Does instantaneous communication implies a preferrred Lorentz frame?

[Demystifier]

Didn't mean to imply anything about the perspective itself. Just trying to see all of the alternatives as clearly as possible.

So, let me rephrase my question.
Would you say that now you see this alternative - clearly? (If not, then I failed.)

 

[DrChinese]

Yes, probably as clearly as I am capable.

 

[Demystifier]

Good! That's all I want, to make others understand how is that possible. (Not to convince others that this is how nature really works.)

 

[Fredrik]

In the examples here it looks like you're writing (Y,X) instead of (X,Y) for your pairs,

My pairs are of the form (t,x). That's seems to be the most common way (by far) to write them, probably because it ensures that time is always x^0 instead of being x^4 when we're considering 3+1-dimensional spacetime and x^2 when we're considering 1+1-dimensional spacetime. There's certainly nothing wrong with writing time as the last component, and I know that there are some authors who do that, but I never do. I like to write four-vectors as (x^0,x^1,x^2,x^3)=(t,\vec x), and when I'm dealing with 1+1-dimensional spacetime I just drop the vector symbol from the right-hand side.

...but that doesn't jive with your previous statement, "A line through (1,10) with slope 10 goes trough (0,-99), not (0,0)." If you're simply switching the X-Y order of the pairs, the slope of the line through (1,10) and (0,-99) would be (1-0)/(10- -99) = 1/109, not 10 as you claim.

Oops, you're right about that. I got it wrong in that attempt to explain these things to you. Everything I said in #24 is correct, and as I said in #41, the two statements that you considered contradictory are actually equivalent. I also got it right in #38, where I said that Bob's world line intersects Alice's at t=-99, but the coordinates of that event are (-99,0), not (0,-99). So everything in #17, #24 and #38 still looks completely correct to me, but I finally made a blunder in #41 when I tried to say the same thing that I had already said in #38 in a different way, and accidentally wrote (0,-99) when I meant (-99,0).

I hope that clears it up.

 

[Fredrik]

If you have past and future light cones in your manifold then cause and effect are meaningful.

Ok, if I will ever observe as tiny pieces of glass are jumping up from the ground to form a glass bowl I will agree that your view is more natural. But until then I will consider mine to be more natural. Is it ok?

Your post doesn't contain any actual arguments. I at least tried to explain why I consider my view more natural. You haven't said anything about why you like yours better. Light cones do not give us a reason to prefer the statement "A caused B" over "B caused A".

So for you future is only ever increasing entropy and nothing new? Then maybe we can make a deal - let's the past be for you and the future for me if it the same for you?

It's not the same for me. I remember the past but not the future...because entropy is increasing towards the future. (We have chosen to define "future" as the direction of time in which entropy is increasing). At least that's how we would have to describe it using a classical theory like SR. It seems to me that QM doesn't invalidate this, but I could be wrong about that.

 

[Fredrik]

Fredrik, it seems to me that we have just arrived at an agreement, or at least that we are very close to it.
Let me clarify. Assume that relativistic superluminal influences associated to nonlocality of quantum entanglement - exist. Then, due to decoherence at the macroscopic level, in MOST cases of everyday macroscopic life these superluminal influences are NEGLIGIBLE. Therefore, in most cases there is nothing that can destroy the illusion of free will, because the illusion of free will is an emergent phenomenon that appears only at the macroscopic level. That is why the illusion of free will is so effective in practice. Nevertheless, significant superluminal influences at the macroscopic level are possible, at least in principle. This means that, in principle, they can destroy the illusion of free will. But in practice, it does not happen due to decoherence. In this way, hypothetic relativistic superluminal influences (at the fundamental microscopic level) are compatible with the fact we feel the existence of free will (at the emergent macroscopic level).
This is similar to the fact that quantum mechanics at the fundamental microscopic level is compatible with classical mechanics at the emergent macroscopic level.

Do you agree now?

I don't know. I don't think entanglement has a lot to do with what we've been dicussing, since entanglement can't be used to send superluminal messages. It seems to me that the only thing that can be a problem for the illusion of free will is the existence of easily detectable tachyons. Note that if we take the ensemble interpretation of QM as our starting point, we can't even conclude that there are superluminal influences in tests of entanglement over spacelike separations (since this interpretation assumes that QM isn't a description of what actually happens).

On the other hand, that assumption strongly suggests that there's an underlying theory that does describe what actually happens, and that the variables of that theory aren't observables. (If they were, I think we'd have problems with Bell inequalities and that sort of stuff). Since we have no idea what that theory is, it's hard to speculate about it, but I'll offer two thoughts about it anyway.

I don't see a reason to think that superluminal influences between unobservables would force us to behave in a certain way to avoid a violation of "global consistency" like the paradox described in #24. So such influences wouldn't necessarily be a problem for the illusion of free will, like easily detectable tachyons would.

It also seems to me that Minkowski spacetime may be the illusion here. Maybe concepts like "distance" or "spacelike separation" don't even enter into the description of entanglement in the fundamental model. Maybe the two entangled particles are just a single system there, and the separation between them is the illusion. I think that concepts such as "distance" are unavoidable in a falsifiable theory, because it must include a mathematical representation of operationally defined equivalence classes of measuring devices (i.e. an algebra of "observables"), and it's impossible to define anything operationally without mentioning space and time. But they do not have to be a part of a description of what actually happens in terms of unobservables.

Now let's take the MWI as the starting point instead. (I define the MWI as the assumption that the usual axioms about Hilbert space and the probability rule can be interpreted as a description of what actually happens). Decoherence can cause correlations in spacelike separated measurements, but does it make sense to describe those as "superluminal influences"? I don't know. I don't understand the MWI and decoherence well enough.

 

[dmtr]

Now let's take the MWI as the starting point instead. (I define the MWI as the assumption that the usual axioms about Hilbert space and the probability rule can be interpreted as a description of what actually happens). Decoherence can cause correlations in spacelike separated measurements, but does it make sense to describe those as "superluminal influences"? I don't know. I don't understand the MWI and decoherence well enough.

AFAIK all these causality-loop paradoxes simply go away under the MWI interpretation. In the example above 'Alice sending 0' would be from a different subjective history from the 'Alice sending 1'.

 

[Fredrik]

AFAIK all these causality-loop paradoxes simply go away under the MWI interpretation. In the example above 'Alice sending 0' would be from a different subjective history from the 'Alice sending 1'.

That may be true, but I suspect that it's not that simple. If it's true, then it should be easy enough to write down a QM description of the sequence of events I described in #24 that makes it obvious that what you said is correct. I expect that the result would be the opposite, i.e. that if we tried to prove this, we would end up proving that the same paradox is present in the MWI too. I haven't tried to do this yet. Maybe I'll give it a shot later.

 

[zonde]

Your post doesn't contain any actual arguments. I at least tried to explain why I consider my view more natural. You haven't said anything about why you like yours better. Light cones do not give us a reason to prefer the statement "A caused B" over "B caused A".

But Frederik, you are unjust. I gave my argument previously you just dismissed it.
I said that time can be measured directly with clock while entropy do not have means of direct measurement.
And there is straight forward motivation for that argument. Because while mathematical model hangs in the air it is useless but to make some relation to physical world you are facing conversion process where you are using measurement equipment. And the more straight forward is this conversion process the less (interpretation) error prone it is.
To be more direct - how can you state that entropy is increasing if you can not measure it? How can you state that system is closed? And as to closed systems I think there are no closed systems in physical world and so there is no physically meaningful way to define change in entropy. Consider for example a possibility that fields can be carriers of disorder - can you shield any system from fields?

And if you need a reason why one direction in time is preffered over the other - the reason is that we do not care about changing the past but we care about changing the future (we care about our survival in the future).

 

[zonde]

Demystifier,
before you make conclusion I propose to consider another scenario.
Take Frederik's example with two computers communicating "1" and "0". But modify it so that it becomes non contradicting i.e. computers are just storing the value they get and then transmit it back to the past of other computer.
As I understand you allow possibility of such scenario.
In that case we can have this causality loop in two states:
1. first computer have stored "0"; it transmits "0" to second computer; second computer have stored "0"; it transmits "0" back to first computer.
2. first computer have stored "1"; it transmits "1" to second computer; second computer have stored "1"; it transmits "1" back to first computer.
Separately each of two computer do not have free will but the whole causality loop has free will to be in one of two states. The state at any point is not random and is not caused by any physical law outside this causality loop.

What you can say about this scenario? Do you see any inconsistency in my reasoning?

 

[Fredrik]

I gave my argument previously you just dismissed it.
I said that time can be measured directly with clock while entropy do not have means of direct measurement.

So if the events A and B are causally connected, and all the clocks agree that A is earlier, how does that give us a reason to say that A caused B instead of B caused A? Ah, I see you answered that too:

- the reason is that we do not care about changing the past but we care about changing the future (we care about our survival in the future).

We feel that way because we feel that it's possible to change the future and impossible to change the past. Why do we feel this way? Because we have memories of the past and none of the future. Why is that? Because storing a memory is a process that increases entropy, and entropy is increasing towards the future. Why is entropy increasing? Because the universe started out in a low entropy state. Why did it do that? No one knows.

 

[Demystifier]

Demystifier,
before you make conclusion I propose to consider another scenario.
Take Frederik's example with two computers communicating "1" and "0". But modify it so that it becomes non contradicting i.e. computers are just storing the value they get and then transmit it back to the past of other computer.
As I understand you allow possibility of such scenario.
In that case we can have this causality loop in two states:
1. first computer have stored "0"; it transmits "0" to second computer; second computer have stored "0"; it transmits "0" back to first computer.
2. first computer have stored "1"; it transmits "1" to second computer; second computer have stored "1"; it transmits "1" back to first computer.
Separately each of two computer do not have free will but the whole causality loop has free will to be in one of two states. The state at any point is not random and is not caused by any physical law outside this causality loop.

What you can say about this scenario? Do you see any inconsistency in my reasoning?

I think it is consistent. And it can be called "free will" if you like. However, unlike the ordinary free will, this free will acts in a nonlocal manner. That's why I do not prefer to call it "free will".

Or let me be more clear. In order to avoid inconsistencies, I need an absence of local free will. Since by "free will" one usually means "local free will", it is simpler to say a simpler statement that "there is no free will", rather than a more precise statement that "there is no local free will".

 

[Demystifier]

We feel that way because we feel that it's possible to change the future and impossible to change the past. Why do we feel this way? Because we have memories of the past and none of the future. Why is that? Because storing a memory is a process that increases entropy, and entropy is increasing towards the future. Why is entropy increasing? Because the universe started out in a low entropy state. Why did it do that? No one knows.

The most elegant explanation of this stuff I ever seen.

 

[Demystifier]

I don't think entanglement has a lot to do with what we've been dicussing, since entanglement can't be used to send superluminal messages.

It seems that you have not understood the whole motivation for this thread. As said in the first post, it is quantum nonlocality that suggests the existence of superluminal messages. (I guess I don't need to explain that quantum nonlocality is a consequence of entanglement.) Of course, such messages, if exist, cannot be CONTROLLED by humans (which is what you probably meant when you said the above), but the idea is that these messages are exchanged between the microscopic particles anyway. In this context, my requirement that "there is no free will" actually means that microscopic particles do not have free will, i.e., that these particles cannot CONTROL the messages. But the inability to control does not mean that they do not exist.

Concerning the relation between human free will and microscopic-particle free will, there is a "free will theorem" discussed several times on this forum.

 

[Demystifier]

It seems to me that the only thing that can be a problem for the illusion of free will is the existence of easily detectable tachyons.

With that I agree. And decoherence explains why the effect of tachyons responsible for superluminal communication among entangled particles, if they exist, cannot be easily detected.

On the other hand, that assumption strongly suggests that there's an underlying theory that does describe what actually happens, and that the variables of that theory aren't observables. (If they were, I think we'd have problems with Bell inequalities and that sort of stuff). Since we have no idea what that theory is, it's hard to speculate about it, but I'll offer two thoughts about it anyway.

Well, Bohmians do have an idea what that theory might be. I have given a link to a concrete theory in the first post.

I don't see a reason to think that superluminal influences between unobservables would force us to behave in a certain way to avoid a violation of "global consistency" like the paradox described in #24. So such influences wouldn't necessarily be a problem for the illusion of free will, like easily detectable tachyons would.

If you think so, then we actually agree. It is possible that superluminal influences exist without a preferred frame, without affecting the illusion of free will.

Decoherence can cause correlations in spacelike separated measurements, but does it make sense to describe those as "superluminal influences"? I don't know. I don't understand the MWI and decoherence well enough.

Well, decoherence actually DESTROYS correlations.

 

[zonde]

So if the events A and B are causally connected, and all the clocks agree that A is earlier, how does that give us a reason to say that A caused B instead of B caused A?

Well maybe it was a mistake to talk about reasons for time asymmetry.
Maybe the reasons does not matter so much as long as we agree that time is asymmetric. We just extend this asymmetry to causally connected events. You can call all the effects the causes and all the causes the effects as long as you treat them asymmetrically.
The clock just gives us easy way to track this asymmetry.
Maybe we have disagreement because I am more interested about operational definition but you about theoretical definition.

Anyways I looked back at the start of our discussion and it occurred to me that there is certain difference between the past and the future. We take present as evidence about our past but we do not take present as evidence about our future. And that is basic fact of our existence that should be taken as is.

 

[Demystifier]

We feel that way because we feel that it's possible to change the future and impossible to change the past. Why do we feel this way? Because we have memories of the past and none of the future. Why is that? Because storing a memory is a process that increases entropy, and entropy is increasing towards the future. Why is entropy increasing? Because the universe started out in a low entropy state. Why did it do that? No one knows.

Since your understanding of this stuff is so clear, I hope the following argument will look convincing to you.

There are two types of messages: those the receiving of which increases entropy, and those the receiving of which does not increase entropy (*). The former type cannot be received from the future (because it would contradict the fact that entropy is larger in the future), but the latter type can. In fact, messages the receiving of which does not increases entropy are not detected by humans. I mean, they may influence microscopic degrees of freedom of our bodies, but they are not perceived by consciousness. Since the illusion of free will is one of the manifestations of consciousness, it follows that messages from the future cannot influence the illusion of free will.

(*) One may doubt that messages which do not increase entropy can exist. But they can, provided that they occur at the microscopic level, at which the nature is symmetric under time inversions (up to irrelevant effects of weak interactions). Indeed, instantaneous messages betwen distant entangled particles in Bohmian mechanics are of this sort.

 

[Demystifier]

We take present as evidence about our past but we do not take present as evidence about our future. And that is basic fact of our existence that should be taken as is.

The question is: Is it only our subjective experience of time, or is it an objective property of the time itself?

Fredrik and me are proponents of the idea that it is only our subjective experience, an illusion. You seem to think that it is an objective property. But you know that an objective (i.e., observer independent) distinction between future and past is not compatible with relativity (viewed as a fundamental law). That's probably the main reason why many physicists prefer the idea that the distinction between "future" and "past" is only an illusion and does not make any sense in the absence of increasing entropy. Just like the distinction between "up" and "down" is only an illusion, and does not make any sense in the absence of gravity.

 

[Demystifier]

AFAIK all these causality-loop paradoxes simply go away under the MWI interpretation. In the example above 'Alice sending 0' would be from a different subjective history from the 'Alice sending 1'.

This is actually not true. According to MWI, the universe splits ONLY when decoherence takes place. On the other hand, causality-loop paradoxes may also appear in cases in which decoherence does not take place.

 

[heldervelez]

As time goes by the level of organization and complexity in the universe is increasing.
I see Life ( and US with our increased capacity to work on knowledge) as the ultimate organization, the last creation of the Universe.
In the beginning of the universe there are no organization at all.
How can we say that entropy increases?

The universe is not closed as required by 2nd law. Also it is not open because ...
the relation open/close involves more than one entity nor we can not find frontiers, nor any other 'object' surrounding the universe.

At least in the scale of the universe we can not say 'the entropy increases'
-----------------------
The whole story about 'arrow of time' is misleading.
We construct equations that describe events, in an observed order. It gives us even some capacity to preview how events will be.
But the 'arrow of time' of our equations are a mental construction, not a real object and we can not manipulate it, reverse it, etc. It is an index we assign to events.

-------------------------
Does a system like the universe could survive if instantaneous communication were possible ? I think not.
Instantaneous communication can not be faster then light. I can live with this.

--------------------------
Entanglement ?
I see it this way :
particle pair A, B created at the same time. Latter we find and measure A. Latter if we find B don't we measure it as B? The properties of A and B are determined from the beginning on the moment of their creation.

 

[PTM19]

Consider the following scenario: (You probably have the spacetime diagram in your head already, so I won't describe everything in detail).

A computer that's hooked up to a tachyon transmitter is running a program that tells it to send a 1-bit message at t=1 that contains the NOT of the 1-bit message it received at t=0. (If we receive 1, send 0. If we receive 0, send 1). The computer at the other end is set up the same way, except that it's running a program that tells it to send the same 1-bit message that it has received, immediately after it has received it. This scenario contains a paradox:

Receive 1 → send 0 → receive 0
Receive 0 → send 1 → receive 1

You're suggesting that a way out of this is to say that free will exists only as an illusion. How does that help? Do you mean that computers running these two programs can't be built? Do you think they can't be hooked up to tachyon emitters or tachyon detectors? What do you think happens if I try? Do I get hit by a meteor? What if we all try? More meteors?

Do you think we would be unable to choose to try? I have two problems with that option. First of all, that doesn't just eliminate free will, but also the illusion of free will. How can we resolve a paradox by assuming that something we know is true (that we have the illusion of free will) is false? Second, if our brains are just physical systems that gather and utilize information, then we're not very different from the computers mentioned above. So why would it be more likely that the event that prevents the paradox happens in my brain than in the computer?

Such a paradox may also be resolved in a slightly different way, any scenarios which are not consistent may simply interfere destructively with each other. So in the above example the tachyons won't reach the detector, they will probably slightly miss, just enough to not register, because the actual path between the sender and the receiver will have 0 probability of being realized.

This would mean that future is naturally restricted by interference to scenarios which do not lead to paradoxes and it does not say anything about free will as the actual selection happens on the level of physics.

Quantum interference in time has been demonstrated which I think gives some support to this hypothesis: http://arxiv.org/abs/quant-ph/0507044

 

[Demystifier]

Instantaneous communication can not be faster then light. I can live with this.

So you can live with
infinity < 300.000 ?
Good for you, but I can't.

 

[Demystifier]

Quantum interference in time has been demonstrated which I think gives some support to this hypothesis: http://arxiv.org/abs/quant-ph/0507044

The work of Horwitz (though not this particular paper) has been also influential on the paper mentioned in the first post in this thread.

If you like the work of Horwitz, you may like this paper as well. In fact, this paper also solves some problems of the Horwitz approach.

 

[RUTA]

My pairs are of the form (t,x). That's seems to be the most common way (by far) to write them.

Can't say I've ever seen them written that way, but I don't do much with SR. I do teach, publish and referee in GR and you're right, we (almost) always put time first in the four vector so it seems odd that we switch them in the SR graph. But, as I said, I've only seen it (x,t) there.

I hope that clears it up.

Crystal clear, thanks. I will use this simple example when showing SR to the non-science students in my general education courses.

 

[heldervelez]

So you can live with
infinity < 300.000 ?
Good for you, but I can't.

Demystifier :
A obvious nonsence I've written and you did spot. I'm not imune to errors. This one is a a comunicational 'lapsus linguae' and, for sure, I will add some errors about spelling.

Why should the entropy be increasing, as you said, and against all evidence as I said?

All errors are not at same level. I do not know yet what "smile" to choose to your error.

 

[dmtr]

We feel that way because we feel that it's possible to change the future and impossible to change the past. Why do we feel this way? Because we have memories of the past and none of the future. Why is that? Because storing a memory is a process that increases entropy, and entropy is increasing towards the future. Why is entropy increasing? Because the universe started out in a low entropy state. Why did it do that? No one knows.

That's a common misconception. Only erasing a memory is a process that increases entropy (look-up Landauer and Bennett).

 

[dmtr]

That may be true, but I suspect that it's not that simple. If it's true, then it should be easy enough to write down a QM description of the sequence of events I described in #24 that makes it obvious that what you said is correct. I expect that the result would be the opposite, i.e. that if we tried to prove this, we would end up proving that the same paradox is present in the MWI too. I haven't tried to do this yet. Maybe I'll give it a shot later.

Just skim over a relevant chapter of 'the Fabric of Reality' by David Deutsch (one of the MWI proponents).

 

[PTM19]

The work of Horwitz (though not this particular paper) has been also influential on the paper mentioned in the first post in this thread.

If you like the work of Horwitz, you may like this paper as well. In fact, this paper also solves some problems of the Horwitz approach.

Thanks, I'll check it out.

That's a common misconception. Only erasing a memory is a process that increases entropy (look-up Landauer and Bennett).

I think what you refer to is only true in information theory, it certainly does not hold for biology. Human brain increases entropy when storing memories - neurotransmitters are being released into synapses, neurons change connections, genes are switched on and off and plenty of of ATP is hydrolyzed.

 

[dmtr]

I think what you refer to is only true in information theory, it certainly does not hold for biology. Human brain increases entropy when storing memories - neurotransmitters are being released into synapses, neurons change connections, genes are switched on and off and plenty of of ATP is hydrolyzed.

I prefer to be very careful with the entropy. As Landauer and Bennett have shown our intuitions can easily be wrong. For instance new entanglements in between the biological system and the environment are continuously generated (decoherence). So you have to consider the contribution of the entanglement entropy. This contribution is negative (entropy of the entangled systems is less).

So my answer to your 'certainly' would be: not necessarily.

 

[Demystifier]

Why should the entropy be increasing, as you said, and against all evidence as I said?

You have argued that the evolution of life is an evidence against increasing entropy. Even if you are right that this is AN evidence, you still cannot claim that ALL evidence is against it. I think that it is safe to say that MOST phenomena in nature confirm increasing entropy, even if the evolution of life is an exception.

Now, how the evolution of life may be compatible with increasing entropy? To be honest, I don't know. But that's because I don't have a good understanding of the evolution of life. After all, I am a physicist, not a biologist.
Nevertheless, I do know that total THERMODYNAMIC entropy in the Universe is increasing with time, despite the fact that life becomes more and more ordered.

Even though I don't know the answer to the question above, I think I can still say something nontrivial about it.
There are two possibilities:
Either life is a quite improbable and rare event in the Universe, or a very probable and frequent phenomena happening on a large number of planets with appropriate physical conditions (liquid water, atmosphere, etc.).
If it is quite improbable, then life is a statistical fluctuation. As the law of increasing entropy is a statistical law, not the exact law, a rare fluctuation is not in contradiction with that law.
If it is very probable, then it is in fact a consequence of increasing entropy, even though we do not understand in detail why exactly is that so.