A variable speed of light cosmology has been proposed independently by John Moffat and the two-man team of Andreas Albrecht and João Magueijo to explain the horizon problem of cosmology.[20][21][22] [23] [24] [25] [26] The idea is that light propogated as much as sixty times faster in the distant past, and thus distant regions of the expanding universe have had time to interact since the beginning of the universe. As such, it was proposed as an alternative to cosmic inflation, although it is less clear how it reproduces the other successes of inflationary cosmology such as the monopole and flatness problems and how it reproduces the observed homogeneity and isotropy of the universe, and the scale invariance of the spectrum of initial perturbations.
There is no known way to solve the horizon problem with variation of the fine-structure constant, because its variation does not change the causal structure of spacetime. To do so would require modifying gravity by varying Newton's constant or redefining special relativity. (See equivalence principle for further details.) Varying speed of light cosmologies propose to circumvent this by varying the dimensionful quantity c by breaking the Lorentz invariance of Einstein's theories of general and special relativity in a particular way.[27] However, it has been pointed out by Ellis and Uzan[28] that the VSL cosmology is an ad hoc modification of various equations of physics without a consistent underlying scheme, such as a Lagrangian from which the equations of motion can be derived. It has been suggested out[29] that a modification of the Einstein-Maxwell action can cause light to propagate at a speed faster than the speed of light defined by the metric, but this necessarily causes problems with causality and quantum mechanics.[30]