Paradoxes of the theory of relativity
Rotation Effects
In SRT, light speed is supposed to be c. But the wellknown Sagnac effect belies that supposition. The Sagnac effect constitutes rotation sensing by means of the rotation-induced shift in the interference pattern produced by two light beams traveling in opposite directions around an optical loop. The implication is that light has speed c only in a nonrotating coordinate frame. The troublesome Sagnac effect can be excluded from the domain of SRT by excluding the rotational motion that produces it.
The argument goes: rotation involves acceleration, and hence motion that is not inertial. Also true although not usually mentioned, rotation invites rotating coordinate frames, and, for large enough radii, rotating coordinate frames imply relative motions at speeds in excess of c, precluded by the speed limitation accepted by SRT.
But rotation exclusion is not really feasible. The SRT Lorentz group of coordinate transformation naturally includes rotations, and noncollinear Lorentz transformations generate rotations. Indeed, those details are involved in explaining another experimental effect: the socalled anomalous magnetic moment of an electron in an atom.
All these things can become consistent only within the context of two-step light propagation. Although it goes beyond the scope of a single paper to discuss all possible rotation related effects, let us consider one of them here: the above mentioned anomalous magnetic moment of an electron in a hydrogen atom. The situation is that from the vantage point of the electron, the proton nucleus is seen to circulate around and create a magnetic field. The problem is that the energy associated with the electron coupling to this magnetic field is only about half what would be expected on the basis of the energy associated with the electron coupling to some exogenous magnetic field.
The resolution emerges as follows. Two-step light propagation implies coordinate transformation matrices that are not the same as Lorentz transformations. Instead of the coefficients characteristic of Lorentz transformations, there are coefficients , and v/2c*(1+v/8c) . This latter coefficient determines the magnetic field seen by the electron due to the nucleus in the atom "circulating", and it is about half the coefficient SRT uses.
Thus the magnetic field communicated by two-step light propagation in this situation is only about half what SRT predicts. So the electron doesn't really have an anomalous magnetic moment, but rather it sees an "anomalous" magnetic field. But this field is "anomalous" only as judged by SRT.
