http://www.ias.ac.in/jarch/jaa/20/91-101%20.pdf
John Stachel: "At first Einstein looked for a scalar generalization of
Newton's theory, based on the gravitational potential. By the middle
of 1912, he had worked out what he regarded as a satisfactory theory
for the case of a static gravitational field. He developed a field
equation for the gravitational potential, which he identified in this
case with a VARIABLE SPEED OF LIGHT c(x,y,z) INSTEAD OF THE CONSTANT
SPEED OF THE SPECIAL THEORY..."

In Einstein's zombie world "the external world exists only in the
mind", as George Orwell would put it. When Einstein and his hypnotists
call the speed of light "variable" they mean "variable so far as WE
call it so and deal with it accordingly". The fact that the speed of
light varies with the gravitational potential makes sense only so far
as Einstein used it in the period 1907-1915. Then Einstein stopped
using it and therefore for later periods this fact makes no sense at
all. If in some period Einstein had applied his equivalence principle
and deduced that, since the speed of light varied with the
gravitational potential, it also varied with the relative speed of the
light source and the observer, in the absence of a gravitational
field, then Einstein's second postulate (the speed of light is
independent of the speed of the light source) would be false for that
period. However for other periods Einstein's second postulate would be
true. That is a speculation of course: Einstein does not seem to have
deduced anything like that and therefore his second postulate is true
forever.

Pentcho Valev