Unless I am missing something (and I could very well be), the light had to travel a different distance in r-p, v/s p-r. This would be due to the target moving toward the laser beam. A "snapshot" in space to record positions at t=laser, t=P/receive and t=R/receive should show this.

"... the light had to travel a different distance in r-p, v/s p-r."

Not according to Einstein - the new frame is inertial and all inertial frames are equivalent. The distance between the observers must still be 2L meters, like before and this can be easily checked by the observers. So where is the catch?

-J

Isn't this the Michelson-Morley experiment? How does the CMB affect your apparatus? In other words, move the universe and leave your apparatus fixed since motion is relative.

"... move the universe and leave your apparatus fixed since motion is relative."

Problem is, the apparatus was accelerated, not the 'outside world' (or CMB, for that matter), so it is not a symmetrical situation.

-J

I missed that it was still under acceleration when the second test was done.

If acceleration is equivalent to a gravitational field per General Relativity, then wouldn't the clocks in the "down" direction tick slower?

Not correct. I wrote: "Next, gently accelerate the whole structure in the direction p->r until it has a constant speed v relative to the CMB, as measured by the change in CMB redshift."

It means it is coasting inertially again.

Ok, during the acceleration phase, the "down side" clock slowed down, so your times no longer agree. So you can't compare the start time with the end time because the clocks are different, even though they're now in free fall and running at the same rate.

That's pretty close, but it's not quite correct to say that the "down side" clock slowed down - acceleration without gravity can be shown not to change the rate of clocks. It has been done in centrifuges, where the slow-down was directly related to the tangential velocity of the clocks relative to the non-rotating frame, not the g's that they suffered.

A further clue: how would one know that the "clocks are different" after the acceleration?

Rixter, I neglected this question of yours: "Isn't this the Michelson-Morley experiment?"

No, the M-M was a two-way test; this is one-way only.

-J

It is counterintuitive to imagine that a photon would take the same time to strike an object traveling toward it, as an object traveling away from it.

"It is counterintuitive to imagine that a photon would take the same time to strike an object traveling toward it, as an object traveling away from it."

Precisely, with the operative words "the same time". But then, the photon source is traveling at exactly the same velocity as the target here...

-J

That sounds right.

If velocity v is sufficent to cause any change to the result, it will also be causing time dilation, which will 'counteract' any 'change' in c. Each observer will see the experiment run exactly the same in both 'static' and 'moving' conditions.

Let us take two observers, one standing on a planet that is 'static to the Cosmic background radiation,' the other on a rocket traveling at an appreciable fraction of c. each is holding a yardstick out so they can measure each others stick to see which one is compressing due to acceleration. They measure each other's sticks at the same instant as they pass. Which yardstick is shorter?

BOTH are shorter than the other stick. Each observer measures the other's stick as being shorter than his own. The constant c is a constant for a reason, it shows the relationship between time, space, and relative motion. You want to find the exact center of the observable universe? You're sitting on it. I'm sitting on it. Each member of the Vulcan Senate is sitting on it. Yoda is sitting on it. The exact center of the observable universe is WHERE THE OBSERVER IS, no matter which observer you are talking about. What's outside the observable universe? Who can tell? We can't see it, can't feel it, can't even be influenced by its gravity, for all we can tell there is NOTHING out there, not matter, not energy, not even emptiness. And if there IS a 'something' out there, just past where we can observe, a galaxy, with a solar system, with a planet, with an observer on it looking at the observable universe, WE don't exist to HIM, WE are out in the 'unknown, unknowable NOTHING.'

It seems to me the 'problem' is that you are switching inertial coordinate systems between an outside observer that sees the travelers, and the inertial coordinate system of the travelers themselves. The 'v' you use is with respect to an outside observer at rest within the CMB framework. Observers p and r are not moving at v with respect to each other, so it is not correct to add that term into their calculations for the light travel times. They should not be recording the light travel time as 2L/(c+v) or 2L/(c-v).

"The 'v' you use is with respect to an outside observer at rest within the CMB framework."

True, but the 4 observers also directly measured that 'v' by looking at the CMB redshift (actually, the dipole that their movement created). They also timed the light travel delay by using their respective clocks and the results were 2L/(c+v) and 2L/(c-v) seconds in the two directions in question. Honestly.

-J

I have most difficulty with this. e.g.

If you see the set-up as being inside a container, then the "whole lot" is moving
"as one," i.e. there will be no apparent difference (inside) at all, to the speed.

Now, remove the container, and say the set-up speed is, e.g. half the speed of light,
why is it not true, that the targets are moving away, and towards? (respectively)

Importantly, at what time, effect or appearance does the "container" become "open"?
I.e. to have a real increase in speed in relation to the targets? Or, is the light
inextricably bound to, and relative to, the targets, in or out of any container?

i.e. by removing a container have we just put it all (one lot) into a bigger container?

Only want the armchair in a dark room now... think on it forever. (Jorrie help!!)

jt.

"Now, remove the container, and say the set-up speed is, e.g. half the speed of light, why is it not true, that the targets are moving away, and towards? (respectively)"

Being inside a container is in principle no different than being without one. There are in any case no targets that are moving relative to each other here. All observers were initially static relative to the CMB, but after the acceleration they move in the same way, as far as they are concerned.

If there was an atmosphere of some sorts inside the container, it may obviously make a difference if the container is there or not there...

-J

Hi Jorrie,

We have an 'axiom' in electronics: "If you haven't tried it, it doesn't work." You may have answered this before in prior threads, but I will ask it again. Has this been tried (proven)? Also I am wondering what prompted this thread. Have you been intimidated by the pest?

I'm not quite sure what you are trying to say here. Are you saying that the speed of light is different if there is acceleration involved?

-S

Hi S, you wrote: "If you haven't tried it, it doesn't work."

Only 'tested' in thought experiments, AFAIK. Such a clock sync device has probably never been used - this should give a clue to the theoretical resolution of the puzzle...

"Have you been intimidated by the pest?"

Not the 'pest', some other Blog correspondent. Sure, the acceleration plays a role, but we are testing before and after, where it does not influence the speed of light per se.

-J

PS: Sorry for keeping up the 'teaser status'; it contains physics that many people understand, but have not quite thought through. This is supposed to help with that process.

So, have we shown that the speed of light is not isotropic and that the one-way speed depends on the laboratory's velocity relative to the CMB?

You haven't "shown" anything (but I trust your formulas).

Who is we? Do you have a frog in your pocket?

Was Einstein mistaken?

About what?

"Who is we? Do you have a frog in your pocket?"

Ah, yes, should have introduced the crew at p, q, r and s; most impolite of me. We are Pest, Quest, Ric and Steve (the latter also looking after the standards).

I think your puzzle can be resolved as follows.

In the static case the light travelled from its source r to p a distance of 2L in 2L/c seconds. In the moving case the light travelled from its source r to the receding position of p some time later, a distance greater than 2L that would take light a longer flight time. In each second light only makes (c-v) progress toward the receding position of p, the distance r to p is always 2L so the time taken is 2L/(c-v).

The reverse transmission of light takes 2L/(c+v) seconds following similar logic.

Putting this another way we have two interesting inertial frames of reference, one static for both experiments and one co-moving with the second experiment for both experiments. Where the frame or the objects move time, distance and simultaneity are relative creating new effects along the axis of movement.

From the static frame in the first case the apparatus is static and light travels from r to p, in the second case the apparatus moves to the right so light has to travel from r (or rather where r was) to a point past the starting point of p because p is moving away from the light, the distance is further and takes longer.

From the moving frame in the first case the apparatus is moving to the left so light travels from the initial position of r to a mid position because p is moving toward the light so the distance is shorter and takes less time than the second case where light would travel from r to p (neither of which move in the co-moving frame). The events of transmitting light are no longer witnessed as simultaneous from different positions long the axis of movement. So p and r record different transmission times for each other's light but can later review each other's experimental log and reconcile their different perspectives using relativity.

Both frames produce identical results and light only ever travels at c as Einstein prediced.

Cedar: "The events of transmitting light are no longer witnessed as simultaneous from different positions long the axis of movement."

GA

You've made heavy work of it, but it is correct. Although acceleration does not influence clocks, it influences the synchronization of clocks located along the axis of the acceleration. Those clocks need to be resynchronized after the acceleration by means of EM waves - only then will the speed of light again be observed as the same in both directions. Which is of course a self-fulfilling prediction. In fact it is only the two-way speed of light that is isotropic in an absolute sense. One-way isotropy is only so by Einstein's clock sync convention.

Interestingly, if the pellet gun of the OP were to be used for such a resynchronization, the one-way anisotropy of light would have remained. We simply should not use things with mass to synchronize clocks. Light is a strange beast...

-J

Hi Jorry,

Just for fun....

You have contrasted the behaviour of light and particles brilliantly in this example. You have teased us now can you answer a related question?

I have always been intrigued by the contrast between light and gravity waves. They are said to travel at the same speed but light speed is held to be independent of any medium while it seems to me that gravity waves must travel through the medium of spacetime. How could gravity waves travel at the same speed with respect to a body moving through space in the way light does in the Michelson-Morley experiment? Gravity waves are waves caused by changes in the local distortion of spacetime travelling through spacetime to update the distortion of remote spacetime, surely they must travel in the medium and be governed by the movement of spacetime just as sound is governed by the movement of air.

If at some day in the future the observers in your experiment had gravity wave generators and detectors what would they see? Would the gravity waves reveal the movement through a Doppler effect just as sound would in air?

(For those engineers who would like an outline design to understand the question I would propose the gravity wave generator might be two masses spinning like a watt governor for steam engines and the detector would be an array of hollow balls with a solid ball in the middle. The contrast of behaviour would isolate kinematic forces and in the freefall of your experiment these would be the gravity waves.)

Hi Cedar, you wrote: "Gravity waves are waves caused by changes in the local distortion of spacetime travelling through spacetime to update the distortion of remote spacetime, surely they must travel in the medium and be governed by the movement of spacetime just as sound is governed by the movement of air."

Likewise, EM waves are changes in the local vacuum, traveling through spacetime...

AFAIK, general relativity predicts that they will have the same inherent properties. If represented by particles, gravitational waves^[1] are surely without any rest mass, like photons and unlike my 'pellets'. We must not forget that light is also affected by the medium, e.g. bending, compression of wavelengths, Shapiro time delay, etc.

This brings us to the crux of the teaser: who says that our movement relative to the medium does not affect light? Why do we have to re-sync our clocks if we change inertial frames? In some subtle way, light must be affected, despite the null MM result. Einstein reportedly said: if nature conspires against us like this (making the aether undetectable), we can just as well forget about it. My best guess is that if it was possible, the MM-experiment would also have given a null result with gravitational waves, while with particles like electrons, it would not.

-J

(1) 'Gravity waves' are traditionally something else, operating in the atmosphere, I think.

Thanks for the correction, I did not know the distinction between Gravity and Gravitational waves and from the web it seems I was not alone!

You quoted Einstein: "if nature conspires against us like this (making the aether undetectable), we can just as well forget about it." Many people seem to treat denial of the existence of an aether as an article of faith rather than a pragmatic thought short cut. It is held that the MM experiment proves no medium can exist. But the MM experiment formed its conclusion using coordinate geometry, or to be pedantic rigid Euclidian coordinate geometry. This can only model the behaviour of a rigid Euclidian aether (the type people expected at the time) which is indeed excluded, but it cannot match and hence cannot exclude Einstein's wibbly-wobbly spacetime.

"But the MM experiment formed its conclusion using coordinate geometry, or to be pedantic rigid Euclidian coordinate geometry."

Not quite. MM used an L-shaped apparatus which was roughly orthogonal and had roughly equal length arms. It did not care about coordinates systems, AFAIK. As long as light traveled those two arms (away and back) and produced an interference pattern at one point, it was sufficient to show that the two-way speed of light did not depend on the direction of the arms. This was an absolute fact, not coordinate-dependent.

This was also the only conclusion that M&M have drawn AFAIK. People like Lorentz and Einstein interpreted it in terms of the aether's existence, or not...

-J

Hi Jorrie,

You said "It did not care about coordinates systems, AFAIK" and "This was an absolute fact, not coordinate-dependent."

I absolutely agree - but you misunderstood me.

I did not say coordinates were used but "coordinate geometry" - the relevant field of maths. I am sure you know the story. Maxwell had thought the Earth's movement in the one-way speed of light might never me detected because it was too small and one-way measurements of "c" are damn near impossible. The MM experiment was designed using (rigid Euclidian) coordinate geometry to prove a contrast between unequal return paths through the medium along the "aether flow" and marginally longer diagonal paths across the "flow". The maths proved this small contrast could be reliably detected by an interferometer as the apparatus was rotated. So I think my statement "MM experiment formed its conclusion using coordinate geometry" is just about valid and so is yours that it is not dependent upon particular coordinates.

The MM experiment cannot eliminate aether drag (the aether sticks to the Earth and so doesn't move in the lab), Lorentz contraction of the apparatus in the direction of travel (which would eliminate path differences) or alterations in the medium along the direction of motion which I was discussing (i.e. the medium is not rigid and/or Euclidian).

For you amusement…

I doubt M&M formed any new conclusions at all from their famous experiment in 1887 as it was an improvement on Albert Michelson's 1881 experiment that had already given a null result (but not eliminated enough error to convince sceptics).

Kind regards

Cedar