I Is the Alcubierre Warp Drive possible?

[swampwiz]

This Quora post seems to say that this type of warp drive could allow faster-than-light travel, but somehow I think this is junk science:

https://www.quora.com/Will-humans-e...ple-light-years-away-to-even-the-nearest-star
(author is Robert Williscroft)

 

[Dale]

The Alcuiberre warp drive, and most other similar solutions require matter with a negative energy density, sometimes called exotic matter. No such matter has ever been detected and it is unlikely that such material would be stable if it did exist.

These solutions should be seen as interesting mathematical solutions that likely have no physical realization.

 

[m4r35n357]

This Quora post seems to say that this type of warp drive could allow faster-than-light travel, but somehow I think this is junk science:

Here is a video by the Author himself. See if you think he takes it seriously!

 

[pervect]

Here is a video by the Author himself. See if you think he takes it seriously!

An interesting video. I haven't had the time to watch the whole thing, but his remark that this one just one paper he did back in 1994 and that the majority of his work is in gravitational waves gives a a good indication to me of how seriously he takes the paper. The word "serious" is a bit ambiguous, but the message I get from what I did watch is that this one paper is not the focus of his work by any means, but I think he still stands by the logic and mathematics of this paper, or he wouldn't have agreed to present it.

 

[Dale]

I agree. His logic and math is correct. It is a valid solution to the EFE. I don’t think that he ever claimed that it was feasible to construct.

 

[bobob]

The Alcuiberre warp drive does not actually result in faster than light travel in it's local reference frame. What it does is to provide a region of spcetime which in which the distances in front and behind it are shortened and lengthened respectively while inside the bubble, the occupants remain at rest.

 

[DrStupid]

What it does is to provide a region of spcetime which in which the distances in front and behind it are shortened and lengthened respectively while inside the bubble

But the warp bubble must already exists in order to do that. Is it possible (in theory) to chose a random target and reach it faster than a light signal outside the bubble?

 

[bobob]

To answer that question would be to speculate on the possibilities of a device which itself is highly speculative and thought to most likely not be possible. Or, in other words, since the device itself depends on physics that might not be possible in this universe, there is no way to say what might or might not be possible if I assume something that is already probably incorrect.

 

[1977ub]

Whatever the technical issues, does not the existence of FTL not lead to causality violation? That's a good enough reason to think it's not possible.

 

[bobob]

There is no faster than light travel using this so called warp drive.

 

[PeterDonis]

Is it possible (in theory) to chose a random target and reach it faster than a light signal outside the bubble?

As I understand how the Alcubierre spacetime works, yes, this is possible, but in a curved spacetime with exotic matter in it, this does not mean anything is traveling "faster than light" locally. A light ray emitted inside the bubble will still outrun any timelike object inside the bubble. All it means is that curved spacetimes with exotic matter in them can have highly counterintuitive properties.

 

[1977ub]

As I understand how the Alcubierre spacetime works, yes, this is possible, but in a curved spacetime with exotic matter in it, this does not mean anything is traveling "faster than light" locally. A light ray emitted inside the bubble will still outrun any timelike object inside the bubble. All it means is that curved spacetimes with exotic matter in them can have highly counterintuitive properties.

But if it is outrunning light just outside of the bubble then it is FTL for all practical purposes

 

[PeterDonis]

if it is outrunning light just outside of the bubble then it is FTL for all practical purposes

The term "FTL" is misleading. I would say it's moving in a curved spacetime geometry with exotic matter in it that has highly counterintuitive properties. (And most physicists do not think exotic matter is possible, so this kind of spacetime geometry cannot actually exist.)

 

[Dale]

Whatever the technical issues, does not the existence of FTL not lead to causality violation? That's a good enough reason to think it's not possible.

The technical term is a closed timelike curve. It is not FTL anywhere, but it does have that causality issue.

 

[DrStupid]

As I understand how the Alcubierre spacetime works, yes, this is possible, but in a curved spacetime with exotic matter in it, this does not mean anything is traveling "faster than light" locally.

Of course there is nothing traveling faster than light locally. But the bubble propagates faster than light in the flat space around it. This implies all the nasty things of FTL travel, including violation of causality.

But that's not my point. I wonder if the warp drive could move the ship to a random destination faster than a light signal in flat space after the destination has been selected.

 

[PeterDonis]

the bubble propagates faster than light in the flat space around it.

Again, "faster than light" is misleading. It is a curved spacetime with exotic matter in it that has highly counterintuitive propereties.

This implies all the nasty things of FTL travel, including violation of causality.

No, there is no violation of causality; there is simply a curved spacetime with exotic matter in it.

 

[1977ub]

One such vehicle leaves Earth and goes to alpha centauri, arriving sooner than a photon emitted from Earth at the same moment. Also let's say you have another such vehicle leaving alpha centauri toward earth, also arriving sooner than a photon emitted at the launch event. As the two vehicles pass one another, interactions could occur which violate causality.

 

[PeterDonis]

One such vehicle leaves Earth and goes to alpha centauri, arriving sooner than a photon emitted from Earth at the same moment.

A photon emitted from Earth along a different path that doesn't cross the path of the warp bubble.

As the two vehicles pass one another, interactions could occur which violate causality.

Like what? Remember that "causality" in a curved spacetime is not the same as in flat spacetime. Spacetime with warp bubbles in it is not flat.

 

[1977ub]

A photon emitted from Earth along a different path that doesn't cross the path of the warp bubble.
Like what? Remember that "causality" in a curved spacetime is not the same as in flat spacetime. Spacetime with warp bubbles in it is not flat.

The bubble region is a tiny % of a sufficiently large flat-ish region of the universe, and so the standard FTL / causality issues arise.
https://en.wikipedia.org/wiki/Alcubierre_drive#Causality_violation_and_semiclassical_instability

 

[PeterDonis]

The bubble region is a tiny % of a sufficiently large flat-ish region of the universe

The spacetime is not flat. Waving your hands and saying "tiny % of a sufficiently large flat-ish region" doesn't change that.

so the standard FTL / causality issues arise

You're going to have to be more explicit: just saying "the standard FTL/causality issues" doesn't cut it. Give a specific scenario that you claim violates causality. Remember that the standard rigorous meaning of "violates causality" is that objects move outside the light cones, and that does not happen in a spacetime containing Alcubierre regions.

 

[PeterDonis]

the standard FTL / causality issues arise

If you're referring to the possible presence of closed timelike curves, "causality violation" is a misleading term to use for those. They certainly violate many common intuitions about causality, but that's not the same thing. There has been significant work, such as that associated with the Novikov Self-Consistency Principle [1], that shows that it is perfectly possible to have consistent models that include closed timelike curves. Whether those models can actually be realized is a separate question (but that's just the same question as whether an Alcubierre warp drive can actually be realized).

[1] https://en.wikipedia.org/wiki/Novikov_self-consistency_principle

 

[Dale]

Remember that the standard rigorous meaning of "violates causality" is that objects move outside the light cones, and that does not happen in a spacetime containing Alcubierre regions.

I think that closed timelike curves could also reasonably be described as “violating causality”. Those can occur in Alcubierre spacetimes.

 

[DrStupid]

Again, "faster than light" is misleading. It is a curved spacetime with exotic matter in it that has highly counterintuitive propereties.

I'm not sure if I understand you correctly. Are you saying there is no way to keep the reference light signal in flat spacetime and reach the target first?

 

[PAllen]

If you're referring to the possible presence of closed timelike curves, "causality violation" is a misleading term to use for those. They certainly violate many common intuitions about causality, but that's not the same thing. There has been significant work, such as that associated with the Novikov Self-Consistency Principle [1], that shows that it is perfectly possible to have consistent models that include closed timelike curves. Whether those models can actually be realized is a separate question (but that's just the same question as whether an Alcubierre warp drive can actually be realized).

[1] https://en.wikipedia.org/wiki/Novikov_self-consistency_principle

Even with Novikov, CTCs allow the possibility of e.g. a play or symphony that has no author. I think many, if not most, physicists would consider this a form of causality violation.

 

[George Jones]

I think that closed timelike curves could also reasonably be described as “violating causality”. Those can occur in Alcubierre spacetimes.

The Alcubierre's warp drive spacetime does have closed timelike curves. This spacetime can be modified to produce spacetimes that do have closed timelike curves, but loads of spacetimes, including Minkowski spacetime, can be modified to produce spactimes that have closed timelike curves.

 

[1977ub]

The question is - is this a technology which can be operated at will. That would be different from the issue of whether nature might generate something like this, outside of human intentions. If I have a friend 4 light years away, and one day I've got a black box I invented which can send him a letter in what he and I both agree is 3 years, then we are in causality violation territory. Say we've been sending chess moves back and forth at the speed of light then one day I have a black box (containing Alcubierre drive with warp bubble) and I put my chess move in it and send it to him in 3 years. Then he sends the object back to me in 3 years with his move.

 

[hmmm27]

The question is - is this a technology which can be operated at will.

Since it isn't a technology, and the mathematics that it is make no overt references to "will" or not . . .

 

[Heikki Tuuri]

If you are allowed to choose the metric freely, you can make a patch of space to travel faster than light. That is clear.

Then you can use the Einstein field equation to calculate the stress-energy tensor T which would produce such a metric.

The Alcubierre drive, of course, allows closed timelike loops and causality violations. It breaks the principles of special relativity since we would have a spaceship which travels faster than light. The energy of such an object in special relativity is imaginary. These contradictions strongly suggest that the stress-energy tensor T is impossible to realize.

The stress-energy tensor T for the drive is very weird: there should be something which makes space to contract in front of the rocket and stretch behind it. I think matter with a negative energy density does not accomplish this. Ordinary matter has stretched the radial dimension r around it, and negative energy matter has contracted r. But it is static - there is no dynamic stretching or contracting around such matter.

One may imagine that the rocket would move inside a solid rod where a lever mechanism keeps negative pressure in front of the rocket and positive pressure behind it. One problem is how a rocket can fly inside a solid rod? Maybe we should instead of a rocket use a neutrino which can fly trough solid rock.

The negative pressure has to be huge to win the positive gravity of the material in the rod. Such material does not exist in nature. Can it exist theoretically? I think such material would allow faster-than-light travel for the neutrino. Since faster-than-light travel involves the causality violations, that strongly suggests that the required material for the rod cannot exist, even in theory.

 

[PeterDonis]

Even with Novikov, CTCs allow the possibility of e.g. a play or symphony that has no author. I think many, if not most, physicists would consider this a form of causality violation.

But that is a different usage of the term "causality" than the standard one, which is that nothing moves outside the light cones. IMO it would be better to just say "closed timelike curves".

 

[PeterDonis]

It breaks the principles of special relativity since we would have a spaceship which travels faster than light.

First, SR is inapplicable since we are talking about a curved spacetime.

Second, as I've repeatedly said in this thread, nothing moves "faster than light" in this spacetime. Nothing moves outside the light cones.

The energy of such an object in special relativity is imaginary.

SR is inapplicable; see above. What makes the stuff that produces the Alcubierre warp bubble "exotic matter" is that its pressure is greater than its energy density in the rest frame of the bubble. There is no imaginary energy anywhere. (There are frames where the energy density is negative, but none in which it is imaginary.)

I think

Don't "think". Do the math and see. The properties of the Alcubierre spacetime are well understood.

One may imagine

Don't "imagine". Do the math.

The negative pressure has to be huge to win the positive gravity of the material in the rod. Such material does not exist in nature. Can it exist theoretically?

As multiple posters have already said in this thread, most physicists think the answer to this is no.

I think such material would allow faster-than-light travel for the neutrino.

Please review the PF rules on personal speculation.

 

[PAllen]

The Alcubierre's warp drive spacetime does have closed timelike curves. This spacetime can be modified to produce spacetimes that do have closed timelike curves, but loads of spacetimes, including Minkowski spacetime, can be modified to produce spactimes that have closed timelike curves.

All you need to do is link up two bubbles as in the tachyonic antitelephone to create a CTC. Long ago, @bcrowell posted a paper on this.

 

[Janus]

The question is - is this a technology which can be operated at will. That would be different from the issue of whether nature might generate something like this, outside of human intentions. If I have a friend 4 light years away, and one day I've got a black box I invented which can send him a letter in what he and I both agree is 3 years, then we are in causality violation territory. Say we've been sending chess moves back and forth at the speed of light then one day I have a black box (containing Alcubierre drive with warp bubble) and I put my chess move in it and send it to him in 3 years. Then he sends the object back to me in 3 years with his move.

No, that does not involve any causality violation. For a causality violation, you would need two separate inertial frames, in relative motion with respect to each other and sending FTL messages back and forth. This is the result of the simultaneity of relativity.
So let's say that you have two spaceships traveling at 0.5c relative to the Earth, and a string of space buoys at rest with respect to the Earth strung out along its path. Space ship 1 passed Earth when its both clocks read 0. 1 year later by his clock, when he is 1/2 light year from the Earth, as he measures the distance, he sends a instantaneous message (any FTL signal will produce similar results, it's just simpler to assume instant messages) back to to spaceship 2 just as it is passing the Earth. Both spaceship 2 and Earth will agree that the Earth clock reads 0.866 yrs as spaceship 2 passes. The message is transferred to the Earth and it then sends a reply instantaneously to a space buoy that is 0.433 ly away in the same direction as spaceship 1. In the Earth frame, this buoy is right next to spaceship 1 at that moment and the reply is transferred to spaceship 1. Assuming that the clocks on the buoy and Earth are synchronized in the Earth rest frame, the buoy clock reads 0.866 years at that moment. However, according both buoy and spaceship 1, spaceship 1's clock only reads 0.75 yrs, 1/4 of a year before he sent the original message. He gets his reply before sending the message.

I'm not sure how the Alcubierre drive would work in this scenario, as you would need to use it to deliver the messages back and forth.

 

[George Jones]

The Alcubierre drive, of course, allows closed timelike loops and causality violations.

Give an explicit, mathematical demonstration the Alcubierre warp drive spacetime contains closed timelike curves.

 

[George Jones]

All you need to do is link up two bubbles as in the tachyonic antitelephone to create a CTC. Long ago, @bcrowell posted a paper on this.

This is not the Alcubierre warp drive spacetime.

 

[Dale]

loads of spacetimes, including Minkowski spacetime, can be modified to produce spactimes that have closed timelike curves.

Well, Minkowski can only be modified to have a CTC with unusual topology, like a torus. The Alcuiberre spacetime can get a CTC with standard topology.

 

[Heikki Tuuri]

https://en.wikipedia.org/wiki/Faster-than-light
The Alcubierre drive allows superluminal communication, and according to Wikipedia, that is equivalent to time travel. Then we will have all the paradoxes of time travel. And that is in an asymptotic Minkowski space - we would not need any wormhole to make a causal loop. The Alcubierre drive is enough.

 

[PeterDonis]

according to Wikipedia

Wikipedia is not a valid source. You need a textbook or peer-reviewed paper.

 

[PAllen]

This is not the Alcubierre warp drive spacetime.

True, but if more than one drive in the universe is possible, it suggests CTCs would be possible.

Here is the paper:

http://exvacuo.free.fr/div/Sciences...tt - Warp drive and causality - prd950914.pdf

 

[DrStupid]

For a causality violation, you would need two separate inertial frames, in relative motion with respect to each other and sending FTL messages back and forth.

A single message could be sufficient. With two events A and B, where one of them is the cause of the other, causality is violated if the order of the events is frame dependent. In Minkowski space this would be the case if the interaction is faster than a light signal traveling in vacuum directly between A and B. However, in the current discussion this would be only helpful if the Alcubierre warp drive can be used in a way that keeps the spacetime between A and B flat. PeterDonis sounds like this is not possible.

 

[PeterDonis]

this would be only helpful if the Alcubierre warp drive can be used in a way that keeps the spacetime between A and B flat. PeterDonis sounds like this is not possible.

Of course it's not possible. Alcubierre spacetime is not flat. If an Alcubierre drive travels between A and B, then spacetime between A and B is not flat.

 

[Heikki Tuuri]

https://arxiv.org/abs/gr-qc/9810026
The paper by Matt Visser et al. suggests that the Alcubierre drive breaks the Null Energy Condition.

The paper is not peer-reviewed. The fact that if you can do unrestricted superluminal travel then you can make causal loops, is folklore that everybody seems to know, but it is hard to find a peer-reviewed reference. The math is easy and can be found from my own blog, year 2013.

If you restrict the superluminal travel, then you can prevent causal loops. One may, for example allow a superluminal trip on Earth as long as it takes you forward in the UTC time coordinate. Causal loops cannot happen because the UTC time always advances on every trip.

 

[DrStupid]

If an Alcubierre drive travels between A and B, then spacetime between A and B is not flat.

My question was not limited to this special case. The ship can also go a long way around, running the warp drive only far away from A and B and flying the rest with conventional engines.

 

[George Jones]

https://arxiv.org/abs/gr-qc/9810026
The paper by Matt Visser et al. suggests that the Alcubierre drive breaks the Null Energy Condition.

I am not disputing the fact that Alcubierre warp drive spacetime violates one or more energy conditions, but violation of an energy condition is not a (mathematically) sufficient condition for closed timelike curves.

The paper is not peer-reviewed. The fact that if you can do unrestricted superluminal travel then you can make causal loops, is folklore that everybody seems to know

What does "superluminal travel" mean? If it means "woldline outside a local lightcone", then I agree. Alcubierre warp drive spacetime, however, does not have this property. Folks (e.g., @bobob , @PeterDonis ) have noted this in this thread.

The math is easy and can be found from my own blog, year 2013.

This statement is not true.

http://meta-phys-thoughts.blogspot.com/2013/
This bolgpost most certainly does not demonstrate mathematically (i.e., using actual GR spacetimes) that, specifically, Alcubierre warp drive spacetime contains closed timelike curves.

 

[1977ub]

Of course it's not possible. Alcubierre spacetime is not flat. If an Alcubierre drive travels between A and B, then spacetime between A and B is not flat.

A small kernel of heavily warped spacetime in a large ocean of flat-ish spacetime such as between here and alpha centauri. This is why i described the vehicle to be within a black box. At the quantum level, spacetime may be heavily warped but that doesn't enable FTL communication or travel at the macroscopic scale.

Adherents of the drive accept that it will allow me to send messages to my friend 4 ly away in what he and I both agree to be 3 years. This is a causality violation, as agreed by Alcubierre and others.

 

[PeterDonis]

The ship can also go a long way around, running the warp drive only far away from A and B and flying the rest with conventional engines.

A and B are points in spacetime, not space. You are stipulating that a light ray not going through the warp bubble could not reach B from A. That means the warp bubble has to affect the spacetime geometry between those events.

 

[PeterDonis]

A small kernel of heavily warped spacetime in a large ocean of flat-ish spacetime such as between here and alpha centauri.

See my response to @DrStupid just now.

 

[PeterDonis]

The math is easy and can be found from my own blog

Your blog is not a valid reference. Please do not reference it again. If you do you will receive a warning.

 

[PeterDonis]

Adherents of the drive accept that it will allow me to send messages to my friend 4 ly away in what he and I both agree to be 3 years.

A lot of complexities that you are blithely ignoring are lurking underneath your simple-looking statement "in what he and I both agree to be 3 years". As has already been pointed out multiple times, nothing moves outside the light cones in Alcubierre spacetime. You cannot just wave your hands and ignore the spacetime curvature involved.

 

[DrStupid]

You are stipulating that a light ray not going through the warp bubble could not reach B from A.

No, I'm stipulating, that a light ray could reach B from A, not going throuth the warp bubble. There may also be light rays, reaching B from A, going through the warp bubble. But I'm not interested in them because I'm currently checking causality in the flat parts of the space-time.

 

[Ibix]

Alcubierre's 1994 paper notes that his spacetime is globally hyperbolic and therefore does not contain any causal loops.

Everett's paper (linked upthread) points out that Alcubierre's solution can be written as $\eta_{\mu\nu}+h_{\mu\nu}$ and that $h$ is very small except near the bubble. So you can have two warp bubbles traveling in opposite directions essentially as a linear superposition of two Alcubierre spacetimes - i.e. $\eta_{\mu\nu}+h_{\mu\nu}+h'_{\mu\nu}$ where the two $h$s are warp bubbles moving in $\pm x$ directions and offset by some distance in the $y$ direction so that they do not interfere with each other. Furthermore, he argues that there's no reason that the two warp bubbles have to share a notion of "at rest" and both can arrive at their destinations in arbitrarily short coordinate times for their rest frames. Thus he says you can build a tachyonic anti-telephone with two warp drives, although you can't do it with one.

I don't know whether or not Everett's analysis is correct. It seems plausible on the face of it, but if I understand correctly Alcubierre's paper assumes a spacetime in which CTC's are impossible before showing that the warp drive is a solution. So it's not completely clear to me that simply adding two warp bubbles can work quite as Everett claims - he seems to have ended up with a spacetime that isn't consistent with the one Alcubierre used.

I may be wrong - this is just my impression from reading the papers this morning.