Can light accelerate in vacuum?

[Isaac0427]

I think these answers may be confusing to the OP. Considering the prefix (B) on the thread, I would assume the question is "Can you apply a force that would increase the speed of a photon?" Also keeping in mind the prefix, I think we can assume inertial coordinates to answer this question.

 

[davenn]

" Light being emitted from a source in vacuum, DO photons accelerate ?"
the answer emphatically is "Yes".

Jon, That is very incorrect
photons are emitted at the speed of light, they DO NOT accelerate to that speed
or to put it the other way ... the EM field is emitted at the speed of light, it does not accelerate to that speed
D

 

[PeterDonis]

I would assume the question is "Can you apply a force that would increase the speed of a photon?"

That still doesn't pin down a specific question, or capture actual physics, because "speed" depends on the coordinates you choose.

I think we can assume inertial coordinates to answer this question.

If by this you mean that "vacuum" implies inertial coordinates, globally that is only true in the absence of gravity. In the presence of gravity, you can only set up inertial coordinates locally, in a small patch of spacetime.

Even in the absence of gravity in a vacuum, however, inertial coordinates are still coordinates, and no actual physics can depend on the coordinates you choose. A question about actual physics would just be along the lines of "can you apply a force to a photon?", and the answer to that in a vacuum (whether or not gravity is present) is "no". (Even in a material medium, it's not clear that "applying a force" is the best way to understand the medium's effect on light; but since we're talking about vacuum here, we don't have to get into that issue.)

 

[David Lewis]

gatopardos wrote: Velocity is the vector of speed, the absolute value of the vector is constant there is no acceleration. Speed is constant.

David Lewis wrote: You may have meant the magnitude of the vector, not the absolute value.
(A.T.'s post #16 claimed that positive and negative values of acceleration tell you whether the direction of motion is up, down, left, or right.)

 

[Jon Richfield]

Jon, That is very incorrect
photons are emitted at the speed of light, they DO NOT accelerate to that speed
or to put it the other way ... the EM field is emitted at the speed of light, it does not accelerate to that speed

D

Dave, that was not my wording, but the wording of the question. I interpreted it as meaning "can photons undergo acceleration".
It never even occurred to me to think in terms of the emission of a photon involving acceleration as such.
In a broader sense it takes a very unnatural interpretation in physics to argue that any particle such as a photon or electron literally accelerates, rather than undergoing acceleration. Granted we might (often do) say that eg an electron in an electric field accelerates, but to me that wording suggests the electron actively "stepping on the gas". I reckon that is just a loose way of saying that it passively undergoes acceleration. Given such loose wording in practice, I am not inclined to be as picky as my natural pedantry might lead me to prefer.

 

[Jon Richfield]

I think these answers may be confusing to the OP. Considering the prefix (B) on the thread, I would assume the question is "Can you apply a force that would increase the speed of a photon?" Also keeping in mind the prefix, I think we can assume inertial coordinates to answer this question.

I understand the assumption, but if it is correct, then answering the question literally and exclusively in those terms is presumptuous at best, and then a responsible reply should make it plain to the questioner that that was in fact the presumption, plus the facts that instead of accelerating, the colour of the photon would change, and that gravitational refraction could involve acceleration in a less naive sense.

 

[davenn]

Dave, that was not my wording, but the wording of the question. I interpreted it as meaning "can photons undergo acceleration".

it's you answer that is incorrect

" Light being emitted from a source in vacuum, DO photons accelerate ?"
the answer emphatically is "Yes".

your response ... the answer emphatically is "Yes"

is totally incorrect ... photons aka EM radiation is emitted AT THE speed of light

 

[Jon Richfield]

No, the answer is "it depends on how you define acceleration, and if you define it as coordinate acceleration, it depends on the coordinates you choose". But no actual physics can depend on the coordinates you choose. So the only version of the question that is asking about actual physics, as opposed to just your choice of coordinates, is the version that asks whether light rays in vacuum can have path curvature. And the answer to that question emphatically is "No".

Not so, but far, far otherwise. Any physics that can have "physical" effects is "real" physics, or what I assume you mean by "actual physics". It is what I might call "physics with measurable implications". I you know of a better contrary definition, please let's have it.

Now, consider a particular beam of light as a photon stream in a (gedanken-) experimental setup in a vacuum in a zero gravitational gradient. If you please, you could choose convenient values for polarisation, frequency etc.
I am not much fashed with details as long as it is all nicely controlled and consistent.
The photons simply pass from the source to the target at a conveniently large distance. For as long as we like, our target will register photons that have traversed the same distance in the same time along the same trajectory and with the same frequency/energy etc. For all observers, though they might disagree about the values of the parameters, they all would agree that the parameters would remain constant and consistent.
Now (still in vacuum, and with no other parametric changes, remember!) we start messing about with gravitationally non-trivial masses that alter the gravitational gradients, and accordingly at least certain classes of the measurements that our target records.
We still do not find c to vary (pace Shapiro of course, which I reckon could be regarded as a coordinate effect anyway, but suit yourself) but you certainly could change the angle, location, and time of arrival of your photons, any of which changes would involve acceleration at constant speed, though not constant velocity.
If our lab equipment included suitably disposed black holes (or possibly pulsars would do if the budget wouldn't stretch to black holes) we even could have photons arriving in the back of our target, 180 degrees out of phase with other photons in the same stream.

And that is what our instruments would show. That is physics actual enough for me.
And acceleration actual enough too.
The answer remains emphatically yes, no matter how you jiggle your definitions in actual physics.

 

[Jon Richfield]

it's you answer that is incorrect
your response ... the answer emphatically is "Yes"

is totally incorrect ... photons aka EM radiation is emitted AT THE speed of light

Dave, I am sure we must be at cross purposes. I never was even speaking of emission, but of whether photons in free passage through otherwise empty space, under the influence of gravitational gradients could undergo acceleration, which as far as I can tell, a lot of folks here are stridently denying.

To which indeed, the answer emphatically is "Yes", then now, and still.

 

[A.T.]

We still do not find c to vary

The constant c doesn't vary per definition. But the coordinate speed of a light beam in non-inertial coordinates can be different from c (see Shapiro delay).

any of which changes would involve acceleration at constant speed, though not constant velocity.

Coordinate acceleration of light in non-inertial coordinates can involve direction and speed changes.

 

[Jon Richfield]

The constant c doesn't vary per definition. But the coordinate speed of a light beam in non-inertial coordinates can be different from c (see Shapiro delay).Coordinate acceleration of light in non-inertial coordinates can involve direction and speed changes.

From that point of view, what is your point?
Are you denying the acceleration of light?
Or denying that it is real, or meaningful, or what?
If OTOH you are asserting the acceleration, welcome to the club, but why tell me? It is a lot of other guys denying it.

 

[A.T.]

Are you denying the acceleration of light?

Light can have coordinate acceleration, but no proper acceleration.

 

[Jon Richfield]

Light can have coordinate acceleration, but no proper acceleration.

Come back and tell us when you can show that light (photons if you like!) can have no physically measurable acceleration -- no acceleration with consequences observable in principle (though not necessarily identical in value) to sufficiently equipped observers in every reference frame.
As long as that remains true, it hardly matters whether you call it proper or not.

 

[mfb]

Stop discussing this in this [B]-thread please. We have the other one for that.

Note that you just keep using different meanings of "acceleration", which lead to different results - obviously. I don't understand why that leads to those long discussions.

 

[Isaac0427]

Stop discussing this in this [B]-thread please. We have the other one for that.

Note that you just keep using different meanings of "acceleration", which lead to different results - obviously. I don't understand why that leads to those long discussions.

That is what I have been saying. The answer to the OP's question is no, nothing going at c, the speed of light in a vacuum, can be accelerated. If he wanted to know all the technicalities with non-inertial frames of reference and curved space in which you have to generalize your answers, he would have asked that, put an (I) or (A) prefix on the thread, or put this in the relativity forum. Some people come here for basic answers in layman's terms and not a lecture on relativity (which was me when I first started). I think these answers are doing nothing but confusing the OP.

 

[PeterDonis]

Any physics that can have "physical" effects is "real" physics, or what I assume you mean by "actual physics". It is what I might call "physics with measurable implications".

If you mean that there are measurable implications of the effects of gravity on light in vacuum, yes, there are. So if the question had been "does gravity have a measurable effect on light", then the answer would indeed emphatically be "yes".

But that wasn't the question. The question was "can light in vacuum accelerate?" The answer to that question is not emphatically "yes" because it's an ill-posed question. It depends on how we interpret "acceleration". And the only interpretation that makes "acceleration" equivalent to "measurable effects of gravity on light in vacuum" is a coordinate-dependent one. Your description makes that clear; you have to use terms like "speed", "velocity", etc. in order to describe what's going on. That's a coordinate-dependent interpretation; the terms you're using don't even have meaning in the absence of coordinates.

By contrast, I can describe the same effects without using the term "acceleration" (or the terms "speed" or "velocity") at all. Light in vacuum travels on null geodesics of spacetime; gravity curves the spacetime geometry. That accounts for all observed effects, solely in terms of coordinate-independent invariants; I don't even have to choose any coordinates at all to describe what's going on.

no matter how you jiggle your definitions in actual physics.

You're the one that's trying to jiggle definitions; you're trying to argue that "acceleration of light in vacuum" in the sense you're using the term means something absolute when it doesn't; it only has meaning in a coordinate-dependent sense. All of the light rays you're talking about have zero proper acceleration, i.e., zero path curvature. And "proper acceleration" is the only coordinate-independent, invariant, absolute meaning of the term "acceleration". So in the only absolute sense of the term "acceleration", light in a vacuum does not accelerate, and the answer to the question emphatically is "no", as I said.

 

[PeterDonis]

The question has been beaten enough. Thread closed.