Subject: Feedback

Date: Sun, 06 Feb 2000 01:04:03 -0500

From: Nick Percival <NickP@SMNT.com>

Organization: Semantech

To: umberto bartocci <bartocci@dipmat.unipg.it>

I have reread your paper, "Most common misunderstanding about SR" carefully. My abbreviated thoughts are given below:

(Your) Section 1: I agree completely.

Section 2: On the one hand, I am in complete agreement with you about special relativistic phenomena being applicable to non-inertial frames as well as inertial frames. For example, when I have debated the Twin Paradox, some supporters of SR have contended that the net time difference accumulates due to the SR time dilation equation/effect as applied by the "stay-at-home" twin to the constant velocity parts of the round trip. When I ask why, for the constant velocity parts of the trip, the travelling twin could not apply the SR time dilation equation/effect to obtain a contrary result, the SR supporters contend that since the travelling twin accelerated, he can not use SR. Among other things, I claim, that such a defense is more damning than the original Twin Paradox attack - Since virtually all real world phenomena involve non-inertial frames, the SR supporters' defense of SR, if consistently applied, would hold that SR was NOT applicable to virtually any phenomenon.

On the other hand, I was taught at Harvard, using the standard text books of the time, that the "special" did, in fact, mean that SR applied just to inertial frames and that GR extended the relativity principle to non-inertial frames. I think that this is a generally accepted view even though at a minimum its inexact. Further, that (inexact) view is supported by Einstein's words as the Relativity Principles seem to explicitly restrict the theory to inertial frames. Also, Einstein's description of how he developed GR discusses extending the relativity principle as applied to inertial frames in SR to non-inerital frames. But I agree with you that this conventional view of "What does the 'special' mean?", is inaccurate and misleading. In my view, while SR does NOT describe effects that are due to acceleration per se, SR effects apply to non-inertial frames. Most of those effects are a function of (relative) velocity and if we view acceleration in terms of an integral of increasing/decreasing velocity, we can accurately describe the special relativistic effects for non-inertial frames - remembering that SR may NOT tell the whole story for non-inertial frames - e.g., effects due to acceleration per se are beyond the scope of SR. Hence, I agree with you that the REAL meaning of "special" is that SR applies to flat space-time and while effects due to space-time curvature/gravtitation are beyond the scope of SR, non-inertial frames are NOT beyond the scope of SR. Hence, your title, "Most common misunderstanding about SR", may not be strictly accurate as you focus on the misunderstanding of SR by its opponents, whereas many of the most common misunderstandings of SR are also held by SR's supporters.

[Note: My first paper described my "extension of SR". It was developed to give the "real" explanation of how the net time difference accumulates. Two unexpected by-products of my "extension of SR" were 1) it extends SR "upward" to curved space-time - my "extension of SR" can also be thought of as an extension of GR ("downward" into the domain of SR) - the net result is that SR and GR become a seamless theory - I gave an example of how SR would be extended to the space-time model that represents the Schwarzschild solution and 2) it unified gravitational potential energy and kinetic energy.]

Section 3: Excellent job in dealing with the base case.

However, I do NOT understand why your handling of the base case does NOT also handle the special case that you

raise, namely, "when R is very large". This is probably my fault for not being familiar with this aspect of the Sagnac

literature and NOT understanding the point being raised by others that would require that you give additional logic. (Note: Others may also not be familiar with why you would have to treat this special case.)

Section 4: I did not realize that this was a controversial point. I've only encountered this topic in the context of GR.

Section 5: I contend that when you modify the classic Twin Paradox scenario by using a "uniformly accelerated normalized" observer (instead of having the main portions of the round trip done at constant relative veelocity) that you add an unnecessary element of complexity and confusion that tends to mask the issue raised by Langvin, Dingle and others. The classic Twin Paradox debate goes something like this: Since, in the classic Twin Paradox scenario, the period(s) of acceleration can be made to be of arbitrarily short duration compared to the constant velocity parts of the round trip, its fairly easy to show that the net time difference must be explained in terms of the constant velocity parts of the round trip. Yet, during those constant velocity parts of the trip, the physics should be identical for both twins, according to the relativity principle. However, if one deals with a "uniformly accelerated normalized" observer, this logic argument is circumvented - now the "travelling" twin is clearly in a qualitatively different state than the stay-at-home twin for the entire round trip - the neat little paradox has NOT been resolved, but rather it has been avoided. The classic Twin Paradox has proved complex enough, subtle enough and sophisticated enough such that, for nearly a century, little progress has been made despite both sides having some of the smartest minds in the world. To add another level of complexity and in this case a level of "indirection" just serves to make matters less clear and facilitates the combatants talking at cross purposes. Interestingly enough, even in your modified version, the paradox remains, but it is now deeply obscured and almost impossible to demonstrate to a skeptic.

Section 6: Very well done. Very instructive.

Section 7: Again - Very well done. Very instructive. Note: S.J. Prokhovnik wrote a paper explaining aberration in terms of a space-time model where light traveled in a preferred frame (not necessarily an aether frame).

Section 8: Excellent piece.

Let me state my thesis once again in the context of your remarks here and in your footnote 32. I contend that SR is correct as shown to great precision by a mountain of experiments and by its being one of the theoretical foundation blocks of all modern physics. I also contend that SR's opponents have made many VALID points (e.g., the points you make here). However, I contend that these "problems" with SR point not to its inconsistentcies, but rather to its incompleteness or limited scope. For example, when Einstein says that the laws of physics are invariant with respect to all inertial frames, I read that as saying that the laws of physics that describe how phenomena will be observed are invariant with respect to all inertial frames. I believe that my added words are consistent with the "modern" view of physics. However, I contend that IN ADDITION to the above quoted invariance, there is a physical space-time model where there is a unique frame and that for virtually all phenomena that physical space-time model is hidden. I contend that the net time difference in the Twin Paradox is one of those rare instances where the physical space-time model shows itself. The problem is that the net time difference can also be "explained" by misusing standard SR. It is my contention that had Einstein done a complete job and revealed the full SR, then there would be no Twin Paradox debate and that ironically anyone who tried to explain the net time difference using the methodolgies of the past 90 years would be held in the same regard as Dingle is now held. Further, my view is that even if one develops an experiment that shows an anisotropy in the speed of light that result, by itself, does not necessarily invalidate SR. Also, if one demostrates a relativistic effect that is a function of velocity wrt the electromagnetic field and that is not predicted by SR per se, that, by itself, does NOT proved SR to be wrong - just incomplete.

[My latest paper shows that for a wide range of phenomena that the physical space-time model of my "extension of SR" yields predictions that are in exact agreement with SR - where the two models address the same phenomena.]

I will next view your 2nd paper.

* * * * *

Dear Percival,

here I am at last with my comments to your mail of 6th Feb.. I shall mostly confine myself to the points in which I do not agree with you.

> When I ask why, for the constant velocity parts of the trip, the travelling twin could not apply the SR time dilation equation/effect to obtain a contrary result, the SR supporters contend that since the travelling twin accelerated, he can not use SR.

- You seem to me wrong in the first assertion, the effect is a GLOBAL effect, and it is not possible to "split" it "for the constant velocity part". If the twin comes back, then THERE MUST BE an acceleration somewhere, and that is all. If we want to discuss relativistic time dilations due to CONSTANT VELOCITIES, we must discuss examples of the kind I have discussed at the end of my last mail. "The SR supporters" are wrong in their assertion, the travelling twin CAN use SR, but he cannot use Lorentz coordinates; physicists seem unable to use general coordinates in a manifold (even a flat manifold as Minkowski space-time is)...

> On the other hand, I was taught at Harvard, using the standard text books of the time, that the "special" did, in fact, mean that SR applied just to inertial frames and that GR extended the relativity principle to non-inertial frames.

This is wrong, of course not in the sense that your historical witness is false, but in the sense that this widespread opinion is wrong.

> Also, Einstein's description of how he developed GR discusses extending the relativity principle as applied to inertial frames in SR to non-inertial frames.

This is correct, mut one must not confuse the historical genesis of a theory, or how a theory has been intended (or understood) by his author, with the FINAL THEORY, which is the result of MANY contributions of different clever minds (I share the opinion of Pyenson, that relativity has mostly been constructed by mathematicians like Minkowski, Hilbert, Levi-Civita, Weyl, etc.).

> SR may NOT tell the whole story for non-inertial frames - e.g., effects due to acceleration per se are beyond the scope of SR.

This is not correct in my opinion, ALL answers which involve Minkowski space-time (namely, physics in absence of gravity), can be answered in this very same frame.

> Hence, your title, "Most common misunderstanding about SR", may not be strictly accurate as you focus on the misunderstanding of SR by its opponents, whereas many of the most common misunderstandings of SR are also held by SR's supporters.

This is completely correct.

> However, I do NOT understand why your handling of the base case does NOT also handle the special case that you raise, namely, "when R is very large"...

I wrote: "in order to avoid such complications, one gets further with naive arguments, saying that, even if alpha is not an inertial observer, he would become such when R is "very large"". This is almost the same topic we discussed before. The most of physicists I know seem not willing to use the strict RIGOUR which mathematics require: when one object to their arguments saying for instance that they CANNOT apply some proposition, because there is an hypothesis in the theorem they wish to use which is NOT satisfied in their case (for instance, the ESSENTIAL presence of an acceleration, which is incompatible with the use of Lorentz coordinates), then they try to avoid the objection putting themselves in a context in which there SEEMS TO BE a possible GOOD APPROXIMATED RESULT from a WRONG argument. In the Sagnac case, it is true that, when R is very large, alpha can be considered approximately inertial, but then so only for a SMALL neighborhood of the space-time (namely, for a small space and a small time), while the Sagnac effect is a GLOBAL effect: the light must travel along the WHOLE rim of the rotating platform, before the effect shows itself. I hope that this words would answer even to your quite correct assertion:

> Since virtually all real world phenomena involve non-inertial frames, the SR supporters' defense of SR, if consistently applied, would hold that SR was NOT applicable to virtually any phenomenon.

This is quite true, but inessential. SR is an IDEAL mathematical theory, which can be APPLIED in Nature when it is possible, with good APPROXIMATED experimental results (or, at least, the greatest majority of today physicists say so). But to understand WHEN it can be applied in order to give good results, or not, it does require a deep understanding of the theory, which does not seem very common. For instance, I did not discuss Michelson-Gale effect, which can of course be predicted by SR (and not GR!), but some of my anti-relativistic friends are continuing to think that this is an effect disproving SR, like Sagnac!

> "I contend that when you modify the classic Twin Paradox scenario by using a "uniformly accelerated normalized" observer (instead of having the main portions of the round trip done at constant relative veelocity) that you add an unnecessary element of complexity and confusion..."

Of course I do not agree. The main question is: since there must be an acceleration as a physical reason for the twin coming back, where do you want to HIDE this acceleration? I showed that the must correct thing to do, from a mathematician point of view, is to uniformously distribute this acceleration along the whole trip, I know very well that physicists rather prefer to hide it in a SINGLE POINT (an "infinitesimal" time instant!). I hope you appreciated the mathematical fact that the curvature of the hyperbolic branch does approach INFINITY, whatever the angle of the two straight parts of the travel is.

> The classic Twin Paradox debate goes something like this: Since, in the classic Twin Paradox scenario, the period(s) of acceleration can be made to be of arbitrarily short duration compared to the constant velocity parts of the round trip, it is fairly easy to show that the net time difference must be explained in terms of the constant velocity parts of the round trip".

What you say "it is fairly easy to show" is in fact wrong. Don't forget that in my example - just an example, one can think of many others "smoothings" of the vertex singularity - when g is very large, the "travelling twin" DOES indeed travel to an almost constant velocity, with zero ordinary acceleration, during the "greatest part" of the travel! The quantity which has a physical meaning in SR is NOT the ordinary acceleration, but the 4-acceleration vector, in my case just a 2-vector, and its module. I hope I have shown to you with good computations (I hope to have not made mistakes!) what is the correct relation between the parameter g, the speed k, and the module of this 2-acceleration. I have said more about that in my previous answer to your "proposal".

> the neat little paradox has NOT been resolved, but rather it has been avoided.

In some sense I do agree: as a matter of fact, I claim that there is NO twin paradox at all. I repeat that the only "paradox" is to understand time dilation (or length contraction) in SR for INERTIAL observers (not twins - inertial observers can meet just once!), as an effect due for instance to Lorentz trasnformations.

> The classic Twin Paradox has proved complex enough, subtle enough and sophisticated enough such that, for nearly a century, little progress has been made despite both sides having some of the smartest minds in the world.

As I already sayed, sociological or historical remarks are interesting, but does not constitute "science". Let me add that the "smartest minds" you refer to, have given exactly the same answer to the paradox I did before: I was in no sense ORIGINAL, all "good" people gave the same reply, Dingle was unfortunately quite wrong, and his opponents were quite right, there is not possibility of any revision of the "sentence". I say unfortunately, because to fight SR with wrong arguments is damaging for the whole fight against this theory, which cannot be but correct from an ideal, mathematician, point of view, but I hope that it is wrong from the experimental point of view - but of course even experiments do require a theoretical analysis, or interpretation, and a comparison with SR predictions. Once again one must understand correctly what are these predictions, and correctly remark a possible difference between an experimental outcome and the relativistic expectations. This is why I rather prefer experiments of the kind "there is effect/there is no effect", which of course work only when the effect is not very small, and there no other physical causes hiding, or confusing, it…

> Interestingly enough, even in your modified version, the paradox remains, but it is now deeply obscured and almost impossible to demonstrate to a skeptic.

Of course I cannot agree; as far as the "skeptic" is concerning, before discussing any argument one must know some essential features of that argument. I cannot discuss questions concerning the exactness or not of a translation of a Qumran manuscript from old aramaic...

> I contend that SR is correct as shown to great precision by a mountain of experiments and by its being one of the theoretical foundation blocks of all modern physics. I also contend that SR's opponents have made many VALID points (e.g., the points you make here). However, I contend that these "problems" with SR point not to its inconsistentcies, but rather to its incompleteness or limited scope.

I do agree, if you intend that the experimental trials of the theory are not complete, in the sense that one can think to many others of them. As a matter of fact, I always claim that there is no one real DIRECT prove that the principle of relativity is true for optical or e.m. phenomena, do you know of anything available? (I believe of course that this principle is not true even for mechanical phenomena, but here in the Earth this prove would be quite impossible). I do not agree if you mean that SR is incomplete as a theory; it is incomplete just as ANY mathematical theory, as for instance Arithmetics, or Euclidean Geometry, is incomplete from the point of view of modern logic (Godel's theorem), but no more than that (and that is a VERY SUBTLE point, which most of the common divulgations necessarily misrepresent).

> Further, my view is that even if one develops an experiment that shows an anisotropy in the speed of light that result, by itself, does not necessarily invalidate SR. Also, if one demostrates a relativistic effect that is a function of velocity wrt the electromagnetic field and that is not predicted by SR per se, that, by itself, does NOT proved SR to be wrong - just incomplete.

Here I cannot understand you at all. In mathematics (as in ethics), white is white and black is black, as yes is yes and not is not. If you have some physical effect relative to some observers, and different physical effects of the SAME phenomenon, for an observer moving with relative uniform speed with respect to the first one, then SR IS irrimediably WRONG, unless one can find a reason for that experimental divergence in the approximations which cannot but exist in connection with the IDEAL case. But this "game" of finding reasons, should always be a FAIRY game, and one cannot introduce ad hoc explanations at the infinity! I wish to quote to you the words that Einstein himself sayed as far as the famous Miller's experiments was concerning:

"If Dr Miller's results should be confirmed, then the special relativity theory, and with it the general theory in its present form, fails. Experiment is the supreme judge" (Science, about 1925, I have no time to check).

Today, after Shankland's analysis (1955), Miller is considered wrong, but if one would reproduce Miller's results, avoiding the "mistakes" that Shankland pointed out as a possible cause of systematic experimental error, then things would change completely…

I say it once again, the plain truth is that we do not have (almost) any experiment which has been performed comparing measures in TWO different (almost) inertial frames, one moving with respect to the other. One exception is the famous Hafele-Keating experiment, but A.G. Kelly ("A New Theory on the Behavior of Light", The Institution of Engineers of Ireland, Monograph N. 2, 1996, p. 8) interestingly remarks - quoting from USNO, Hafele, 1971 - that Hafele himself acknowledged that:

"Most people (including myself) would be reluctant to agree that the time gained by any one of these clocks is indicative of anything … The difference between theory and measurement is disturbing".

Did you know that?

For the remainder, we have only indirect proves, under the assumption that the Earth itself is an (almost) inertial frame, and that experiments carried on in different times on the Earth can be interpreted in some sense as experiments of the kind I claim necessary, but this interpretation is of course quite wrong (the fact that there are possibly no other good interpretations, namely good enough aether theories, or absolute space theries, in which the relative velocity Earth-aether, at least at the Earth's surface, is almost zero, does not mean anything in favour of SR!). Another case of INCOMPLETENESS OF OUR EXPERIMENTAL KNOWLEDGE, is that we never performed measures of the light speed in the far space, at a great distance from any star (some people believes that light speed DOES depend on some physical features of the "empty space", like density and so on, and that these features can change dramatically from one point of the space to the other - one could thing even to artificially modify these features of the space in a terrestrial laboratory, due to electric or magnetic stresses).

Goodbye, I hope you will appreciate at least my effort in commenting, and trying to frankly communicate (in my very bad English!).

[Don't forget that I am waiting for your contribution to my project of book "Putting SR to test", the only hope one has to defeat SR, and to show that ordinary space-time categories are enough in order to explain natural phenomena! (of course, this is just a "philosophical" issue, and one could be of a different opinion; but in this case, I simply do not understand why one should be so eager to change somethings in SR: if one is happy about its theoretical foundations, then things go well enough with relativity, at least until now, and in relation to our INCOMPLETE experience...)]

Best regards, from yours most sincerely

Umberto Bartocci