resurgance2001 said:
Hi Phinds - thanks for your reply. When you say super luminal recession rates are readily demonstrated in SR can you explain this more? I thought that was the whole point, that in SR it is not possible for anything to go faster than the speed of light, no?
This is a major weakness in how cosmology is presented in a number of textbooks. To clarify, a few definitions are in order:
1) Relative velocity: The tangent vectors of two world lines are compared. This is an unambiguous operation in SR, no matter how far apart in space and time two vectors are, because parallel transport is path independent. This corresponds to the coordinate velocity of one in the instantaneous rest frame of the other. In GR, this operation is only unambiguous if the two world line tangents are coincident, or at least sufficiently close for spacetime to be treated as flat. Otherwise, relative velocity is simply undefinable or at least, inherently ambiguous - vary your parallel transport path and you get wildly different relative velocities. Thus, velocity comparison over large distances is not a possible operation at all in GR.
However, two additional observations are worth making:
a) The result of comparison between a lightlike vector and a timelike vector is path independent and the result is always c, with total generality in GR. Thus, the relative speed of a light and any material body is identically c, with no exceptions in GR.
b) Even though relative velocity of distant world lines is fundamentally ambiguous in GR, this ambiguity does not allow superluminal relative velocities because comparison of any timelike vectors over any possible parallel transport path is always less than c. Thus, a precise statement is that the precise relative speed is ambiguous, but still always less than c. Different comparison paths produce different values, but they are all less than c.
2) Celerity: given some foliation of spacetime, the rate of change in proper distance measured on this foliation between one world line and another by the proper time time on one of them. This is slighly more general that the most common definition of celerity in SR, allowing general use in GR. Cosmological recession rates are a particular instance of celerity: the standard cosmological foliation is used, and the world lines are the comoving ones. In SR, there is no upper bound to the value of a celerity, and the same is true in GR. A trivial example in SR is to use the foliation of a solar inertial frame, use the sun's world line as a reference world line, and the world line of a rocket traveling near c relative to the sun as the one whose celerity we measure. Then as the rocket travels one light year in the chosen foliation, its corresponding proper time can be made arbitrarily small, with corresponding arbitrarily large celerity. Of more relevance to cosmology, you can introduce cosmological style coordinates in SR (Milne coordinates) that have many features of FLRW solutions; in fact, this coordinate system is simply the result of the zero density limit of FLRW solutions. The resulting spacetime is simply flat Minkowski space, foliated by hyperbolic slices. Recession rate computed in these coordinates in
flat spacetime (pure SR) grows with distance for 'comoving' world lines, with no upper bound, just as in realistic FLRW solutions.
There are many differences between such a fake flat spacetime cosmology and a realistic one, but superluminal recession rate is
not one of these differences. The unfortunately common claim that this is a distinction is a fundamental category error: comparing relative velocities in SR with recession rates in an FLRW solution.
An important difference, for example, is that given a congruence of world lines with isotropy and homogeneity and positive expansion scalar in flat spacetime, the foliation of common proper time from initial coordinate singularity (in flat spacetime, it is only coordinate singularity not a true singularity) must have a unique hyperbolic geometry. In curved spacetime, any constant curvature geometry is possible for different mass densities, including flat spatial slices such as our universe appears very close to.