consistency of the speed of light

Aether

I don't think so, Hans. We actually perceive only dimensionless ratios (http://arxiv.org/abs/hep-th/0208093), and all judgements of simultaneity ultimately depend on one's choice of clock synchronization convention.

Suppose that you are right, doesn't this "simple rule" amount to the choice of Einstein's clock synchronization convention?

Tom Mattson

How's that? Right now, sitting at my desk, I've just observed someone tying his shoe, followed by someone else clicking a mouse. Why do I need a clock synchronization scheme to say which one comes first? And if some other observer goes zipping by at high speed and observes the events in reverse order, why would he need a clock synchronization scheme to say the opposite?

Hans de Vries

In fact, The Lorentz transformations them self can be derived directly from
the single statement that "A wavefront is always at right angles with the
direction of the wave in all reference frames."

(This beauty already laid unrecognized in the M&M experiment)

The example I use here (with the Broglie wave functions) works at any speed
let is be at meters per second or centimeters per second.


Regards, Hans

Aether

Lorentz transformations amount to an arbitrary choice of Einstein's clock synchronization convention. Lorentz symmetry is equally well represented by transformations in which absolute simultaneity is maintained.

Hans de Vries

Imagine,

Looking at one of the images in my post above sitting right in front of it.
Take a picture from sufficiently far away. The photo will group together all
the space/time points which have equal time in the particular reference frame.
There's no choice here.


Regards, Hans

PS: For two points, say one left and one right of the center, having the same
distance to the viewer. You can take a picture from any distance and you'll
always fetch the same two space/time points togeter. Even though you have
c going at all kinds of angles, with all kinds of different ratios of its x and y
components.

Aether

How do you know that these two points have the same distance to the viewer? What are you assuming about the speed of light? Aren't you assuming that its speed is constant and isotropic?

Aether

You interpreted what you observed using a set of assumptions including that the speed of light is very fast, and that it is isotropic; right? That is where you adopted Einstein's clock synchronization convention. If you didn't use these assumptions, exactly which ones did you use? If I postulate that the speed of light is a function of the cosine of the angle that your velocity makes with some other velocity, say the CMB rest frame, then clocks are synchronized absolutely and judgments of simultaneity are generally different. You can transform freely between Einstein clock synchronization and absolute clock synchronization, and in doing so you change the directional speed of light.

A dimensionless ratio can not be freely transformed, and is the only truly invariant outcome for any measurement. For example, say that I measure my desk to be 2 meters wide; where is the dimensionless ratio? 2 meters per 1 meter of my meter stick: the ratio of my desk's width to my meter stick's length is 2/1. In any frame, this ratio sticks.

The speed of light is not a dimensionless ratio, and therefore it is not something that you can ever measure without reference to one artificial convention or another!

"When clocks are synchronized according to the Einstein procedure the equality of the velocity of light in two opposite directions is trivial and cannot be the subject of an experiment." - M-S I p.499

pervect

OK, I've snagged the files (both sources), and will look them over when I get a chance. The .pdf definitely looks readable, I haven't tried the new .doc yet.

Aether

There are two things going on here, Hans. First there is Lorentz symmetry, and that is what is actually measurable in experiments. Galilean transformations fail those tests, and I am not implying otherwise.

The second thing that is going on is that Lorentz symmetry is conserved by at least two types of transforms: Lorentz transforms are one type, and they are characterized by postulating that the speed of light c is isotropic in all inertial frames (e.g., SR). However, there is a second type of transform (e.g., LET) where clocks and rods are isotropic in all inertial frames, but the speed of light is not. These two type of transforms are equally valid, and they have nothing whatsoever to do with any experimental result (so far). Interestingly, BOTH theories (SR & LET) would need to be modified upon the detection of a violation of Lorentz symmetry because they are empirically equivalent.

Violations of Lorentz symmetry are empirically measurable in principle, but the conventional choice as to which terms in the transform are constant and which terms are allowed to vary is not. So which one should we use? SR is convenient as long as you don't have a locally preferred frame to use as a "handle", but LET would be better if we ever find a handle.

It is wrong to claim that the constancy of the speed of light is proven by experiment. Does this go for GR as well? I suspect that it does, but I'm still studying GR myself and can't say for sure at this time.

pervect

If you look at Einstein's original paper, Einstein *assumed* isotropy, which he did not define in great detail, to arrive at his theory of relativity.

Mansouri and Sexyl, from the papers I skimmed (thanks for posting them) are basically exploring the realm of physics of "what happens if one does not assume isotropy", though they don't discuss it in those exact terms. (I personally think their paper would be improved if they did at least mention the term "isotropy", it's got a lot of history).

A perfectly isotropic space-time can be made to appear non-isotropic by the proper (or improper) choice of clock synchronziation methods. So what Mansouri and Sexyl are basically doing is to *not* automatically choice a coordinate system that conforms to the (apparent) isotropy of space-time, by instead considering arbitrary clock synchronizations.

By not assuming isotropy as a given, (as Einstein did), Mansouri and Sexyl's work helps provide a framework for testing it. (Of course I should add that there is currently no evidence that there is any physical aniosotropy in space-time).

Their (M&S) general approach may also be useful in rotating coordinate systems, where the usual assumption of isotropy has issues. While one can always chose not to use rotating coordiantes, they are convenient enough that sometimes it's worth giving up the conveniences of isotropy for the convenience of using rotating coordinates.

Didyoueatpaintchips

Zapper,
I believe what i said was correct. To throw out a universe with constant time and space, there had to be a reason. Every one accepted a constant time and space as fact. So it was the the MM experiment which led to SR, which led to variable space and variable time. If you have a constant velocity for the observer, then something else must be a variable, Einstein chose space and time. So the whole reason for coming up with SR is the MM experiment. Without the MM experiment, SR was not needed, and thus would never have been accepted in the scientific community.

And yes, relative to the constant speed of light, the observer's speed was 0.
Remember, everyone expected something other than 0, because they were looking for the so called aether. But the experiment was done in air, so relative to the air, the velocity of the interferometer was 0. Not a good experiment.
mike

JesseM

What difference does it make what its velocity relative to the air is? Neither relativity nor the aether theory would predict your velocity relative to the air would make any difference in terms of the velocity you measure for light.

And "relative to the constant speed of light, the observer's speed was 0" doesn't make sense. You can only talk about your velocity relative to a thing like a car or a light wave, talking about your speed relative to a speed is meaningless. What is my speed relative to the speed of 70 mph? Maybe the idea you're trying to express is that the experimenters originally believed that only an observer whose velocity relative to the aether was 0 would see light moving exactly at c in all directions, whereas relativity predicts that every observer will see that. But this means that every observer has a speed of c relative to light, not a speed of 0 relative to light.

DrChinese

Welcome to PhysicsForums, Mike.

This forum is intended primarily for mainstream discussions of physics topics. It is not intended for presentation of new theoretical ideas, which are normally presented in the context of peer-reviewed journals. Anyone is free to publish their ideas as they like on their own sites; that is what I do with my work (although most of my stuff is in fact mainstream).

And if you think this forum is devoted to Einstein, you haven't followed the discussions here sufficiently. There are skeptics here, but the presentation should be made within the context of legitimate topics. For example, SR and GR are currently generally accepted theories within the physics community. Therefore, a thread about why SR is wrong is not welcome here. On the other hand, EPR is considered to be outmoded (at least as to the incompleteness of QM) due to Bell's Theorem. Therefore, a critique of Einstein's position on this is acceptable.

Please keep in mind that there are many readers here with varying levels of knowledge. This forum is devoted to those interested in learning more about what is going on in physics and science.

...

Regarding your later post: I wonder why you don't feel the M/M experiment is good. Based on your historical description, how do you see that the speed of the air is an issue? Are you saying that the experiment would only be meaningful in a vacuum? Isn't the air itself also moving "faster" relative to the ether in the direction of the earth's movement?

Hans de Vries

You can do that by using a ruler That's the way we define length.
Under all conditions, even if you might question how we do know that the
ruler is always the same.

The example in the postscript shows that you get the same result under
various mixes of c_x and c_y components.



Regards, Hans

Hans de Vries

Aether,

All Lorentz transforms are equally well valid. They are just different
representations of the same, single, reality. Just different 3D slices from
the single 4D universe.

What are you looking for? A preferred reference frame? what about Einstein's
beloved Machian reference frame, the center of gravity of all mass in the
universe. Or the modern day cosmologist's co-moving frame. more like the
center of mass of the local universe, so there are different co-moving frames
at different places in the universe.

Are you looking for aether? like your name implies. The vacuum is far from
empty. Look at the vacuums from Quantum Field Theory, All the many
different vacua people discuss in Quantum Gravity research. Just never
call it aether. That's a name which is reserved for a substance with a
classical gas like behavior. It has be shown over and over that that's
not the way how it works with incredible accuracy.

Are you maybe looking for the absolute "NOW" ?
Why should two different space/time events separated billions of light years
from each other be connected instaneously by an invisible link? You may
define a mathematical space in any arbitrary way that does so but does
it matter to physics anymore then a statement that two different events
have the same value for x or y? All what matters to physics is how different
events in space time communicate with each other, and they do so from
neighbor point to neighbor point to neighbor point. Two events at different
sides of the universe are completely disconnected.

Special Relativity holds up an illussion of a "NOW" in each reference frame
but in General Relativity there's no global "NOW" any more in any reference
frame.


Regards, Hans

Aether

I'm working on a unification theory that makes predictions that are so precise and consistent that I tend to take them seriously, but a locally preferred frame seems to be implied. So, I raised the Aether flag and went forth to test the waters and found that LET is empirically equivalent to SR. Whether or not my own theory pans out, this empirical equivalence of LET and SR is something that I think people should be aware of.

Aether is used as shorthand for both a rarified gas (which is not how I am using it), and a locally preferred frame (which is how I am using it).

This is getting close to where I'm going, that events on opposite side of the universe are connected instantaneously. Every event has a temporal coordinate on the farthest edge of the universe, so why shouldn't events be connected out there? There is cosmological evidence of such a link (e.g., the "horizon problem").

And, 98% of the matter-energy in the universe is still missing.

Hans de Vries

Aether,

Mansouri & Sexl are plainly wrong with the claim that their Lorentz
Ether Theory is empirically equivalent to Special Relativity. There is only
one frame in which it makes some right predictions and that's the frame
in which they did their calculations: The preferred reference frame.
They then erroneously extrapolate that it does work in all the
reference frames.

In the other frames it gives results that are completely different than
those from Special Relativity. The LET is nothing else than a Galilean
transformation with a scaling factor (gamma)

A clear example in where it goes wrong is when you've got two objects,
one moving from left to right and the other from right to left, both with
the same speed. In SR both will have the same Lorentz contraction for
the observer at rest. In their LET however there is no such symmetry.
One object will typically get extended while the other gets contracted.
The only frame where LET and SR give equal results is in their preferred
reference frame.


It's only with the non-simultaneity of SR that two observers both
see each other contracted. With LET one observer will be contracted
and the other will be extended in an asymmetric way.

To see how SR works we can imagine that we instantaneously "freeze"
a bypassing traveler. Walking around him we can now see him "hanging
in the air", indeed being contracted in the direction in he was moving.
The traveler however will complain that his front was stopped first,
before his back was frozen, and argues that this is the reason of his
compressed state.

This now is a symmetric situation. If you'll lend you're "freezing device"
to the traveler then you'll see that you get frozen in the same way:
contracted by non-simultaneity.

With absolute simultaneity this will always lead to a paradox: If A is
contracted in respect with B. Then B is extended relative to A.
It is the non-simultaneity of SR which resolves this paradox.



Regards, Hans


P.S: R.Mansouri and R.U.Sexl
A Test Theory of Special Relativity, I. Simultaneity and Clock Synchronization
General Relativity and Gravitation, Vol 8, No 7 (1977) pp. 497-513