson Point that's another 15.73 hours. A 31. 46 hour delay is more than sufficient to counteract a 13 minute time discrepancy. Another possibility: the Alderson point itself collapses if a jump is attempted that goes back in time. Alderson points may restore or themselves or reform after a suitable lapse of time, say, the light-time distance between the two systems. $\endgroup$
    – a4android
    Commented Jul 10, 2016 at 10:09

If you can do FTL travel, you're going back in time according to some potential observers. If you can do FTL travel in different reference frames, you've got time travel. If you say FTL only works in one reference frame, you're ditching Special Relativity, and I'd rather not do that. It is the solution of choice for fiction, since it isn't completely intuitive and audiences usually don't notice it being broken.

The idea of putting the Alderson points all in one reference frame seems brittle to me, at least if you like relativity. It's extremely unlikely that a universal reference frame will match ours reasonably well, and so it works better if the reference frame can change gradually over space.

Given that you have a fixed number of discrete Alderson points, you can fake an ansible pretty easily. You have information-carrying drones that do endless round trips between Alderson points. They pop out of the communication station near point A, hop to B, transmit and receive information with the communication station near point B, then hop back. Repeat as desired. Lightspeed delays over 0.3 AU could be more significant than drone transmission delays.

In my story, I just had an FTL drive that I slapped a name on. It could propel things at up to a certain speed. There were no theoretical limitations, just "this is how fast we can go with current technology". FTL was relative to the mass in a large volume around the ship, which means the reference frames can't be very far within, say, a thousand light-years of Earth. With FTL but not instantaneous movement, and a limited choice of reference frames, there's no time travel going on right now. Increase possible speeds too much and it becomes possible. Go sufficiently far away to get a different reference frame and it becomes possible. With what we've got now? Not a chance.


Disclaimer: I'm not a physicist.

That begin said:

Short Version: From what I understand from the various expeditions in the world of Physics.SE FTL-questions, I'd say any kind of FTL travel or communication will allow causality violation.

Longer version:

There are a few potential problems I see with your description: First of all, how can points belonging to the same network share the same inertial reference frame, and still move relative to their respective stars? Check with your trusted physicist, but as far as I understand, inertial reference frame means no relative motion between each other. Also, I don't know what being in the same inertial reference frame, but far apart, would do, and if it matches your expectations of having a well-defined "now"/"simultaneous"/"instantaneous".

Typically, violating causality goes like this(crude explanation, look at the many answers on Physics.SE for a better one): Move from point A to point B at superluminar speed. Now move back. You end up in a reference frame at position A but before you left.

Remember that "simultaneous" is something very unintuitive in Relativity (not sure whether that's General or Special Relativity), and depends on the observer (=is relative).

But this is fiction, you could just say: We found these things, they don't allow for time travel, so all this Relativity stuff wasn't 100% right after all. What's the problem with that?

  • $\begingroup$ I was hoping that there was some clever way to arrange them which would make the transit time between points in different networks enough to render that specific version ineffective, but I'm not sure how to definitively prove that it's impossible within a certain arrangement. $\endgroup$ Commented Jan 21, 2015 at 15:03
  • $\begingroup$ I second what @kutschkem said, your game is a work of fiction, you're seriously overthinking it by trying to solve FTL travel paradoxes for real. Just move the player instantaneously from point A to B or with some time delay to simulate actual travel and thats all. Isn't it? The communication thing is the same as the travel thing. If you allow matter to travel at FTL speed there is no reason to not allow information to do the same. If you do not want to send matter/energy in a channel look at entanglement as a means of comms (plus.google.com/114068886964011047974/posts/BQ4ZnKuh5Zs) $\endgroup$
    – bor
    Commented Jan 21, 2015 at 19:42
  • 1
    $\begingroup$ To answer your first point, the points are not mouths embeded in our normal spacetime with a world line. They are mathematical positions that map the world line of one to the world line of another, such that they are the same t in some special reference frame. $\endgroup$
    – JDługosz
    Commented Jan 22, 2015 at 3:22
  • $\begingroup$ @AdamHovorka There is a clever way to arrange the transit time between points to prevent causality violation, and the universe already does it. That’s one definition of the speed of light. Just have transits between your Alderson points take as much time as a photon would take to move between the points, and you will have no causality violation (and also of course no FTL). $\endgroup$
    – Mike Scott
    Commented Dec 24, 2018 at 6:54

The key requirement for FTL time travel is making two separate FTL jumps in two different reference frames. Thus, to prevent paradoxes, you simply need to pick one reference frame (e.g. that of the cosmic microwave background, which the closest thing to a universal stationary reference frame in the universe) and declare it as the only frame in which FTL travel can take place.

With your Alderson-based system, this is quite simple: Make all swaps instantaneous in the CMB frame. And you're done.

The Alderson points themselves don't have to be stationary relative to anything. If one point is moving rapidly or deep in a gravity well (e.g. orbiting close to a black hole), it will experience time more slowly than a point nearly at rest relative to the CMB in interstellar space on the outskirts of the galaxy. However, this just means that clocks attached to these points will disagree on how much time passes between consecutive swaps. There's no way to use this sort of system to create any kind of truly reality-breaking paradox. No meeting yourself, no murdering your grandfather; nothing stranger than the classic "twin paradox" from special relativity.

If you want to use this system to take a few good close-up photos of Sagittarius A* (the supermassive black hole at the center of the Milky Way galaxy), you might arrive back home a few weeks (or months, or years, or centuries, depending on how close up your your photos are) after you left, despite only experiencing a few minutes yourself, but that's about the worst that might happen.

For FTL communications, the simplest solution given the existing Alderson system would indeed be Alderson couriers- your "space pony express". If you want to make it less of a hassle, you'll need some additional mechanic. Maybe the energy required to power an Alderson swap increases with the size and mass of the object being transported. Flash drives are small and don't weigh much, so sending them through would be pretty cheap. Alternatively, you might be able to rig the Alderson points to swap extremely rapidly, with a period approximately equal to the time it takes for light to traverse its area of effect. Light beamed at the point will enter the point's area of effect, get swapped to a point in the target system, then exit the area of effect before being swapped again. This makes the pair of points act like a two-way wormhole that only light can cross- and since light has no mass, these swaps may be cheap enough to make the rapid cycling feasible.

For more info on FTL time travel, check out this video from PBS Space Time, which explains how a ship equipped with an Alcubierre drive with a maximum speed of only twice the speed of light could fly off into space and return home before it was built. Note that to pull it off, the ship must burn its Alcubierre drive twice: once in one reference frame (the Earth's, in this case), and once in a reference frame moving very close to the speed of light relative to Earth. This change of reference frame is the key, and an Alcubierre-equipped ship can probably pull it off by using conventional sub-lightspeed engines to accelerate itself to a substantial fraction of the speed of light between the Alcubierre burns.

If, however, the only available FTL tech is locked to a single frame of reference, all these paradoxical time-traveling shenanigans become impossible. This means no Alcubierre drives that don't cook themselves in infinitely-hot Hawking radiation (or just collapse into black holes the moment you turn them on), no stable wormholes that don't explode with virtual particle feedback (or, once again, collapse into black holes) the moment they're time-dilated and brought close enough together to permit backward time travel, and definitely no tachyonic antitelephones. But your Alderson system, with the restriction that all instantaneous FTL swaps occur in the same inertial reference frame, works just fine. It doesn't necessarily have to be the CMB frame; the CMB is just a nice example of a frame that anyone can measure their velocity relative to no matter where they are in the entire universe, as long as they have the right kind of telescope.


The short answer to the title question is no. I don't even need to know what you've built or how it works, there's a thing called Time's Arrow, or macro-system temporal asymmetry; basically it says 3-Dimensional objects move forward through time, period. No travel back in time is possible if you're a 3-Dimensional being. It's the reason Entropy and a couple of physical constants like Planck and Boltzmman exist.

As far as I can tell while we can see how Time's Arrow effects the universe we don't know why it exists (it does seem to allow us to see the universe in the first place). It means that causality violating "space-time geographies" are inaccessible so all FTL has to be locked into the same Inertial Frame of Reference so you can't get out of you "Light Cone" to violate causality.

  • $\begingroup$ Your Time's Arrow link doesn't say that going back in time is impossible, just that there is a way to tell the future from the past. If all FTL is locked into one inertial frame of reference, you've just invalidated Special Relativity, which seems to be ditched almost routinely in fiction. $\endgroup$ Commented Dec 21, 2018 at 22:06
  • $\begingroup$ @DavidThornley The first sentence is "The arrow of time, or time's arrow is the concept positing the "one-way direction" or "asymmetry" of time." Meaning time travels in one direction and the three dimensional universe does as well, you have to violate this idea to travel backwards through time. $\endgroup$
    – Ash
    Commented Feb 24, 2019 at 11:20

Any kind of FTL at any "velocity" of FTL can potentially create a temporal paradox, so it doesn't matter what type you use so long as you set up some rule to prevent it from causing one. Any kind of warp drive is probably the worst example because at first blush it seems hard to make sure you can't do it due to the relative freedom of travel. Wormholes and any sort of jump points make it easier: you simply state that the second (or 3rd, 4th, whatever) trip that would cause a causality violation--a one-way trip really won't do it--prevents the jump/wormhole from allowing passage until such time as it would not violate causality.


If you limit your Warp Points (WPs) to occur only around starts and can't be moved then you only have to deal with causality if you can move the star.

Also, give the WPs a non-zero transit time.

The non-mobility of the WPs should do away with any meaningful time shenanigans since you can't accelerate one WP much more than the other one is being accelerated. The transit time can be manipulated to cancel out any distortions due to acceleration (the greater the difference in acceleration between two WPs, the greater the transit time). This should negate the effect even if they can some how move the WPs.

  • $\begingroup$ If you can 2-way travel through a warp point at FTL, you can potentially cause a paradox, regardless of how much faster than light you are going ie, no matter how long it takes to transit the wormhole, so long as you're exceeding the speed of light. C+1 m/s can potentially result in a paradox. $\endgroup$ Commented Dec 21, 2018 at 22:27
  • $\begingroup$ @KeithMorrison, I disagree on a functional level. The closest star system is 4.6 LY from Earth. That's too far away for information from Earth to cause a causality issue unless the other end had been accelerated to cause 4.6+ years of dilation. Any dilation from the movement of the wormholes around the stars should be in the millisecond range. So if the wormhole has a transit time => the dilation, you can't get back before you left. $\endgroup$
    – ShadoCat
    Commented Dec 22, 2018 at 0:56
  • $\begingroup$ It doesn't matter what the ends of the wormhole do. Add in someone travelling who shifts reference frames by going at relativistic speeds in normal space. It's possible to arrange things so that they come out of one end of the wormhole before they enter it, (or send a message telling themselves not to go in, or tell someone to stop them from first going in, or whatever) thus causality violation. $\endgroup$ Commented Dec 23, 2018 at 4:23
  • $\begingroup$ @KeithMorrison, I suppose that you will have to explain how traveling at relativistic speeds is going to cause causality issues over a 4.6 LY distance. My understanding is that you have to accelerate one end and then make sure that the ends are close enough together for communication to affect the initial entrance. $\endgroup$
    – ShadoCat
    Commented Dec 28, 2018 at 1:48
  • $\begingroup$ I'm not talking about the wormhole itself except that the wormhole is used to transmit info faster than light. I'm going to give you an example of what I mean. Here's the setup: you have two ships traveling at relativistic velocity relative to each other such that each would see the other's clock running 50% slower. From each ship's perspective, their clock is running normally, the other guy's is running slow. Basic relativity, okay? Now on to part 2. $\endgroup$ Commented Dec 29, 2018 at 20:04

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .