Visit With Paul Davies on Block Time

[lmoh]

However, we know that a photon, traveling at c does not experience time.

I don't know if it is right to talk about the reference frame of a photon in SR. From what I know about the maths, it looks like c is something like a limit that you approach but never reach. If you ever do talk about an object that is traveling at c in terms of the maths, you would get something like division by zero (but this is just from my limited understanding, hopefully somebody can chime in here). I would leave it to an actual professional to settle this matter though.

 

[PeterDonis]

I don't know if it is right to talk about the reference frame of a photon in SR.

It isn't. PF has a FAQ on this:

https://www.physicsforums.com/showthread.php?t=511170

Does this not mean that a newly-created photon could travel across a Universe essentially frozen in time and that this would constitute a universal *now*?

No. A photon's worldline is a null curve. Any curve, surface, etc. that could possibly qualify as a "universal now" would have to be spacelike.

 

[harrylin]

[..] [...] we know that a photon, traveling at c does not experience time.

Does this not mean that a newly-created photon could travel across a Universe essentially frozen in time and that this would constitute a universal *now*?

I could imagine a stream of newly-created photons each making own their way across the Universe within their own *now* which we , forced to travel at less than c, experience as a flow of time. [..]

I can offer a slightly different and less technical answer: while it is not right to talk about the reference frame of a photon in SR, neither did you mention one. In fact, your description doesn't need it insofar as you discuss the proper "photon experience". But surely that isn't Block Time. The time is only frozen for the photon, and that has little or nothing to do with a "universal now"; it's merely a photon "now".

 

[PeterDonis]

The time is only frozen for the photon, and that has little or nothing to do with a "universal now"; it's merely a photon "now".

The concept of having a "now" doesn't apply to a photon, for the same reason that the photon can't have a reference frame. Saying that "time is frozen" for a photon isn't really correct. What is correct is this: "The invariant length of a photon's worldline is zero." Interpreting that invariant length as "time elapsed" for a photon implicitly assumes that the photon's worldline can serve as the "time axis" of a reference frame, but it can't.

 

[harrylin]

[..] Saying that "time is frozen" for a photon isn't really correct. What is correct is this: "The invariant length of a photon's worldline is zero." Interpreting that invariant length as "time elapsed" for a photon implicitly assumes that the photon's worldline can serve as the "time axis" of a reference frame, but it can't.

I disagree with your claim that "time elapsed" for a photon should implicitly assume anything about worldlines; "time is frozen" or "zero time elapsed" for a photon simply means that the photon "clock" counts zero time between events, such that its celerity dx/dtau is infinite (or, as Einstein put it: in that sense the speed of light plays the role of an infinitely great speed). However such a discussion over words has nothing to do with the Block Time discussion, and I won't comment further.

 

[PeterDonis]

I disagree with your claim that "time elapsed" for a photon should implicitly assume anything about worldlines; "time is frozen" or "zero time elapsed" for a photon simply means that the photon "clock" counts zero time between events,

Events where? Along the photon's worldline. But that requires you to interpret the (zero) interval between distinct events on the photon's worldline as "elapsed time", which would require the photon's worldline to be a possible "time axis" for a reference frame.

such that its celerity dx/dtau is infinite (or, as Einstein put it: in that sense the speed of light plays the role of an infinitely great speed)

But this "sense" does not *justify* the term "elapsed time" to refer to the length of the spacetime interval traversed by a photon; it implicitly *assumes* it, by interpreting dx/dtau as "change in distance with elapsed time", even though you have passed to a limit where the prior justification for that interpretation no longer holds. (There is also a problem with how "dx" is to be interpreted in this limit, btw; interpreting it as "distance traveled" is subject to the same problems as interpreting dtau as "time elapsed"; it assumes that dx is a coordinate interval in an inertial reference frame, but it can't be if dx/dtau is infinite.)

 

[bobc2]

It was discounted by many on philosophical grounds, and by some due to mistaken reasoning (or both). And while Lorentz did say something to that effect, it was meant as criticism on how Einstein presented the theory.

However, as Einstein's derivation was simpler than his own, he simply referred his students to Einstein's derivation (although with the above-mentioned criticism). In case that you are familiar with Einstein's 1905 paper: not long ago I presented a few subtle points of reformulation of that paper as I imagine would have been to the taste of Lorentz. The result is essentially what many people call "LET":
https://www.physicsforums.com/showthread.php?p=3756233#post3756233
Note also the addendum in that post: the simplest and most straightforward is to start with the light postulate as meant by Maxwell, and the rest remains quite the same.

I hope that it will be clear; but if not, we should start it as a separate topic (or perhaps there is still an open, existing thread on that topic).

harrylin, here's a copy and paste from another thread on the SR vs. LET subject:

Originally Posted by DaleSpam
I agree. This is why my understanding is that LET regards the PoR as being something that is violated in nature but cannot be detected.

harrylin response:
I'm fed up with the strawman called "LET"; I now conclude that it is less than useless - thus I will from now on ban it completely.

Does this mean we can evaluate block universe without the need to consider other theories? If there are no competing theories, block universe would be required by SR.

(I know, I've yet to deal fully with PeterDonis's objections)

 

[PAllen]

harrylin, here's a copy and paste from another thread on the SR vs. LET subject:

Originally Posted by DaleSpam
I agree. This is why my understanding is that LET regards the PoR as being something that is violated in nature but cannot be detected.

harrylin response:
I'm fed up with the strawman called "LET"; I now conclude that it is less than useless - thus I will from now on ban it completely.

Does this mean we can evaluate block universe without the need to consider other theories? If there are no competing theories, block universe would be required by SR.

(I know, I've yet to deal fully with PeterDonis's objections)

NO, I've never found LET attractive at all, find SR beautiful, and think block time is an over-interpretation of SR that is neither necessary nor plausible (especially in light of QFT). Note, QFT is a pure SR based theory. I agree with all of Peter Donis's points in this thread.

 

[PeterDonis]

If there are no competing theories, block universe would be required by SR.

(I know, I've yet to deal fully with PeterDonis's objections)

Just to make it clear (it probably already is, but just in case), I am most certainly *not* arguing that the only alternative interpretation to SR other than "block time" is LET.

 

[bobc2]

Just to make it clear (it probably already is, but just in case), I am most certainly *not* arguing that the only alternative interpretation to SR other than "block time" is LET.

Understood. If LET works it would definitely rule out the block universe. Is that also true of the other theories you have in mind?

 

[PeterDonis]

Understood. If LET works it would definitely rule out the block universe. Is that also true of the other theories you have in mind?

"Interpretations", not "theories". LET, at least as I understand it, makes all of the same predictions as standard SR does, so it would also qualify as an "interpretation", although it's certainly not normally thought of that way. But since all of the physical predictions are the same, there is no way for any interpretation to "rule out" others; to do that, it would have to actually graduate to an alternative "theory" by making a different prediction somewhere.

So I think the answer to your question is "no, and neither does LET". All of the different interpretations of SR contradict each other somewhere, but none of the contradictions have any physical consequences.

 

[harrylin]

harrylin, here's a copy and paste from another thread on the SR vs. LET subject:
[...] Does this mean we can evaluate block universe without the need to consider other theories? If there are no competing theories, block universe would be required by SR.

Quite the contrary: I disagree with revisionist history that presents SR as the block universe interpretation of the same, falsely suggesting that Lorentz did not teach SR - and thus I disagreed with the title of that thread.

Instead, once more, here I explained Lorentz's interpretation of special relativity as presented by him: https://www.physicsforums.com/showthread.php?p=3756233#post3756233

 

[harrylin]

"Interpretations", not "theories". LET, at least as I understand it, makes all of the same predictions as standard SR does, so it would also qualify as an "interpretation"

Yes - but:

[..] All of the different interpretations of SR contradict each other somewhere, but none of the contradictions have any physical consequences.

Not exactly: different interpretations of a theory have nothing different to say about predictions of that theory, but as Bell showed, a different theory such as QM may be incompatible with one interpretation but compatible with another one.

 

[PeterDonis]

different interpretations of a theory have nothing different to say about predictions of that theory, but as Bell showed, a different theory such as QM may be incompatible with one interpretation but compatible with another one.

Good point, I was only talking about SR specifically. It's good to bear in mind that SR itself, as a theory, is known to be false: it assumes flat spacetime and spacetime is actually curved, and it assumes determinism and reality does not appear to be deterministic at the quantum level. We still use SR because it's a good approximation within its domain, but that's all.

 

[PAllen]

Good point, I was only talking about SR specifically. It's good to bear in mind that SR itself, as a theory, is known to be false: it assumes flat spacetime and spacetime is actually curved, and it assumes determinism and reality does not appear to be deterministic at the quantum level. We still use SR because it's a good approximation within its domain, but that's all.

I don't see that SR requires determinism. QFT is strictly based on SR as an exact framework and rejects determinism. Further, there are approaches to treating GR as a gravity theory without curvature (separate and apart from quantum approaches). Of course, I agree that however you view gravity, since there is no SR based theory that fully accommodates SM + gravity, SR (at present) can only be used as an accurate local approximation. Yet, I don't consider it established that there cannot be a exist SR based quantum theory that agrees with all verified predictions of SM + GR, and thus can be taken as complete adequate theory of nature.

 

[PeterDonis]

I don't see that SR requires determinism.

Perhaps a better way to put it is that SR, by itself, has no way of modeling non-determinism. See below.

QFT is strictly based on SR as an exact framework and rejects determinism.

QFT in its original form uses flat Minkowski spacetime as its background spacetime, yes. However, you can do QFT in curved spacetime as well. The only treatments I am familiar with for QFT in curved spacetime use Schwarzschild spacetime, which is asymptotically flat; however, I believe quantum cosmology includes models that basically do QFT over a set of Friedmann-like spacetimes. I am not very up to date in this area, though.

Also, the way in which QFT models non-determinism is basically to assign amplitudes to *different* possible spacetimes--that is, different configurations of background spacetime plus quantum fields. (In the original form, the background spacetime is always Minkowski but the field configurations can vary; in the curved spacetime form, the background spacetime can be affected by "back reaction" terms, as in the case of Hawking radiation.) But each individual possible spacetime is "deterministic"; it's a fully determined solution to the appropriate field equations on the appropriate spacetime. So as far as the theory of the underlying spacetime is concerned, there is no indeterminism; each solution to that theory is fully deterministic. The quantum indeterminacy comes from assigning amplitudes to *different* solutions. The classical theory of the underlying spacetime has no way of modeling indeterminism by itself.

Further, there are approaches to treating GR as a gravity theory without curvature (separate and apart from quantum approaches).

Are you referring to the theory of gravity as a spin-2 field on a flat background spacetime? As presented, for example, in the Feynman Lectures on Gravitation? The foreword to that book is now online at Caltech's website:

http://www.theory.caltech.edu/~preskill/pubs/preskill-1995-feynman.pdf

Of course, I agree that however you view gravity, since there is no SR based theory that fully accommodates SM + gravity, SR (at present) can only be used as an accurate local approximation. Yet, I don't consider it established that there cannot be a exist SR based quantum theory that agrees with all verified predictions of SM + GR, and thus can be taken as complete adequate theory of nature.

I'm not sure that an "SR-based" quantum theory, in the sense of a quantum theory that required a fixed flat background spacetime (Minkowski), could be constructed with a classical limit that was, for example, an FRW spacetime. So if predictions about the early universe count as "verified" (which some certainly seem to be), this would count against the possibility of an "SR-based" quantum theory that would be completely adequate.

 

[harrylin]

Some elaboration:

[..] If there are no competing theories, block universe would be required by SR.[..]

Not sure if I read that correctly, but it seems to contain a logical error (or two). If there was only one proposed explanation of why SR works, or even none, that would not mean that SR would require the one that was proposed, or that SR would require no explanation. As it is, there are several competing models and the most popular one (Minkowski's block universe) is often confounded with SR by association.

I gave you a link to a competing model; regretfully it involves going through the rather elaborate presentation in Einstein's 1905 paper if you are not familiar with it. But it's easy to present a much more compact space-time presentation based on that model, as follows (even keeping the historical sequence of development of SR):

1. Postulate a causal, spatial background for physical processes in which light propagates like a wave with speed c (Maxwell-Lorentz).

Define S as a hypothetical reference system that is in rest in that background. Similarly, define S' as a system that is in rectilinear uniform motion with respect to S.

2. From pragmatic considerations, distant clocks are synchronised by assuming that the speed of light dx'/dt' is uniformly c in all directions. However, that is only true with respect to S. Consequently, make a distinction between "local time" t' in S' and "true time" t in S.

3. Based on some experiments, propose that moving objects contract in length:
L'/L = √(1-(v/c)^2)

4. From symmetry (PoR), propose that the natural frequency of moving resonators decreases:
f'/f = √(1-(v/c)^2)

Combining these assumptions yields the following transformation equations for coinciding origins at t'=t=0:

x' = gamma (x -vt)
t' = gamma (t - vx/c^2)

Still one subtle step is needed: so far the meaning of x and t there is not the common one. However, due to the form of these equations, if they are valid for an S in "true rest" then they will also be valid for a system S that is in inertial motion. With that generalisation you obtain the Lorentz transformations for x,x' and t,t' and the related space-time intervals.

As far as SR concerns, the "true time" model is empirically indistinguishable from the "block time" model.

Harald

 

[harrylin]

[..]
3. Based on some experiments, propose that moving objects contract in length:
L'/L = √(1-(v/c)^2)
4. From symmetry (PoR), propose that the natural frequency of moving resonators decreases:
f'/f = √(1-(v/c)^2)
[..]

Oops I see now that I had put the prime sign ' on the wrong side and I should have written "are contracted" and "is decreased". Sorry for that!

The point is that since both models work, it can't be right that Block Time must be the correct one just because it works.

 

[Russell E]

It's easy to present a much more compact space-time presentation based on that model, as follows:
1. Postulate a ... spatial background for physical processes in which light propagates [in time] ... with speed c...
4. From symmetry (PoR) [principle of relativity]...

If your objective is a compact presentation, you already have the complete basis for special relativity with just those two trimmed propositions - properly understood. Note that "like a wave" has been removed from 1, because that is both unnecessary and false, in view of the quantum properties of em radiation, as Einstein had already realized by 1905. Also your items 2, 3, and the rest of 4 are removed because they are consequences of just the above, with the understanding that the "pragmatic considerations" in 2 are simply the choice to use the unique synchronization of standard inertial space-time coordinate systems in terms of which the principle of relativity holds good.

As far as SR concerns, the "true time" model is empirically indistinguishable from the "block time" model.

Empirical indistinguishability is theory-independent (that's why it's called empirical), so the indistinguishability is not limited to "as far as SR concerns".

The "alternate model" you've described is not a different model at all. It's nearly the standard interpretation of the special theory of relativity, albeit presented in an overly convoluted and elaborate way, with many superfluous postulates and assumptions, revealing a lack of understanding of the essence of the theory, which is Lorentz invariance arising from the empirical equivalence between inertia-based and light-based measures of space and time. The only non-standard aspect (setting aside the superfluous meandering) is the pointless metaphysical assignment of the undefined word "true" to one particular (albeit unidentified and unidentifiable) coordinate system.