Talk:Special relativity/Archive 9

From Wikipedia, the free encyclopedia

Archive This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

A suggestion regarding the postulates section

I propose that the postulates section of this article contain the original (translated) version of the postulates, as written in "On the Electrodynamics of Moving Bodies" as well as the generally accepted 'modern' version of the postulates. Further, I propose that there should be little or no additional information in the postulates section. Edits to the postulates section should be discouraged here in this article but encouraged in the article Postulates of special relativity. In other words, the less said about the postulates here, the better IMHO. Let this article emphasize the implications of the STR rather than the postulates themselves. Thoughts? Alfred Centauri 23:01, 12 May 2006 (UTC)

We're still going in circles, for it was stable like that for a long time: the postulates of 1905 with the definition of "speed of light" as used in those definitions. We had in January:
1. First postulate (principle of relativity)
The laws of electrodynamics and optics will be valid for all frames of reference in which the laws of mechanics hold good (non-accelerating frames).
In other words: Every physical theory should look the same mathematically to every inertial observer; the laws of physics are independent of the state of inertial motion.
2. Second postulate (invariance of c)
Light is always propagated in empty space with a definite velocity c that is independent of the state of motion of the emitting body; here the velocity of light c is defined as the two-way velocity, determined with a single clock.
In other words: The speed of light in vacuum, commonly denoted c, is the same to all inertial observers, and does not depend on the velocity of the object emitting the light. An observer attempting to measure the speed of light's propagation will get the same answer no matter how the system's components are moving.
If we want to say the least as possible about the postulates needed for deriving the LT, then Pauli's version is preferable as it doesn't require the clarification about c. Harald88 09:01, 13 May 2006 (UTC)
In fact after comparing the with the treatment in a number of books, IMHO the postulate based intro in its current form seem out of place, it is more a history of science and philosophy of science topic (to be handled in History of special relativity and Postulates of special relativity).
The emission theory is already incompatible with pre-SR electrodynamics, Maxwell's equations make the velocity of EM radiation independent of the velocity of the source. Ritz had to modify ME to give a (somewhat) consistent emission theory.
The only thing to be added to Maxwell's equations to arrive at the Poincare group as invariance group, is the postulate that they hold in every inertial frame, not only in the ether frame. That is the way, the topic is introduced e.g. in the books of Born and Møller or Einstein's 1921 Princeton lectures.
There is only a need for separate second postulate if one doesn't want to pull in the full Maxwell equations but is looking for the weakest possible statement, which will result in full SRT.
Pjacobi 10:15, 13 May 2006 (UTC)
It was the very purpose of the postulates to derive the LT from minimal asumptions. Harald88 16:31, 13 May 2006 (UTC)
Yes, but is this pupose to be presented in the main article Special relativity, if it is no longer only oer even main form to preset SR in text books?
Anyway, in your formulation, the 2nd postulate is redundant, as the 1st already pulls in Maxwells'equatation.
Pjacobi 18:13, 13 May 2006 (UTC)

This is going to be harder than I thought... OK, I made two proposals.

(1) Include the translated original form of the postulates of the STR as given in 'On the Electrodynamics of Moving Bodies".

Yes or No?

(2) Include a generally accepted modern version of the postulates of the STR.

Yes or No?

Alfred Centauri 22:18, 13 May 2006 (UTC)

Ad 1.: That's essentially the version as existed (here above), and which has subsequently been edited beyond recognition. Apparently there was disagreement about them.
Thus it seems wise to merge the discussion of Einstein's formulation with the Postulate article; moreover, I now plan to include a separate section in this article on "Speed of Light".
Ad 2.: Effectively, that was here above proposed by Pjacobi: Pauli's version may probably be regarded as one of the first "modern" versions of the postulates.
Cheers, Harald88 11:30, 14 May 2006 (UTC)
PS: Eventhough Einstein's paper was a milestone because of the derivation of the LT, his presentation shouldn't be regarded as more "authentic" or "true" than others. In that respect Pauli's discussion is interesting (although not without errors), see http://www.questia.com/PM.qst?a=refresh&docId=88787429&type=book Harald88 11:40, 14 May 2006 (UTC)

Yes, they have been edited beyond recognition. Thus, my proposal to put in the translated original version of the postulates. Any edit to the translated original words of Einstein will be reverted. There is no reason to edit those words - they are his words and there is no room for interpretation of whether those are his words or not, right?

If we then follow with Pauli's more modern and transparent version of the postulates, we once again have a source and there would be no reason to edit those words. Any change to Pauli's words would be reverted for good reason. Do you see what I'm getting at? Alfred Centauri 13:19, 14 May 2006 (UTC)

Which of Einstein's versions of the first posulate do you want to use?

Let me throw into this discussion the 1916 version of the first postulate. In 1916, in the Annalen der Physik article describing general relativity, Einstein wrote:

"The special theory of relativity is based on the following posulate, which is also satisfied by the mechanics of Galileo and Newton.
"If a system of coordinate K is chosen so that, in relation to it, the physical laws hold good in their simplest form, then the same laws will also hold good in relation to any other system of coordinates K' moving in uniform translation relative to K. This postulate we call the 'special princuiple of relativity'. The word 'special' is meant to intimate that the principle is restricted to the case when K' has a motion of uniform translation with respect to K' ...
"Thus the special theory of relativitty does not depart from classical mechanics through the postulate of relativity, but through the constancy of the speed of light in vacuo, from which in combination with the special principle of relativity, there follow ... the Lorentz transformations, and the related laws ... ."

The problem is that the understanding of what Einstein did in devising the special relativity has evolved since, and did even for Einstein. Come 1916, he now knows that this is the special principle of relativity. So we need to note that in the aritcle and in the aritcle of the principle of relativity. I also advise choosing a definite modern paraphrasing of these laws and enforce that. I find it hard to consider Einstein's 1905 version to the best, especially when when Einstein ended up tweaking it to make room for general relativity.

In a related vein, I wish to make it clear that I strongly oppose any mention of round-trip time for the speed of light. Harald88 is being confused by a definition that Einstein given early in the 1905 paper related to the synchonization of clocks. The raw definition given by Einstein, which directly states that the one-way speed of light is constant, is what is important. (I also consel people not to use the work "velocity" in the posulates. That is used due to a translator's choice, and is not the correct term given the context in which the german word was used. In english, "velocity" is a vector, and "speed" is a scalar which is the absolute velocity of a vector. It is the speed of light that is constant, not the velocity.) --EMS | Talk 04:39, 15 May 2006 (UTC)

We do have History of special relativity and Postulates of special relativity. This article should use contemporary language and flow of presentation. There is little reference to history necessary here. --Pjacobi 06:16, 15 May 2006 (UTC)
That is basically what I am after here. I have done some work on the first postulate statement in the article, and removed the excess commentary. The second posulate I am leaving alone for now, although the phrase "independent of the motion of the emitting body" impresses me as being redundant even though it is Einstein's words. To me, the statement that the "speed is light (in enpty space) is a constant" suffices. --EMS | Talk 14:52, 15 May 2006 (UTC)
I similarly strongly oppose any suggestion that with "speed of light postulate" the one-way speed is meant, as that would be pseudo scientific. EMS confuses his lack of understanding (or even knowledge) of the abovementioned references with my alleged lack of understanding. According to Einstein, OWLS is a stipulation, and Poincare as well as modern papers call it a convention. As opinions are devided on this issue, obviously some of us have a lack of understanding about the relativity/absoluteness of one-way light speed; which underscores the need for a section on this topic (but preferably the issue is avoided in the postulates section!). What papers do the "OWLS=absolute" proponents refer to?
Note also that "the speed of light is constant" is meaningless without the precision relative to what. Thus the formulation by Pauli is definitely the best. Harald88 19:40, 15 May 2006 (UTC)
Might I propose that the issue of one-way versus two-way be not mentioned in this article, be mentioned in a subsection of this article with both sides being represented, or possibly have the issue discussed in another article, considering how it seems to be in contention. The second postulate could then be stated in the postulates section in a way that doesn't necessarily specify which one it is. Certainly neither of the wordings being batted about by Harald88 and EMS favor one or the other in such a way that jumps right out at you.
As to calling explanations of the second postulate in terms of the one-way speed of light pseudo-science, that's a bit much. There have apparently been (and given my limited access to articles, I could be wrong) numerous attempts to verify the isotropy of the one-way speed (http://prola.aps.org/abstract/PRL/v60/i2/p81_1, http://prola.aps.org/abstract/PRD/v42/i2/p731_1, and http://prola.aps.org/abstract/PRA/v56/i6/p4405_1 that I have found), with isotropy in this case implying constancy (if the one-way speed were isotropic but not constant, then the two-way speed would also not be constant as a two-way trip is composed of two one-way legs). So it seems to be verifiable, and apparently verified. Given this, the constancy of the one-way speed of light does not appear to be pseudo-science.
As to what the second postulate says, and whether it addresses the one-way speed or two-way speed is obviously a matter of much debate (if you don't believe me, look at the rest of this page :) ), and to be honest when Einstein proposed the postulate, he didn't specify, though what he went on to say could be taken to imply one thing or another. Given this, I reiterate my proposal that the second postulate be worded in a way that simply does not explicitly address the issue, and ideally does not implicitly favor either side.
To put my money where my mouth is, so to speak, and to start the discussion along these lines, might I propose: "The speed of light in vacuum is taken to be a universal constant, c, independent of the motion of the source of the light."
There is my two cents, on top of the five dollars I've already put in.  ;) DAG 22:33, 16 May 2006 (UTC)
DAG, thanks for your contribution, and I think that it's not too bad in view of an upcoming clarifying section on the issue.
Apart of that, probably you overlooked in the overflow of text my clarifications to you above (22:42, 11 May 2006) for a good article by Ives as well as the straightforward statement from Foundations of physics (21:15, 9 May 2006) :
we simply think that the lack of observability allows a multiplicity of conventional assumptions, encapsulated in some synchronization gauge, which are consistent with any possible experimental evidence.
IMO the label pseudo science is a very strong but correct description for a prediction that "lacks observability" so that it is "consistent with any possible experimental evidence" (and apart of citing such statements I can also explain why they are correct; but such elaboration results in a separate article of course). Harald88 23:07, 16 May 2006 (UTC)

Harald, if the second postulate doesn't pertain to the one-way speed, then to what does it pertain? (It could not pertain to Maxwell's c because this case was closed to postulation prior to special relativity. Likewise, it couldn't pertain to the round-trip case because this case was also closed before SR. One does not need to postulate regarding past experimental results. Einstein had to be postulating regarding some future result, and the only one left was the OWLS result.)

The following points could help:

Since Einstein invoked the PR in connection with the one-way, two-clock light speed (in his book he said that light's one-way speed would be c - v given classical clocks, and he went on to say that this result conflicted with the principle of relativity), Einstein - at that point - had to have considered his one-way result to be a law of physics because the PR pertains only to laws. But confusion arises when we are given (as you noted) Einstein's claim that his one-way invariance is a mere stipulation. How can a mere stipulation be a law of nature?

This brings us to the conventionality problem, the necessary attempt to find some sort of physical content within the second postulate. This can be done only by looking behind the simple-but-confusing face-value stuff just mentioned. An important clue is Einstein's statement that we must discard absolute time, i.e., we must discard the absolutely synchronous clocks of classical physics because they don't fit in with Einstein's null-results-always scheme, even though he replaced (good)synchronous clocks with asynchronous (incorrect) ones!

This tells us that half of the physical content of the 2nd postulate has to be Einstein's assumption that clocks are properly related only if they get a one-way null result. And the other half of the physical content has to be Einstein's assumption that we cannot absolutely synchronize clocks.

Only by doing this can we eliminate the total confusion (from Einstein himself) re his 2nd postulate. Otherwise, we simply have no scientific postulate because it is certain that all we would have is a mere definition (of synchronization), a nonlaw, a nonpostulate. Cadwgan Gedrych 19:09, 16 May 2006 (UTC)

I already explained that above: the second postulate simply states the wave nature of light and the outcome of MMX as facts of observation. Of course, the first postulate already implies the outcome of MMX; but the value c is required as constant in the LT.
And there is no "absolute" OWLS result, as the abovementioned articles explained. OWLS is as you like, as is - in consequence - simultaneity; that doesn't affect observations. One shouldn't confuse convenience with predictions!
I would have to look it up, but probably you misunderstand Einstein's argument; he knew very well the connection between OWLS and simultaneity.
It would help you if you read up related articles; then you'll realise that Einstein didn't really state anything new there. Reading the papers that Einstein had as background helps to follow his argumentation.
Or you can read again Pauli's formulation of the same; what issue do you have with that one? Harald88 22:49, 16 May 2006 (UTC)
Harald - Cadwgan is quite right in the above. My only quibble is in his remarking that asynchronous clocks are "incorrect", but either you live with that consequence for co-moving clocks with are in motion with respect to each other or you can't accept SR (and we already know which choice Cadwgan is making). The current wording of the second posulate in the article (the initial version of which was placed there by you BTW) is
The speed of light in a vacuum is a universal constant (c) ...
The speed of light in a vacuum cannot be a universal constant if it is going at a different rate in one direction than the other. So that blatantly is a OWLS definition, one that appropriately reflects Einstein's intent and is the operative part of the second postulate.
It's time to give this business a rest. I thank you for helping us to refine the postulates section. The work is done, and I strongly advise that you rest on your laurels for a while. --EMS | Talk 14:57, 17 May 2006 (UTC)

The second postulate has not been stated

It is important to remember that many people who read the article may not go on to read the separate section about the postulates.

This means that the article's postulate section must be valid and unambiguous even if brief. But this is certainly not the case.

For example, look at the current (5-15-2006) version of the second postulate:

Second postulate - Invariance of c - In empty space, Light always propagates
with a constant speed c, independent of the state of motion of the emitting 
body.

It is vital to mention which light speed. Einstein did not postulate regarding light's round-trip speed. Here are his own words from his special relativity paper:

"In agreement with experience we further assume the quantity 2AB/(t'A-tA)=c [light's round-trip speed] to be a universal constant ..."

Einstein merely and simply accepted the Michelson-Morley round-trip null result as an experimental fact ("In agreement with experience...").

This left only the one-way speed for special relativity. The entire purpose of special relativity was to show how light's one-way speed could also be invariant.

Einstein did this in two steps, as follow:

[1] Seeing that absolutely synchronous clocks would get c ± v for light's one-way speed, Einstein proposed that such clocks cannot be obtained. (It is important to note that he did not prove this.)

[2] Additionally, he maintained that clocks will be properly related only if they are set by definition to obtain one-way light speed invariance.

We can now make sense of Einstein's version of his second postulate, which follows (slightly paraphrased for clarity):

Any ray of light moves relative to any coordinate system with the one-way, two-
clock speed c as follows:
speed = light path/time interval
where the two-clock time interval is given by the definition in Section 1.

Since the definition applied to all inertial coordinate systems, all must get c simply by definition, but the important thing here is Einstein's belief that his clocks are properly related. This is extremely important because all of special relativity's consequences are based on the use of at least two clocks per frame, two clocks which are related via Einstein's definition.

For example, since the lack of simultaneity depends on the use two clocks per frame, Einstein felt that this is a proper result, i.e., that we simply must live with relative time (or relativity's asynchronous clocks).

Similarly, since Einstein's transformation equations are based on the use of two clocks per frame, he maintains that these are the proper equations for physics.

Also similarly, since Einstein's addition (or composition) of velocities theorem is based on at least two clocks per frame, he believes that this result is valid.

However, to be fair to both sides, it must be emphasized that the absolutely synchronous clocks of classical physics would not get c for light's one-way speed, no one has proved that such clocks cannot exist, and they would entirely do away with special relativity. Cadwgan Gedrych 20:32, 15 May 2006 (UTC)

I have no big quibble with much of the detail here, although I see your version of the second postulate as being redundant and self-referential. The more direct version is much more my preference.
As for your last paragraph, please see Special_relativity#Status. Based on the overall success that SR has had over the years, it is now your job to show that absolute synchonization exists instead of bwing the job of those who support SR to show that it does not. --EMS | Talk 20:59, 15 May 2006 (UTC)
Einstein's philosophy (as well as those of editors) is irrelevant for scientific theories. SRT relies on the observation that despite the wave property of light (thus, source independence) the PoR still applies to it (as was found by M-M): the return-speed of light is independent of the intertial frame of reference. SRT is based on these simple principles, as the Lorentz transformations can be derived from them. Some modern derivations use only the first postulate, but in such derivations the constant in the LT is undefined. Harald88 06:45, 16 May 2006 (UTC)

Harald, please see my reply to you in the "Which of Einstein's versions ..." section above.

Also, please note that it was not my version of the 2nd postulate, but was Einstein's. (I merely paraphrased it).

Furthermore, please note that there has been no real test of SR because the only real test is to test the real 2nd postulate. Your cited source (#Status) mentioned things such as round-trip cases and time dilation, etc. SR's effects such as time dilation do not pertain to physical clock slowing but merely to observer-dependent "effects" due to the use of absolutely asynchronous clocks.

REF: [From Tom Roberts, April 30 2006] When you look at a building from directly in front, it appears to be wider than when you look at it from a corner. This effect does not affect the building itself, of course, and is purely due to your point of view. And because this is purely a geometrical effect, it would not matter if the front corners of the building were replaced with markers, there being no object between them. Similarly in SR, "length contraction" is purely a geometrical effect, an artifact of one's point of view, and it does not matter if there is an object between the endpoints, or if there are just markers with no object between.

http://groups.google.com/group/sci.physics.relativity/browse_frm/thread/69494989e198f9b0/fa366ff94909a8ad#fa366ff94909a8ad

Tom continued with this: Because the rod is moving, naturally the observer in A must mark both ends _simultaneously_ in frame A (which requires assistants) and then use a meterstick at rest in frame A to measure the length between the marks. Ultimately it is the difference in simultaneity between frames A and B that is the source of the length contraction (an observer at rest in frame B would say that the observer in A marked the front of the rod before the back end, so naturally the distance between marks in A is shorter than the rod at rest in B).

If one used absolutely synchronous clocks, then there would be no relativistic "length contraction, "time dilation," or "mass increase," and light's one-way speed would be c±v, not c.

Since, contrary to your #Status, relativity has had zero success regarding real physical phenomena, the burden of proof is actually on folks such as you, as is always the case with any theory. Cadwgan Gedrych 19:34, 16 May 2006 (UTC)

Opinions from people in newsgroups have no say in Wikipedia, and the statement (by you?) that no effects would be detected if one chooses another convention is erroneous. In which peer reviewed paper did you read that? Harald88 22:30, 16 May 2006 (UTC)
I won't deny that the experimental proofs of SR as listed in the "Status" section are "observer dependent effects". However, no such effects are predicted by classical mechanics, while in relativity they are. Look at it this way: In classical mechanics, synchronized clocks should always stay synchronized, but in the Hafele-Keating experiment not only did they become unsynchronized, but did so in accord with the predictions of relativity theory. You see what you want to see, but exepriments like that have very much placed the burden of proof and of explanation on you. --EMS | Talk 21:04, 16 May 2006 (UTC)
P.S. I have gained some respect for you due to recent events, but you are still an anti-relativist arguing for your view in the discussion page of an article that is obliged to accept relativity theory. I reserve the right to stop responding at any time, and once again ask you to take your opinions elsewhere. --EMS | Talk 21:04, 16 May 2006 (UTC)
The Hafele-Keating experiment seems to say the experiment is invalid, due to invalid data-gathering and other technical reasons. --GangofOne 06:24, 17 May 2006 (UTC)
I agree with EMS except for his P.S. (and this is the second time that I point this out) ... ok let's try it with an example this time: the article flat earth is not obliged to accept "flat earth theory"! No such thing as "sympathetic point of view" in wikipedia; instead NPOV as based on peer reviewed literature. Harald88 22:30, 16 May 2006 (UTC)
To quote WP:NPOV:
If a viewpoint is held by an extremely small (or vastly limited) minority, it doesn't belong in Wikipedia.
Harald88 - Cadwgan's view is part of such a minority. I also kindly submit to you that the peer reviewd literature overwhelmingly supports special relativity as an established and true theory. Those are the reasons why I say that this article is obliged to accept special relativity. At the least, even a "flat earth" article is obliged to express the theory in its own terms, even if there is also an obligation to describe why the mainstream view disregards it. --EMS | Talk 23:43, 16 May 2006 (UTC)
We need not to pull in the minority view/undue weight clause. IMHO it is best to point to the Bogdanov affair precedent: Editors who are only coming to Wikipedia to continue a controversy started elsewhere are not welcome. Editing only one (or nearly only) article can be seen as evidence for this. --Pjacobi 07:09, 17 May 2006 (UTC)
Kindly show me where Wikipedia policy explicitly states this. (I do think that this is a good idea for a policy, but if it does not exist as such I cannot in good faith use it as such.) Otherwise the minority view/undue weight clause has much the same effect. --EMS | Talk 14:34, 17 May 2006 (UTC)
It only exists in so far, as you accept the meta-poliy, thatArbCom rulings set precedents. In my own interpretation, it is widely accepted on warnings level, but if someone chooses to ignore the warning, ArbCom must be involved again.
Note that each ArbCom ruling includes a section, which general principles apply. This would be in our case:
Pjacobi 16:49, 17 May 2006 (UTC)
I am loathe to bring ArbCom down on Cadwgan at this time. If nothing else he has at least twice made positive contributions to this discussion through his excellent knowledge of the postulates of SR. For that reason I would prefer that he control himself and his urge to raise inappropriate questions here. However, it is nice to know that this case and the related precedents exist if we should need to use them. --EMS | Talk 18:08, 17 May 2006 (UTC)

Thanks for the kind words, EMS. As for my agenda, actually, I am simply pushing for "truth in advertising," not some pet point of view. I see no need to control myself given this agenda, but I have been, and will continue to be, very respectful, unlike a certain Dirk person, who was supposed to help in the "case against me," but that backfired (as it well should have because I am not pushing anything here but the truth).

The article's second postulate section still mentions measuring light's speed, but Einstein never mentioned doing this. What he did mention was presetting clocks to obtain a prechosen value "c." (This is given explicitly in John Wheeler's book Spacetime Physics, in case you need a ref.) (Of course, it was also stated by Einstein, but not as clearly as by Wheeler.) (Think about it: If the travel "time" is given up front, then there is no actual speed measurement being made.)

Do you honestly think the article is OK in this case? (And this is not the only major problem, as I have tried to point out.) Cadwgan Gedrych 19:56, 17 May 2006 (UTC)

I thank you for being respectful and considerate in your words. I just wish that I could get you to extend that to the content, as that is what makes you somewhat distruptive here. While I understand that these remarks reflect your honest (but mistaken) opinion, you are none-the-less asking me to debate the theory here as opposed to how the theory should be described in the article. The former is not what this page is intended for. --EMS | Talk 20:21, 17 May 2006 (UTC)

As you should know, Einstein’s definition was given ‘’prior to’’ his second postulate.

As you should know, the second postulate cannot exist without the definition.

As you should know, and I have said, the "time" portion of light's one-way "speed" was given by definition (again, prior to the 2nd postulate.)

And Einstein's quotes around "time" indicate that it is not an actual time measured by actual clocks, but is merely a stipulated time.

But none of this is mentioned in the article's 2nd postulate section even though it was directly given in Einstein's SR paper.

In fact, the article's 2nd postulate section does not even mention a definition, and yet this is critical to the very existence of the postulate. You may not see this as the serious sin of omission that it is, but surely someone here can. Cadwgan Gedrych 19:14, 19 May 2006 (UTC)

Cadwgan - It's "time". I will not debate you. I will thank you for making some astute remarks that showed up the errors of others, but Wikipedia is not as soapbox. You cannot keep trumpeting your anti-relativity here. What I will say (if only in appereciation of your assistance in improving this article be it intended or not) is that you do know the real answer to these issues, even if you do not know that you know it. --EMS | Talk 03:41, 20 May 2006 (UTC)

What you mean is that you cannot debate me. And I challenge you to point out any "anti-relativity" remarks. See more on my TALK page. Cadwgan Gedrych 20:23, 22 May 2006 (UTC)