I saw in a chapter on special relativity a derivation of the transverse Doppler effect, which seemed okay, but I have a question concerning this: If the motion of a source is always perpendicular to the position vector connecting the observer to the moving source (i.e. the distance remains constant), that is the source rotates about the obs., does not (technically, at least) the validity of the result (transv. D. eff.) depend on general relativity, rather than special relativity, since the source's rest frame is in non-uniform motion relative to the observer's rest frame?