[yellowapple]

> The wikipedia page has a worked out example. If you can go back and forth between 2 places FTL you can arrive before you left and interfere with your departure.

That example has multiple assumptions that seem to make said example only relevant to a very specific case, rather than to superluminal communication in general (least of all to approaches specifically designed to avoid those assumptions):

1. Alice and Bob are themselves moving at relativistic speeds relative to each other while communicating. That doesn't seem relevant for cases where they are not doing so - in particular when they are communicating across a region of spacetime crafted specifically such that they are stationary relative to each other.

2. Alice's signal to Bob and Bob's reply to Alice are both somehow themselves moving through normal spacetime at a speed greater than c - i.e. the messages are tachyonic (hence: tachyonic antitelephone). Again, when discussing mechanisms of warping spacetime such that no >c movement within any local frame of reference is actually necessary, bringing up tachyons doesn't seem especially relevant.

> You are more convinced by a metaphor than the actual theory of relativity. That can't be helped.

That can be helped, specifically by explaining why the metaphor is wrong, and where it actually contradicts general or special relativity.

[bitdizzy]

The name tachyonic antitelephone is a bit unfortunate because it need not involve tachyons. All it takes is the ability to send a message outside of your light cone. Along the path actually taken need not be the case, we are reasoning about the endpoints of the communication independent of the means.

In particular it does not matter whether you use a warp drive or not. The violation occurs relative to reference frames outside of the warp bubble where special relativity fully applies. Here, explicitly looks like a worked out example: http://exvacuo.free.fr/div/Sciences/Dossiers/Time/A%20E%20Ev...

You can create instances of FTL travel or communication that don't violate causality. But the ability to perform such FTL tricks implies the possibility of constructing a closed timelike curve so long as you're talking about a localized phenomenon like a warp drive or tachyon where special relativity holds outside of a negligible area.

If you're messing with the global geometry of spacetime then you can make very weird things happen and I would not be surprised if there were such a geometry that allows FTL communication in a limited way that doesnt allow CTCs. But that would not like our universe or a warp drive, a specifically localized situation.

[yellowapple]

So first, I'd like to say thanks for bearing with me here. I know it's probably frustrating when someone you're talking to just... ain't... getting... it...

...which is why I hate to say that I still ain't getting it, lol

> Along the path actually taken need not be the case, we are reasoning about the endpoints of the communication independent of the means.

So the means doesn't matter at all? It doesn't matter that as far as the endpoints are concerned they're at a fixed distance from one another? That seems hard to believe - not to mention contrary with the math presented in that antitelephone article, which seems to make a big and explicit deal about the relative movement between the two endpoints.

Even when factoring outside observers...

> The violation occurs relative to reference frames outside of the warp bubble where special relativity fully applies.

...how would this apply if the information in question is constrained to that warp bubble between its transmission and receipt? What would there be to observe? If we're talking about side effects of that communication (say, our Alcubierre messenger pigeon drops a feather back into "normal" space somehow), would the propagation of those side effects not just revert back to being subluminal? That is: from every local perspective, the pigeon is traveling subluminally, so this should still apply in the event the pigeon or any signal from it escapes its bubble and reenters "normal" spacetime, no?

> Here, explicitly looks like a worked out example: http://exvacuo.free.fr/div/Sciences/Dossiers/Time/A%20E%20Ev...

Thanks, this is useful.

...unfortunately, I don't know if it's really agreeing with you. Or me, for that matter.

Namely, it seems to admit that the metric as Alcubierre describes[1] would not produce a closed timelike curve; the paper instead describes ways to produce custom metrics separate from Alcubierre's which introduce the possibility of CTCs, in which case it seems like the answer would simply be to... just not do that, right? Indeed, the paper speculates that there might be other mechanisms that would prevent such a metric from being constructed (specifically naming Hawking's chronology protection conjecture).

That is:

> But the ability to perform such FTL tricks implies the possibility of constructing a closed timelike curve so long as you're talking about a localized phenomenon like a warp drive or tachyon where special relativity holds outside of a negligible area.

Only, from what I can gather, if these equations are indeed the only ones at play, which even that paper admits might not be the case.

----

[1]: https://arxiv.org/pdf/gr-qc/0009013v1.pdf

[bitdizzy]

So the thing that paper does is it assumes you can make warp bubbles in any reference frame. The original paper makes one warp bubble and this doesn't lead to anything paradoxical.

But if I can make two warp bubbles in two different frames, I can make a round trip that arrives before it starts.

So you need to violate special relativity in a global way to avoid these issues. You have to have a spacetime where only certain warp geometries are possible.

I apologize for being a bit impatient before, these issues are subtle. The only real way to get it is to bear with the math onesself.

[yellowapple]

> I apologize for being a bit impatient before, these issues are subtle.

No worries, and thanks again :)

> But if I can make two warp bubbles in two different frames, I can make a round trip that arrives before it starts.

Even for the metric as Alcubierre describes? Or for one that's modified per Everett? Now that I'm rereading the Everett paper, I'm not really sure where he's getting his "Lorentz boost"; if there's technically no actual "motion" (because everything's locally at rest and the warp bubble is outright expanding/contracting space around it), then I'm having a hard time figuring out what there would be to "boost", since the relevant Lorentz transformations should be no-ops if everything's locally at rest. Is Everett moving the ship itself at relativistic speeds within the bubble? Is an observer moving at relativistic speeds outside the bubble?

> So you need to violate special relativity in a global way to avoid these issues.

Which the universe kinda already does, no? The mechanism here (from what I understand) is the same as the one driving universal expansion (the difference being that there's no corresponding contraction in the universal case - right?). If that expansion were to be reversed somehow, would that, too, result in causality violations?

And further, doesn't special relativity already only hold in cases with low gravitational potential - so i.e. not in a gravitational field?

Sorry if these are kinda dumb questions.

[bitdizzy]

A lorentz boost is just a change of perspective. Let's say you have a bubble that is at rest with respect to your reference frame. Well, someone flying by your solar system has just as valid a reference frame as you and they see the bubble moving with respect to them. The relation between their point of view and yours is described by a lorentz boost. Same physical situation, different perspective.

Now, if special relativity holds in the large then there is no reason why they can't also construct a warp bubble. Relative to you and your bubble that bubble is in motion. Since this is allowed, we can construct a closed timelike curve with two judiciously chosen bubbles.

> Which the universe kinda already does, no?

The expansion of the universe does violate special relativity but not in a way that protects warp geometry from creating paradoxes. The kind of violation you would need would distinguish between frames of reference that SR says are equivalent.

> And further, doesn't special relativity already only hold in cases with low gravitational potential - so i.e. not in a gravitational field?

Special relativity holds well enough in most situations that the same argument applies even if you accounted for general relativity except near extreme situations like black holes

Warp drives are not permitted by any deviation from special relativity. You need specific contrived geometries of the entire universe that don't match up with what we know about it.

[SideburnsOfDoom]

> The expansion of the universe does violate special relativity

I am curious as to how?

If, we take as axiomatic that there is no such thing as "an object in motion", only "an object in motion with respect to another object", i.e. motion is not a property of an object, but a relation between two objects.

And that since it's 4.3 light-years to Alpha Centauri, therefor a cause (e.g. a radio wave) sent from Earth today cannot have an effect (e.g. an Astronomer writes a paper about it) at Alpha Centauri for 4.3 years, and vice versa. There is 4.3 years of "Absolute elsewhere" to get through first where there is no possible cause-and-effect relation between events. At the other side of our galaxy it's around 100 000 years. And further out, objects are not just far, but receding from us (or us from them, equivalently)

Assuming an unbounded and expanding universe, for very distant parts of the universe receding from us at lightspeed (or speeds faster than light?) and us from them, equivalently. So lightspeed signals from there never reach us, by definition? That part is is utterly unobservable, permanently Absolute elsewhere. The universe's observable edge is a slowed-down red-shift that trails off into the unobservable. In other words, the at no time in the future will those signals have an effect on Earth, or us on them. So, they can't have an effect in our future, let alone our past. No impact to causality at all?