Tuesday, October 14, 2014

Parallel universes and the arrow of time

In a previous post about quantum mechanics and parallel universes I ended with a puzzle:
All measurements are in principle reversible. Imagine that we could actually carry out this program of undoing the myriad entanglements that constitute your making a particular observation. What would be the subjective sensation, i.e. what would it "feel like" if this were done to you?
If you haven't read the previous post, please do before you read the rest of this one, otherwise this won't make any sense.  The point of this puzzle is not the answer, but the process by which one arrives at the answer.  Knowing the answer and understanding why it is the right answer are not the same thing, and in this case the latter is much more important than the former.

The intuitive answer to the question is that it would feel something like having time run backwards, but this is wrong.  The reason it is wrong goes to the very heart of the nature of reality.  To prime your mind to accept that the answer I am going to give is in fact correct I'm going to start by giving you the answer to what seems like it should be a completely different problem but is in fact almost exactly the same problem.  It is a classic problem first contemplated by Albert Einstein: what would it feel like to travel at the speed of light?

The intuitive answer to this question is that it would feel like a thrill ride, like zipping through the universe really, really fast.  Pluto is about four light-hours away, so if you headed towards it at the speed of light you'd get there four hours later, right?  How could it possibly be otherwise?

The unintuitive but undeniable fact of the matter is that the speed of light is the same in all reference frames.  This was shown experimentally in 1887 by the Michelson-Morley experiment, but it was actually predicted by James Clerk Maxwell 23 years earlier, in 1864.  Maxwell's equations for electromagnetic fields predicted the existence of electromagnetic waves that propagated at the (already well known) speed of light, but the equations made this prediction without any mention of a frame of reference.  The significance of this was not understood except in retrospect: it was forty years before Einstein first took the mathematical prediction and the experimental verification seriously and derived the theory of relativity, which can be summed up in the following pithy slogan: the speed of light is the universal reference.  Everything is always moving at the speed of light through space-time.  When you move faster through space you move slower through time.  When you move through space at the speed of light, time stops.  So what it would "feel like" to travel to Pluto at the speed of light is not that it would take four hours, but that you would get there instantaneously.  In other words, you would arrive at the same time that you left.

Now, an interesting thing happens if you arrive at the same time you left, and that is that you can no longer distinguish between leaving and arriving.  A trip from earth to pluto at the speed of light is indistinguishable (to the traveller) from a trip from pluto to earth at the speed of light.  Both consist simply of being at earth and pluto (and everywhere in between) at the same time (in the traveler's reference frame).  How do we reconcile this with the fact that an observer back on earth can easily tell the difference between a beam of light traveling in one direction rather than the other?

Here's a clue: the solutions of Maxwell's equations that predict electromagnetic waves are time-symmetric, that is, they predict the existence not only of waves traveling at the speed of light c, but they also predict waves traveling at speed minus c, that is, waves that move backwards in time.  These solutions are usually discarded out of hand as being "unphysical", but if Einstein teaches us anything it is that discarding mathematical results just because they don't "feel right" can blind us to deep truths.  (And indeed, if we take seriously the idea that there are electromagnetic waves moving backwards in time what we end up with is a completely self-consistent theory called the transactional interpretation of quantum mechanics.)

So what happens in QM if we really take the math seriously?  Well, we end up with a deeply unintuitive but nonetheless self-consistent description of the universe (except near a black hole -- physics is still working on that).  The math tells us that our classical reality is merely an approximation of the underlying metaphysical truth, just as Galilean relativity (where space and time are separate things) is.  The reason we think particles exist is not because they really exist, but because when you slice-and-dice the wavefunction in a certain way you end up with something that acts like (but isn't really) a bunch of classical entities all of which agree on a set of measurements (i.e. are in classical correlation with each other).

But, like Maxwell's equations, the Schroedinger equation is time-symmetric.  You can run it backwards as well as forwards.  And if you throw relativity into the mix then space and time are equivalent, and you can't distinguish forwards versus backwards any more than you can left versus right or up versus down.  In other words, the wave function is a static four-dimensional "thing" (for want of a more general noun) out of which space, time, and classical reality "emerge" when you slice it up in certain ways.  But there is nothing in the math that says that the slice that we are living in (what we call classical reality) is in any way special.  You can slice up the wave function in a different way and get a different universe.  All of this seems to be very much at odds with the (apparently) undeniable truth that our universe is special (at least to us), and that time only moves in one direction.

But again we must be very careful about trusting our intuitions, which have been proven wrong time and time again (the pun being a quantum superposition of intended and not :-)  The math says time doesn't "flow", just as it says particles aren't real.  So why does it appear to flow?

Again, we will approach the issue of time obliquely by drawing an analogy with particles.  Why do we think particles exist?  Because we can measure them.  But QM tells us that the results of our measurements indicate the presence of particles despite the fact that there are in fact no particles (or, if you're a multiple-worlder, that there must be other universes).  Let us go through the exact same mental exercise with respect to time: why do we think that time moves in one direction?  Because there is this manifest asymmetry: we can remember the past, but we can't remember the future.  And it's not just us: the universe "remembers" the past but not the future, which is to say, the universe contains information about the past (in books, for example) but not about the future.

There is another way to describe this asymmetry: we can "travel" forwards through time but we can't travel backwards.  Except that we actually can (in principle) "travel backwards in time" by "rewinding the universe" i.e. undoing all the entanglements that led to the present situation.  So let us go back and consider the opening puzzle: what would it feel like to "travel backwards in time" this way, to have yourself "reversed"?  Picture the scene: you would have to go into a sealed chamber of the sort that houses Shroedinger's cats.  Once safely ensconced inside, you would make some kind of measurement.  It doesn't really matter what kind, but let's suppose you made a quantum measurement just to keep things simple.  So there is some particle in the box with you, and you arrange for that particle to become entangled with some measurement apparatus and thence with your brain.  Your brain is now, by virtue of having become entangled with the particle in the having-observed-the-particle state.  You can remember making the measurement.  You know what the result of the measurement was.  All of the atoms in your brain are now acting together to make you appear like a classical system in classical correlation with itself and the measuring device.  This is what makes you think you know the state of the particle (even though in reality there is no particle).

Now we turn on the magic reverse-o-matic ray.  One by one, all of the entanglements that led you to be in the having-observed-the-particle state are undone.  This involves reversing the entanglements in your brain, the measuring device, and everything else inside the box (including the magic revers-o-matic ray, but since this is a thought experiment we can ignore the obvious difficulties that entails).  When this process is over, what mental state are you in?  Well, you are in exactly the same mental state that you were in before the whole process began.  You have no memory of having measured the particle.  You would have no memory of having the reverse-o-matic ray turned on.  And since completing the erasure involved undoing all of the entanglements in the box resulting from the initial measurement, you would not be able to find any evidence anywhere inside the box that any of this had happened.

In other words, the subjective sensation of undergoing a macroscopic quantum erasure is exactly the same as the subjective sensation of having nothing happen to you at all!  In fact, the only evidence that you would have that you had in fact undergone a quantum erasure is that when you emerged from the box after the experiment, you would find that more time had elapsed there than could be accounted for by what you experienced.  (Note the similarity to relativistic time-dilation.  Exercise: what would happen if you took a clock inside the box with you?)

So this is an unsatisfying sort of time-travel because it is experimentally indistinguishable from the normal state of affairs.  What we really mean when we fantasize about time travel is moving into the past with the information contained in our brains and bodies intact.  In other words, we're thinking about transporting information about the present into the past, at which point it becomes information about the future.  We think this might be possible because we think we're classical entities, coherent material things that move volitionally through space but are swept inexorably through time by some sort of "flow", and if we could just figure out how to "move through time" the way we move through space that we might be able to swim against the tide.

But we are not classical entities.

To understand the truth you need to adopt a completely different mindset about what you are.  Again, I will approach this obliquely starting with a more familiar concept: you think you're a human being, a physical entity with some coherent identity that remains intact through changes like growing older and learning new things.  (It is this abstract identity that we imagine moving into the past when we fantasize about time travel.)  What does this identity consist of?  It can't be the atoms in your body because those are constantly being swapped out for new atoms.  Maybe it is the arrangement of those atoms, at least at some high level of abstraction.  But the arrangement of your atoms today is radically different from what it was the day you were born.  In what sense, then, are you the "same person"?

The reason we consider ourselves to be in some sense the "same person" throughout our lives is because there is a continuous sequence of "yous" that lead from cradle to grave.  You today may be radically different from when you were born, but in between there is a smooth transition from one to the other.  At any instant in time, the you of that instant is very similar to the you one second previously, and very very similar to the you one Planck time earlier.

Imagine for a moment that you could take a God's-eye view of the universe and see all four dimensions at once.  Consider two different instances of "you" at two different times.  Is there any way you could tell which is the "earlier" you and which is the "later" you?  If the two times are far apart then you could look at which you appears older, but what if the times were just, say, a minute apart?  Well, you could still tell (if you were God) by examining your mental state: the later you would contain information that the earlier you didn't, namely, memories of the events that transpired during the intervening minute.  In fact, if you were presented with an unordered set of all the you's that have ever existed you could easily reassemble them in their proper order simply by looking at which ones contain information about which other ones.  If You-A remembers You-B then You-A must come after You-B.  It can never be the case that You-A remembers You-B and You-B remembers You-A because then one of those you's would be "remembering" the future, and that's not possible.

But the crucial point is not that it is impossible but why it is impossible.  The reason it is impossible is not that time only "flows" one way (that explanation would beg the question).  The reason it is impossible is that memories are entanglements just as measurements are.  In fact, memories are measurements, because every memory is a memory of something.  So it is not that a memory can't be reversed (it can), it is that in order to reverse a memory you have to reverse all the entanglements that comprise that memory.

Now, here's the killer question: given that the observable result of undergoing quantum erasure is indistinguishable from having nothing in particular happen to you at all, how can you be sure that at some point in your life it hasn't happened to you?

The answer is: you can't be sure it hasn't happened to you!  It is possible that you (and the rest of the universe) have undergone quantum erasure at some point.  In fact, it's possible that it happens regularly, that the entire universe is constantly being rolled back and replayed from different points in "time".  In fact, the universe is chock-full of little quantum "isolation boxes" where this happens constantly!  These are called "vacuum fluctuations" or "spontaneous pair generation", where a particle and an anti-particle just materialize out of nothing and almost immediately annihilate each other.  The members of a spontaneously generated pair are entangled.  The annihilation process "undoes" the entanglement and returns the universe to its previous state.

The punch line is this: the statement that we can't time-travel into the past is exactly the same as the statement that we can only remember the past.  It is not the case that one causes the other, it is that the two things are logically equivalent.    Your perception of "traveling through time" emerges from your mental states and not the other way around.  You feel like you are "traveling forward through time" because your mental states have this natural order to them.  You can remember the past and not the future because whatever you remember is your past.  The laws of quantum mechanics (and entanglement in particular) insure that what any given instance of you remembers appears to be a continuous and coherent sequence that behaves according to regularities that we call the laws of physics.  But in fact you do not travel through time, because at root there is no you, and there is no time.  There is only the wavefunction, from which you emerge as an approximation.

25 comments:

ErnestO said...

First, I love your blog. :-)

I really enjoyed what you have written here, and I think I follow what you are saying, but I'm a bit confused. I'm very much an amateur with this stuff, and am not sure how to best express these things.

Your focus on remembering the past is confusing me a bit, since any given person will forget many things from the past, or even everything if their brain is damaged. So I am assuming you are using this as a convenient point of focus to get your (awesome) idea across.

So the image I got from reading this is that all of space time is like a tapestry woven from wave functions, and I am a part of that, and there is no way to just pull me out from it and patch me into a different part of it. Actually tapestry isn't really the best way to describe what I am thinking, as someone could imagine pulling part of that out and patching it at another place in the tapestry. Perhaps just saying there is a solid 4-dimensional block of space time and there is no way to pull part of it out. Not that it's impossible because it's so difficult, but that it's categorically undoable. Given our reality, it makes no sense. Is that accurate to what you have described?

Don Geddis said...

Nice connection between memory, and the arrow of time. Maybe we travel backwards in time, regularly! LOL.

I've heard it before, but I still love this line: "Everything is always moving at the speed of light through space-time." And the followup: "When you move faster through space you move slower through time." I think that clearly distinguishes people who "get it", from people who don't. Most laypeople would react to those sentences with "WTF?". Whereas people who understand Relativity would say, "of course, isn't it obvious?" There's only one speed in the (4D space-time) universe; all you get to do is allocate some fraction of it towards moving through time, and the remaining fraction towards moving through space. (Or, even better: divide into four fractions, for each of the four dimensions.)

Still, ErnestO's question raises the concept of wormhole-like transitions between different parts of the 4D space-time cube. What you've talked about is ordinary transitions between adjacent parts of the cube. But (General) Relativity appears to allow (at least, in principle) transitions between non-adjacent parts of the cube. Which might allow "you" to go back in time, with your memories intact. (Or to jump between points in space, "faster" than light.) Unclear how this interacts with logical causality, of course.

Luke said...

> But in fact you do not travel through time, because at root there is no you, and there is no time. There is only the wavefunction, from which you emerge as an approximation.

Is this necessarily true, or dependent on particular metaphysical assumptions? I'm reminded of the growing block universe as opposed to the [static] block universe, as well as this bit from Back From the Future:

>> It bothered Aharonov as well. “I asked, what does God gain by playing dice?” he says. Aharonov accepted that a particle’s past does not contain enough information to fully predict its fate, but he wondered, if the information is not in its past, where could it be? After all, something must regulate the particle’s behavior. His answer—which seems inspired and insane in equal measure—was that we cannot perceive the information that controls the particle’s present behavior because it does not yet exist. (2)

If we were to run with the holographic principle, it would seem that Aharonov (winner of the 1998 Wolf Prize in Physics) has shown some evidence that new bits appear as time goes forward. Now, I'm not opposed to there being a [static] block universe explanation of this, but I would want to pit it against the growing block universe, and ask what would make one prefer one over the other.

What I'm not sure of is whether the Back From the Future article depends on Hardy's paradox. It definitely involves Interpretations of quantum mechanics § Time-symmetric theories, which include this juicy bit:

>> The collapse of the wavefunction is therefore not a physical change to the system, just a change in our knowledge of it due to the second measurement. Similarly, they explain entanglement as not being a true physical state but just an illusion created by ignoring retrocausality. The point where two particles appear to "become entangled" is simply a point where each particle is being influenced by events that occur to the other particle in the future.

Thoughts? BTW, Back From the Future is deeply dependent on weak measurement; your blog does not appear to contain that term, so I'm not sure you know about it. I was introduced to it in Caltech's Ph12b by John Preskill in the form of the Elitzur–Vaidman bomb tester. I could be wrong, but I think a possible usage of that interaction-free measurement (not to be confused with weak measurement, but perhaps there is a connection?) is to measure biological systems with something like gamma rays, but without 'exploding' the molecules being measured very often. Well, that would be an application a long ways down the road, but the idea seems legit to me on surface inspection.

Oh, there's Can a Future Choice Affect a Past Measurement's Outcome?, which will probably help as well.

Ron said...

@ErnestO: Thanks for the kind words!

> Your focus on remembering the past is confusing me a bit

Yeah, I was afraid of that. I was going to talk about this explicitly but I had already spent three days composing that post and it was already getting too long.

When I talk about "remembering" what I'm not talking about *consciously* remembering. I'm talking about remembering in the technical information-theoretical sense. We can talk about a system A containing information about a system B if the state of A is correlated with the state of B. The canonical example is a flipped coin and a sensor that sees whether the coin is heads or tails. If the sensor is in the "I-see-heads" state whenever the coin is actually heads-up then we say the sensor contains information about the coin (1 bit of information in this case).

"Remembering" is exactly the same thing, except that the two correlated states are separated in time. So if the coin is heads-up at time T0 and the sensor is in the I-saw-heads-at-time-T0 state at time T1 then we say that the sensor "remembers" information about the coin.

Of course, we could just as easily say that the coin "remembers" information about the sensor! There is no way to define an arrow of time for the sensor and the coin. From the perspective of the laws of physics they are exactly symmetric. (The laws of physics don't know the difference between a "coin" and a "sensor" -- they are both just physical systems.)

To draw an arrow of time you have to arbitrarily decide what is sensor and what is sensed. The "natural" way for *you* to do that is to decide that *you* are the sensor, and everything else is sensed. When you do that, the information in your brain -- whether or not you are conscious of it -- defines an arrow of time (for you).

@Don:

> I've heard it before, but I still love this line:

Thanks! I am the author of that line (at least AFAIK :-)

> wormhole-like transitions

Yes, these are theoretically possible, but only in the presence of "exotic matter" with negative mass, which as far as we know does not actually exist.

@Luke:

> Is this necessarily true, or dependent on particular metaphysical assumptions?

Well, it's a story that is consistent with the current state of scientific understanding of our universe. Furthermore, it's the *only* story (AFAIK, and modulo minor variations on the theme, like multiple-worlds) that has this property.

> growing block universe as opposed to the [static] block universe

The classical universe is a growing-block universe. The wave-function (or the multiverse -- same thing) is static.

> Aharonov

I don't have time to dig into this just now, but I will just point out that if he really has done what he claims to have done then anyone could use this technique to send information about the stock market into the past, and use this information to become the richest person on the planet in a matter of days. That no one has done this is strong evidence that Aharonov's technique doesn't do what the popular accounts say that it does.

Luke said...

@Ron:

> Well, it's a story that is consistent with the current state of scientific understanding of our universe. Furthermore, it's the *only* story (AFAIK, and modulo minor variations on the theme, like multiple-worlds) that has this property.

Wait, this sounds you're saying that we now know that the vast majority of interpretations at Interpretations of quantum mechanics are false; is that the case? Or are you instead claiming that the vast number of them share the property under discussion? I'm reminded of Sean Carroll's The Most Embarrassing Graph in Modern Physics, in which 18% of the participants (physicists, philosophers, and mathematicians) held to the Everett interpretation. 24% said "Information-based/information-theoretical".

> The classical universe is a growing-block universe. The wave-function (or the multiverse -- same thing) is static.

What would falsify the "static" claim?

> I don't have time to dig into this just now, but I will just point out that if he really has done what he claims to have done then anyone could use this technique to send information about the stock market into the past, and use this information to become the richest person on the planet in a matter of days. That no one has done this is strong evidence that Aharonov's technique doesn't do what the popular accounts say that it does.

Does his discovery necessarily lead to what you say? From my reading, the only way that future actions could be measured in the past is by looking at a very specific set of weak measurements which were being performed for a while in the past. The very concept of weak measurement was "first developed by Yakir Aharonov, David Albert and Lev Vaidman, published in 1988". And so, it could easily be the case that even if the science can be used to do what you describe, there have been no chances to do it yet.

Ron said...

> Wait, this sounds you're saying that we now know that the vast majority of interpretations at Interpretations of quantum mechanics are false; is that the case?

No. Most of them are *equivalent*, different ways of saying the same thing. (Copenhagen is demonstrably false, however.)

> What would falsify the "static" claim?

You'd have to falsify quantum mechanics (and probably GR too). It's not out of the question that this will happen. We still don't have a theory of quantum gravity, and so it might be the case that when we do get one it will require as much of a conceptual leap as QM and GR themselves required. That sort of thing is hard to predict :-)

> Does his discovery necessarily lead to what you say?

Hard to say because I don't really understand what "his discovery" is. But *any* retrocausality could be used to send information into the past (that's what retrocausality *means*) and so could be used to gain an edge in the stock market. Given the incentives, I'm pretty confident that if it were possible someone would have done it by now.

Luke said...

> No. Most of them are *equivalent*, different ways of saying the same thing. (Copenhagen is demonstrably false, however.)

What experiment falsified Copenhagen, or what contradiction was found within it? Copenhagen interpretation § Acceptance among physicists has:

>> Throughout much of the twentieth century the Copenhagen interpretation had overwhelming acceptance among physicists. Although astrophysicist and science writer John Gribbin described it as having fallen from primacy after the 1980s,[21] according to a poll conducted at a quantum mechanics conference in 1997,[22] the Copenhagen interpretation remained the most widely accepted specific interpretation of quantum mechanics among physicists. In more recent polls conducted at various quantum mechanics conferences, varying results have been found.[23][24][25]

> You'd have to falsify quantum mechanics (and probably GR too).

Would you be a bit more specific? For example, does this critically depend on all time-evolution of quantum state being, as far as we can tell, completely unitary? I'm aware of some attempts to formulate nonlinear time-evolution which evolves different mixtures of pure states in different ways; see for example On (Non)Linear Quantum Mechanics, p2. So far though, it makes the same [experimental?] predictions as linear QM.

I'm reminded again of Hardy's paradox, where an experiment got the equivalent of (I think!) quantum fluctuations to not self-annihilate 1/16 of the time.

> But *any* retrocausality could be used to send information into the past (that's what retrocausality *means*) and so could be used to gain an edge in the stock market. Given the incentives, I'm pretty confident that if it were possible someone would have done it by now.

What if the technology to receive simply isn't available yet?

Furthermore, and I know this is getting a bit conspiracy theory-ish, wouldn't it be in the "cheater's" best interest to keep his/her interference extremely hard to detect? Being the richest person in the world isn't necessarily the best goal; having the most power seems to be the best goal, and power can be deployed in a great variety of ways. It could even be the case that having few if any people knowing that you have great power is important. If you're too obvious, doesn't that paint cross-hairs on your chest?

Ron said...

> What experiment falsified Copenhagen, or what contradiction was found within it?

If Copenhagen is true (and specifically, if wave-function collapse is a real, physical phenomenon) then this can be used to produce faster-than-light communication. See http://www.flownet.com/ron/QM.pdf

> Would you be a bit more specific? For example, does this critically depend on all time-evolution of quantum state being, as far as we can tell, completely unitary?

I don't know. We're at the hairy edge of my understanding here. But if you put a gun to my head and forced me to guess I would say, yes, it does depend on unitarity.

> What if the technology to receive simply isn't available yet?

The article you pointed to claimed that the effect, whatever it is, has been demonstrated experimentally.

> Furthermore, and I know this is getting a bit conspiracy theory-ish, wouldn't it be in the "cheater's" best interest to keep his/her interference extremely hard to detect?

What interference, and what cheating? This wouldn't be cheating, and it wouldn't be interfering. It would simply be making use of a competitive advantage. There would be nothing illegal or even unethical about it. However, time travel is only a competitive advantage if you're the only one who has it. If everyone can predict the future then there's no competitive advantage any more. So the advantage accrues only to the first person (or first few people) to make it work, and only for a narrow window of opportunity. So the only way to win would be to accumulate as much money as you could as fast as possible before others figured out what you were up to and built their own time machines. Discretion is not the winning strategy.

Luke said...

> If Copenhagen is true (and specifically, if wave-function collapse is a real, physical phenomenon) then this can be used to produce faster-than-light communication. See http://www.flownet.com/ron/QM.pdf

Do we know, for sure, that FTL communication is impossible? That seems rather a big guess that lots of physicists have confidence in, but which hasn't been tested extensively. In particular, I predict we'll open up a lot of very interesting doors with weak measurement and interaction-free measurement. Weak measurement especially seems like a "non-classical" way to do measurement.

What we will accomplish with entanglement alone has only started to manifest; see Quantum Microscope Uses Spooky Entangled Photons To See Better, for example.

> The article you pointed to claimed that the effect, whatever it is, has been demonstrated experimentally.

Surely you know that showing something in the laboratory can be a far cry from technological implementation?

> What interference, and what cheating?

I see you have rejected my attempt to inject certain meaning into those words. :-p Most people, I should think, would view Biff Tannen's time-traveling in Back to the Future Part II as 'cheating'. :-p

> There would be nothing illegal or even unethical about it.

This is not at all clear! Star Trek Voyager explicitly lays out time-travel regulations; Star Trek Enterprise does even moreso. Do you think people wouldn't start drafting laws if they saw influencing the past as a distinct possibility? As to what constitutes 'ethical'—well that depends on whether you think there is a Platonic Form that is 'Ethical', whether 'ethical' just means something like the current majority opinion of society, or something else. :-)

> So the only way to win would be to accumulate as much money as you could as fast as possible before others figured out what you were up to and built their own time machines.

I disagree. Money is merely one vehicle of power. Power is much more important than money. Power is real; money is fictional.

Ron said...

> Do we know, for sure, that FTL communication is impossible?

We don't know anything for sure. But It would falsify relativity, so we know it as surely as we know anything.

> Back to the Future Part II
> Star Trek Voyager
> Star Trek Enterprise

If you are seriously suggesting that science fiction is a reliable guide to ethics then you need to read this:

http://www.salon.com/1999/06/15/brin_main/

At the risk of stating the obvious, the problem with using fiction as a source of reliable information about, well, anything is that in fiction you're permitted (indeed expected) to just make shit up. (Take this to an extreme and you end up with Scientology.)

Luke said...

> We don't know anything for sure. But It would falsify relativity, so we know it as surely as we know anything.

Meh, I much prefer to say that relativity might be falsifiable, but that would involve moving to quite a different regime than where we seem to spend most of our time. Too many 'laws' have been found to be only approximations. Why expect this one to be any different? That being said, one must be incredibly intelligent in trying to 'break' such laws; trying randomly is virtually guaranteed to fail. It is much better to probe the 'edges' of current knowledge, which is precisely what we are doing.

> If you are seriously suggesting that science fiction is a reliable guide to ethics then you need to read this:

All I meant to suggest is that it is reasonable to assume that actual time travel would be subjected to legal and ethical reasoning. You are the one who said:

> > > There would be nothing illegal or even unethical about [time travel].

My response was not that all time travel would be outlawed, but merely that restrictions would likely be placed on it. Do you really disagree with this? If you don't, then certain time traveling would indeed qualify as 'cheating'. Precisely which, of course, would be a matter of debate. I am aware of at least one philosopher at USC who specializes in talking about consistency in time travel; whether he/she works with ethics and jurisprudence related to time travel, I do not know.

> At the risk of stating the obvious, the problem with using fiction as a source of reliable information about, well, anything is that in fiction you're permitted (indeed expected) to just make shit up. (Take this to an extreme and you end up with Scientology.)

While I roughly agree, I will assert that there is more- and less-believable fiction, and therefore that fiction can communicate useful information and/or usefully explore concepts. It is a way of playing with future possibilities. It is, of course, full of possible errors. That's simply a property of the imagination. But here, you seem to be summarily dismissing any and all fiction as being useful in any way when discussing issues like these—is that the case, or did you just mean to sweep away my particular use of fiction?

Ron said...

> Too many 'laws' have been found to be only approximations. Why expect this one to be any different?

I don't. That doesn't change the fact that as far as we know right now FTL is impossible.

> restrictions would likely be placed on it. Do you really disagree with this?

No. But "would be placed" is not the same as "have been placed."

> fiction can communicate useful information and/or usefully explore concepts

Of course it can. But what it can't do is be a reliable guide to the *truth*. Star Trek says one thing, Atlas Shrugged says another. How do we decide?

> you seem to be summarily dismissing any and all fiction as being useful in any way

Not at all. Fiction is useful in all sorts of ways. I'm only dismissing it as a reliable arbiter of truth. Truth is not the only thing that has value in this world.

Luke said...

> I don't. That doesn't change the fact that as far as we know right now FTL is impossible.

I simply prefer to say: "We don't yet know how FTL, or FTLI, would work." It is very close to what you said, but also different. Also, for fun: Alcubierre drive. Cheating for the win!

> No. But "would be placed" is not the same as "have been placed."

Well, your original statement contained 'would':

> There would be nothing illegal or even unethical about it.

So if we're allowed to talk counterfactuals, we've gotta be consistent about it. Sorry if I'm being pedantic, but it doesn't seem like I'm being any more pedantic than you are. :-/

> Of course it can. But what it can't do is be a reliable guide to the *truth*. Star Trek says one thing, Atlas Shrugged says another. How do we decide?

I'm a little annoyed that you think my view was so strongly of the form: "a reliable guide to the *truth*"—if indeed you meant to impute that view to me. I very explicitly stayed away from using 'reliable' or any word like it. All I said was 'useful'. That, of course, requires sifting the wheat from the chaff.

Now, how do we decide? Well, we've gotta do things like iron out intersubjective agreements on what terms like "human thriving" mean, and work from there. Atlas Shrugged was interesting in the respect that it advocated rampant individualism on the one hand, but almost violent 'love', on the other. I think studying the two as perhaps defining one another would be an interesting project. Sort of like how particles and fields seem to define each other.

> Not at all. Fiction is useful in all sorts of ways. I'm only dismissing it as a reliable arbiter of truth.

Then let me rephrase: you seem to have jumped to the conclusion that what I originally said (search for "Star Trek Voyager") was stupid/false/ignorant. Can you now see how there was an entirely non-stupid/false/ignorant way to interpret what I said? See, when I read what others say, I do my very hardest to pick the best possible interpretation I can, the one that makes the other person out to be thoughtful, intelligent, and desirous of good results. Conversation moves much faster that way, in my experience.

Ron said...

> Well, your original statement contained 'would':

English is a fluid language and the meanings of words depend on their context. My statement:

"This wouldn't be cheating, and it wouldn't be interfering."

Your statement:

"restrictions would likely be placed on it"

My statement is an assessment, yours is a prediction. My statement means, "It would not be cheating because it would be legal." Your statement means, "It would most likely stop being legal once society realized the consequences of having it remain legal." These are not contradictory.

If society passed pre-emptive laws against time travel then my statement would be false. (There's that pesky "would" again.) But I think that 1) that is unlikely to happen and 2) we're talking about a technology that supposedly already exists today, so we have to discuss in the context of the laws that exist today. And today there is no law against time travel.

> you seem to have jumped to the conclusion that what I originally said (search for "Star Trek Voyager") was stupid/false/ignorant

I was making a claim about the law and ethics. To phrase this as diplomatically as I can, I don't see how fiction can possibly inform a dispute about whether or not time travel is or is not in fact illegal at the present time. And if you intended your fictional citations to inform a dispute about ethics rather than a dispute about the current state of the law, well, you didn't exactly make that clear (go back and read what you wrote).

Luke said...

> And today there is no law against time travel.

Fine, but there remains my point that we are far from being able to get reliable messages via the method pioneered by Aharonov, if indeed his method works. And so, your original scenario that you claim would not be 'cheating' perhaps cannot yet happen. The closer it gets to realistically happening, I claim the closer we get to instituting laws. So... I'm not sure I'm so wrong, after all!

> To phrase this as diplomatically as I can, I don't see how fiction can possibly inform a dispute about whether or not time travel is or is not in fact illegal at the present time.

I was operating on a premise you were not: that it takes time to develop this tech, and the original phases cannot be used for the purpose you outlined.

> And if you intended your fictional citations to inform a dispute about ethics rather than a dispute about the current state of the law, well, you didn't exactly make that clear (go back and read what you wrote).

Well, you did not provide any input on what I said:

>*> As to what constitutes 'ethical'—well that depends on whether you think there is a Platonic Form that is 'Ethical', whether 'ethical' just means something like the current majority opinion of society, or something else. :-)

One could say that law is made to ever-better approximate the ontic Form of 'Ethical'. Under that view (and views like it), the 'unethical' in what you say could be utterly false, now:

>*> There would be nothing illegal or even unethical about it.

Meta-ethics is complex stuff. There are so many possible interpretations, so many ways to get seemingly the same result. Keeping it all in my head is hard, so I just do my best.

Publius said...

Homer Simpson on Time Travel Risks

Ron said...

We're safe. From the abstract of http://arxiv.org/abs/1206.6224:

"Causal loops are avoided by this anticipation remaining encrypted until the final outcomes enable to decipher it."

<irony>Whew.</irony>

And since you asked:

> As to what constitutes 'ethical'—well that depends on whether you think there is a Platonic Form that is 'Ethical', whether 'ethical' just means something like the current majority opinion of society, or something else. :-)

Yes, of course it does.

coby said...

Hmmm...random question that occurs to me while thinking of the flow of time and the universe and the symmetry of things: must there be only one possible past for any given moment?

Just as the current instant leads to one of multiple and exponentially increasing numbers of possible futures, could not this current configuration of the universe have been arrived at via multiple pasts?

Perhaps your reverse-o-matic ray could send you back on one of many pasts, no guarantee it will be the one you had experienced when you pressed the button!

Ron said...

> must there be only one possible past for any given moment?

Yes. That is what is meant by "classical correlation."

> could not this current configuration of the universe have been arrived at via multiple pasts?

No.

> Perhaps your reverse-o-matic ray could send you back on one of many pasts

Not by reversing entanglements. That can only send you back from whence (pun intended! :-) you came.

Luke said...

> Yes. That is what is meant by "classical correlation."

According to Luboš Motl on "Weak measurement and Hardy's paradox", you can't always retrodict a classical trajectory. I wonder if that messes with anything... (Woot for QM paradoxes!)

Ron said...

> you can't always retrodict a classical trajectory

You can't always retrodict a classical trajectory for quantum particles. But that is hardly surprising, since quantum particles don't have trajectories because they aren't classical entities.

Luke said...

> You can't always retrodict a classical trajectory for quantum particles. But that is hardly surprising, since quantum particles don't have trajectories because they aren't classical entities.

What makes the difference between quantum particles and classical particles, other than that the latter are 'observed' [with sufficient frequency]?

I also think Back From the Future is relevant, here: measurements in the future may well be able to change the past. And so, reversing those measurements...

Peter Donis said...

I'm coming to this article late, via a link from a discussion on Hacker News. I wanted to comment on some of the physics discussed in the article.

A trip from earth to pluto at the speed of light is indistinguishable (to the traveller) from a trip from pluto to earth at the speed of light. Both consist simply of being at earth and pluto (and everywhere in between) at the same time (in the traveler's reference frame).

This is not correct. The correct statement is that there is no such thing as a "reference frame" for a "traveler" moving at the speed of light, and no such thing as "travel time" for such a traveler either. The spacetime interval between the two events you describe ("traveler on Earth" and "traveler on Pluto") is zero, but that spacetime interval does not have a valid interpretation as a "time it took to travel". (The more technical way of saying this is that this interval is not a timelike interval--it's a null interval--and the interpretation as "the time it took to travel" is only valid for a timelike interval.)

In other words, the wave function is a static four-dimensional "thing" (for want of a more general noun) out of which space, time, and classical reality "emerge" when you slice it up in certain ways.

This is not correct. The wave function that you "slice up" to get different classical realities is not a four-dimensional thing; strictly speaking, it's a infinite dimensional thing (since it has to include continuous fields), and none of its dimensions are the four dimensions of spacetime. The four dimensions of spacetime form a "framework" on which the quantum fields underlying the wave function are defined.

(Note that some proposed models for quantum gravity do have four-dimensional spacetime emerging from an underlying quantum state. However, that underlying quantum state is not itself a four-dimensional thing, and getting classical four-dimensional spacetime out of it is not a matter of "slicing" it a certain way.)

You can slice up the wave function in a different way and get a different universe.

There is a sense in which this is true, but I don't think it's in any way analogous to choosing different splittings of spacetime into space and time in relativity. The different classical universes in QM are different terms in a superposition; "slicing" means picking one of the terms and discarding all the others, i.e., throwing away information. But choosing a different splitting of spacetime into space and time in relativity does not throw away any information; it just re-expresses the same information (the same 4-dimensional spacetime geometry) in different coordinates.

Your perception of "traveling through time" emerges from your mental states and not the other way around. You feel like you are "traveling forward through time" because your mental states have this natural order to them.

I agree with this, but I don't think quantum mechanics needs to be invoked to explain why it's true. A purely classical model of human memory would have the same property.

Ron said...

> This is not correct. The correct statement is that there is no such thing as a "reference frame" for a "traveler" moving at the speed of light

Cut me a little slack, will you? I'm writing for a (mostly) lay audience. I could have said that the apparent distance travelled and the apparent time elapsed in the traveller's reference frame both approach zero as the traveler's speed approaches c in the reference frame of the origin an destination. That would have been technically correct, but I don't think it would have made the underlying issues any clearer.

> The wave function that you "slice up" to get different classical realities is not a four-dimensional thing; strictly speaking, it's a infinite dimensional thing

Again, do you really think that this detail makes a difference with respect to the target audience?

> The four dimensions of spacetime form a "framework" on which the quantum fields underlying the wave function are defined.

We actually don't yet know whether that is true or not, whether all parallel universes share the same spacetime geometry. Designing an experiment to test this would be quite the challenge.

> There is a sense in which this is true, but I don't think it's in any way analogous to choosing different splittings of spacetime into space and time in relativity.

That depends entirely on how flexible your analogy neurons are. Slicing up the wave function to get classical universes is obviously different in detail from rotating a Cartesian reference frame to get different ideas of what is space and what is time. But in broad brushstrokes, the idea that what appears to be objective truth depends on your perspective is the same in both cases.

> I agree with this, but I don't think quantum mechanics needs to be invoked to explain why it's true. A purely classical model of human memory would have the same property.

There is one important difference, and there is not doubt in my mind that you can figure out what it is if you think about it. Would you like me to tell you, or would you like to puzzle it out for yourself? It makes a fun exercise. Hint: think about the actual physical process of performing a classical erasure of a human brain, and what happens to that human's body while this process is going on.

Peter Donis said...

That would have been technically correct, but I don't think it would have made the underlying issues any clearer.

It would have made a big difference: in all of the cases where the traveler's worldline is timelike (interval getting smaller and smaller, but not zero), you can't even make the argument that the "origin" and "destination" are indistinguishable. It's clear that one is "before" and one is "after". Only when you go to the limit of a null worldline does it apparently (but incorrectly) seem like you can no longer tell the "before" and "after" apart, because the interval itself is now zero.

In other words, if you tried to make your argument using the limiting process you describe, it would have been more obvious that it was wrong. But since you did it directly in terms of the null interval, it seemed like it could be valid. In fact it's not, because even though null intervals are zero, the future and past light cones of any event are still distinguishable, so for any given null interval, the "before" and "after" ends are still distinguishable.

do you really think that this detail makes a difference with respect to the target audience?

Yes, because it implies an analogy which is not valid. See below.

We actually don't yet know whether that is true or not, whether all parallel universes share the same spacetime geometry.

More precisely, if we try to quantize the spacetime geometry as well as the quantum fields, we don't know how to build a consistent theory. So what we do instead is quantize the fields on a fixed spacetime geometry (which does not have to be flat Minkowski spacetime, you can do quantum field theory in curved spacetime). In this model, all of the "parallel universes" do share the same spacetime geometry. And so far, we have not found any experiments that can't be correctly predicted by this model. Theoretical work on quantum gravity is driven by theoretical concerns, not experimental ones. I agree that trying to test any of this experimentally would be quite a challenge.

the idea that what appears to be objective truth depends on your perspective is the same in both cases.

No, it isn't. Classical relativity does not say objective truth depends on your reference frame; it says the opposite. All physical observables are invariants, independent of your choice of reference frame.

think about the actual physical process of performing a classical erasure of a human brain, and what happens to that human's body while this process is going on.

Are you referring to the fact that, in principle, you could make a classical erasure of a brain without also reverting the body's state back to what it was before the memory being erased was formed? (Because classically, you don't have to take into account the entanglement between the brain and the body?)