The problem with FTL is that, due to the tachyonic antitelephone, it allows causality violation: I send you a superluminal message (you are moving away from me at relativistic speed), you reply with a superluminal message that reaches me before I send the first one.

It is crucial to see that this causality violation happens because of the return message.

Now, suppose the very act of sending a FTL message creates a bubble around the sender that may not be entered by FTL signals. Wikipedia on antitelephone gives the delay $T$ between my message and my receipt of the reply as $T = (\frac{1}{a}+\frac{1-av}{a-v})L$. ($a$ is message speed and $v$ is our relative velocity in units of $c$; $L$ is distance). Even if $v$ approaches one (relative velocity approaches $c$), $T$ will be some finite (likely negative) amount. $T$ remains finite even for $a$ approaching infinity. From this I conclude that we don't need to prevent return messages for eternity, just for some finite amount of time.

In other words, there's no need to have FTL travel/communication only into one direction, we can loop back if we wait long enough.

What would be the smallest such bubble and why? What shape would it have, how would it evolve over time?

The simplest assumption would be a sphere surrounding the first sender, shrinking with $c$. However, I see no reason why the bubble should expand in all directions equally, it ought to be flatter on the backside.

As we see, $v$, the relative velocity of both participants in the message exchange, goes into $T$; so assume $v=1$ or close to it so that size and shape of our bubble depends solely on features of the FTL device - namely speed of signal $a$ and distance travelled $L$, not on nearby random objects. Also assume the FTL signal travels point to point without affecting anything inbetween.

Different values for $a$ for the return message need not be considered.

I've looked at this and this question and I don't see my specific idea addressed - the closest is a4android's answer, but they only go into unidirectional superluminal travel and I think we need not go that far, as outlined above.

  • $\begingroup$ Can't the person just use the FTL to travel back and break any rules that you create? $\endgroup$
    – Andrey
    May 15, 2017 at 14:02
  • $\begingroup$ I could possibly answer this if anybody could explain to me why FTL causes time travel. I've not seen an explanation that adds up. Surely all it means is that the light-cone is replaced by the FTL-cone of your new cause-effect structure, a wider flatter cone that sits outside the light-cone and gives meaning to that which previously did not have meaning. $\endgroup$
    – Separatrix
    May 15, 2017 at 19:03
  • $\begingroup$ first link in the question explains the problem. $\endgroup$
    – mart
    May 15, 2017 at 20:42
  • $\begingroup$ I'm still stuck looking at it from an external absolute perspective, the way Lorentz himself did, I can't get into Einstein's relative perspective. $\endgroup$
    – Separatrix
    May 16, 2017 at 7:36
  • $\begingroup$ @separatrix the issue is with choosing such an absolute perspective. For any FTL travel, there will be an inertial frame that says the travel is backwards in time. Once you pick an inertial frame and force it onto everyone, there is no option for relativity paradoxes. $\endgroup$ May 21, 2017 at 17:36

2 Answers 2


The problem is that the length of the reception blackout you need to impose will vary based on the velocity and distance of the other party. In a nutshell, sending a message will blackout reception for a time equal to the conventional round-trip light speed of the channel, or a time based on the distinguished reference frame rule.

Now how do you know how far away the recipient is, and how fast he’s moving when he attempts to reply later?

You end up having to implement the “Empire Time” as an imposed constraint, checking the timestamp on any incoming message and ignoring it if it’s before present. How could you make that happen as a law of nature? It would require a chronology protection rule to be built into the fabric of nature.

An interesting twist would be that the chronology protection doesn’t actually prevent the receiver from working, but causes “very bad things” to happen via fate if it is violated (think of the Final Destination movies. So the rule is programmed into the machine and message protocol.

  • $\begingroup$ I've written that you should assume velocity of the other party = c (worst case), then the neccessary blackout time rapidly approaches c * L for high values of a. Will make more explicit that blackout time will depend on distance as well. $\endgroup$
    – mart
    May 15, 2017 at 13:40
  • $\begingroup$ Cease with the wimpy Final Destination conception. See Larry Niven's "Rotating Cylinders and the Possibility of Global Causality Violation". Think supernova. $\endgroup$ May 15, 2017 at 15:41
  • $\begingroup$ @WhatRoughBeast Nice science-fiction concept, but lousy physics. A star like our Sun won't go supernova -- unless the chronology protection principle does something very strange and otherwise unphysical. $\endgroup$
    – a4android
    May 15, 2017 at 23:37
  • $\begingroup$ @a4android - Oh yeah, the mechanism is never explained, The story employs a future setting in which a researcher has extrapolated the principle from remnant evidence of several previous (extinct) stellar civilizations. He convinces his emperor to leak the principle of a time machine (hence the title) to their enemy, so that the enemy will try to build it and be wiped out. But the simple decision to start the process triggers their own supernova. Oops. Think of it as a really strong equivalent of "Nature abhors a vacuum." $\endgroup$ May 16, 2017 at 17:34
  • $\begingroup$ @WhatRoughBeast it was a Tipler cylinder, artifact from an extinct culture. The thought was to feint interest in expanding borders in that direction, to get the enemy to waste resources. $\endgroup$
    – JDługosz
    May 16, 2017 at 18:58

With superluminal communications in the tachyon antitelephone scenario implies that FTL communications only happen if and only if they don't happen.

However, there are several scenarios where FTL communications can occur and how they work out will be different.

Assume there is a chronology protection principle (CPP), then superluminal messages shouldn't be able to be received in the past. This will remove any problems with causality violation from the picture. This scenario only allows forward in time superluminal communications.

Starting with equation for a round-trip time delay for FTL communications

       $T = (\frac{1}{a}+\frac{1-av}{a-v})L$

There are two possible scenarios that are workable.

If the message velocity a is infinite, then the time delay will be almost effectively zero. Any communications delay will be due to response in the FTL communications itself.

If instead of the spacecraft you are trying to communicate moving a t relativistic velocity, it decelerates and comes to relative rest with respect to the position where the transmitter is located. The relative velocity v is now zero. The time delay equation is now: --

       $T = (\frac{2}{a})L$

The time delay is due to the superluminal message travelling back and forth between your location and spacecraft you are contacting.

Communications in this scenario requires both parties to be at relative with respect to each other. This can only be done if both you and the ship have agreed in advance when you will communicate with each other. There is almost no point in either party simply deciding to call without prior arrangement. (1)

The usual solution used to resolve causality violation problems is the preferred frame of reference. Vide @JDlugosz's answer for this model. However, there is another resolution, but isn't used commonly, because it isn't easy to wrap our heads around. Essentially superluminal messages travelling backwards in time when they are received they do not come from our future. Instead they come from the future of another worldline (effectively parallel timelines) and this thereby circumvents any causality problems. Causality sequences of events can't be violated, but they can be initiated. This is exactly like the branching timelines model of time travel that also avoids causality violation.

Anyone receiving a reply to their initial message will receive that reply before they send their message, but that reply will come from another future. In fact, because the number of worldlines will be large. There won't simply be one reply, there will be a veritable torrent of them. The problem for the receiver is working out which message is the most useful. Their best approach would be to do a statistical analysis of all messages to work out what is the most probable set of circumstances that is the subject of the message.

If there is only one worldline, then the CPP will manifest itself as the Novikov self-consistency principleNovikov self-consistency principle. The universe will be deterministic. Even if you know the future you will still send the same message, anyway, the reply will come back, into your past, unchanged. Basically there are an infinite number of pathways to achieve the same set of events (always). Also, this model turns the interpretation of backwards in time processes on their head. This isn't causality violation it is now retrocausality.

We are accustomed to causality be about event A causes an event B to happen in its future. With retrocausality event D causes an event E to happen in its past. This starts getting quite confusing when both causal processes are working together in the universe. This usually is resolved as a form of determinism. Again this is Novikov self-consistency.

Finally there is the Antippa Corridor solution where superluminal travel and communications can only go in one direction. If you send a superluminal message to a ship they can't reply with a superluminal message. Any reply can only be send at lightspeed or slower than lightspeed velocities. In which case, you won't need to worry about causality violation of any kind.

The Antippa model would allow one way superlumninal messaging. The geometric shape where FTL messages can be send will be in the half-circle (actually it's a half-sphere in three dimensions) in the direction where tachyons can travel. The ship can only receive messages but cannot reply. You can only send messages to ships within ninety degrees of the tachyon unidirection. There will be an increasing difficulty in sending messages as the width of the angle of sending superluminal approaches ninety degrees.

(1) Almost no point, there might be communications systems set up so if the ship decelerated to relative rest with FTL communications station that could be always listening for messages. You could always accelerate in a vehicle to match the relativistic velocity of the ship you are trying to contact.


You must log in to answer this question.

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