0

I watched this very interesting video on the speed of light, which stated that no-one has actually measured the speed of light, but rather the round trip speed of light. In essence, this video suggests it would be possible for light to travel instantaneously in one direction, and at half the "speed of light" in the other direction, and this would have the same experimental result as if the speed of light is constant.

I have one big issue with this idea. Assuming there is a "cosmic north" in which light is faster in one direction than another, wouldn't that be relevant when shooting beams of light at angles off of this "cosmic north"?

Consider the following diagram, where "cosmic north" (an axis in which the speed of light is instantaneous) is the black arrow, a red vector which represents the velocity of a photon, and and a blue vector which represents the velocity of the photon projected onto the "cosmic north".

Black arrow up, blue arrow up but smaller, red arrow diagonal

because there is a portion of the photons velocity which is along the direction of the cosmic north, the velocity of that photon along that axis would have to be some fraction of infinity. As that would remain infinite, in this hypothetical scenario, any angle <90deg and >-90deg would be impossible to obtain.

My question is, does this line of thinking make the "instantaneous one way, half speed the other way" model of light impossible?

Daniel Warfield
  • 103

3 Answers3

1

The idea of the speed of light being infinite one one direction and finite in another is nonsense. However, to take your question at face value, if any beam of light always had two components of velocity, one infinite and the other finite, then it would be impossible to direct a beam of light at any arbitrary point on a sphere around its source, since the infinite component would always dominate the propagation of the beam.

Professor Sushing
  • 29,350
1

the velocity of that photon along that axis would have to be some fraction of infinity

This is not correct. The projection of a vector onto an axis is some fraction of the actual length of that vector. It is not some fraction of the hypothetical length that the vector would be if it were pointing in a different direction. The operation you describe is not any part of caluclating vector components.

However, if you took an actual “north” light vector, which would be infinite, and broke that into components on any non-perpendicular axis then those components would be infinite. There is nothing particularly wrong with that, but it is a little weird.

You may want to study Reichenbach synchronization. It is entirely self consistent, but needlessly complicated.

Dale
  • 117,350
0

There are experiments done using stellar aberration which can measure the speed of light using light that only travels in one direction, basically rules out the loop hole you are attempting to use. In addition if light did travel at different speeds in different directions then you would break special relativity/Lorentz symmetry. Breaking these symmetries breaks conservation laws meaning you shouldn't have conservation of angular momentum (and probably wouldn't even have conservation of energy). Since we don't see those violated that is pretty strong evidence that you shouldn't expect there to be a directionality to the speed of light.

N A McMahon
  • 466
  • 2
  • 6