Pentcho Valev wrote:
A. Einstein, "Relativity", Chapter 22:

"A curvature of rays of light can only take place when the velocity of
propagation of light VARIES with position."

So, according to Einstein, an observer positioned in a gravitational
field will measure a VARIABLE, not constant velocity of light. The
observer can measure, for instance, the frequency shift and, if
Einstein is right, a non-zero frequency shift corresponding to the
changing velocity of light will be detected.

Einsteinians never discuss this problem voluntarily but, when pressed,
they produce a combination of words more or less like this: The
velocity of light is constant only locally, non-locally it is variable.
The meaning of this combination of words, especially when applied to
Einstein's assertion, is a grand secret between Einsteinians.

This is all quite well known, has been for decades, and is no dirty
little secret. (If it were proprietary, then why did you find it in a
book written by Einstein targeted at the general public?)

Even in special relativity it's necessary for the speed of light to
vary between frames of reference having a relative velocity. The speed
of light is invariant locally. WITHIN any INERTIAL reference frame, c
will always measure out to 3.0 x 10^8 m/s.

-Mark Martin