Originally posted by lemon lime
A question about redshift. Why does redshift occurring in relation to earth necessarily mean everything is accelerating away from everything else? If earth itself is in motion relative to some other more or less central point of observation then it might only appear the entire universe is accelerating. If it turns out the earth is simply decelerati ...[text shortened]... g this out here as something to think about... I doubt if it can be proven one way or the other.
Hi lemon lime,
the previous responses to that question were inaccurate. There is no reason that a distant galaxy should not be moving towards the Earth. I can't remember the exact figure, but the Solar System is moving at a speed of the order of 150 km/s compared with the microwave background. So all the observations need to be corrected for that. In any given galactic cluster the galaxies are all moving in different directions from one another. Andromeda is moving towards us. However at a given distance the
average redshift is what matters. The basic assumption is that on average any given cluster has as many galaxies moving one way as another.
In terms of your analogy, you're right if there is only one car. However, if there is a roadful of them then some other cars will be accelerating and some cars will have constant speed. The average speed of all the cars is what the observer on the train should be observing.
Your underlying point about what would be seen from another fairly distant galaxy is a good one. I think that, in general, there is no reason to believe that the universe would look any different from another location. It's a fundamental assumption in cosmology which I assume, but do not know, had been tested as far as one can. The problem is that we can measure radial velocity quite easily, but not tangential velocity at those distances. So it is rather difficult to prove isotropy.
Science has it's limitations, but the theory is more-or-less consistent with observation, so it's scientifically true, which means "provides a reasonable description and not yet disproven".