4.0 Testing

Experimental Testing

Traditional SR testing

Existing experimental tests of special relativity typically compare two sets of predictions: those of "classical theory" (which assumes a flat, fixed aether stationary in the lab frame), and those of SR.

The SR predictions are "redder and shorter" than those of CT, by the Lorentz factor.

CT ........... SR ........... ??

<--- blue

red --->

Traditionally, a shift difference to the "red" of CT whose magnitude is at least as strong as SR is considered to validate special relativity, and overshoots into the region to the right aren't assumed to carry theoretical significance, since no theories are supposed to exist in this range. "Excess" redshifts can be calibrated out of the experimental apparatus, or blamed on assumed hardware problems or complications (such as mirror recoil).

4.0 testing

The proposed "4.0" predictions are in turn "redder and shorter" than SR, by a second Lorentz factor, putting the "range of interest" for testing 4.0 outside the range covered by current test theory.

CT ........... SR ........... 4.0

<--- blue

red --->

4.0 testing treats the range "to the red" of SR (which we normally ignore) as critical, while the range that we do currently test, "to the blue" of SR, is invalid under 4.0, since it corresponds to solutions that would associate positive energy with negative curvature.

Requirement for further testing

The asymmetrical nature of the usual SR-testing procedures (only looking for deviations from SR to the "blue" side) means that the current experimental data supporting SR over CT (see McArthur review, 1986) doesn't indicate that the SR predictions are necessarily better than those of 4.0 . Results from complex experiments designed to test "CT vs SR" can't be safely reanalysed for "4.0 vs SR", because of the difficulty of knowing all the potential undocumented steps to eliminate "excess" redshift that the experimenters might have carried out.

This leaves us with the option of reanalysing only the very simplest tests, or performing a new round of testing. The only two "basic" tests available both have reported problems.

Ives-Stilwell (non-transverse test, 1938) appears to support SR and rule out 4.0, but reports an unresolved issue with spectral lines, and its accuracy has since been queried.
Hasselkamp et al (transverse test, 1979) found twice the predicted SR transverse redshift, as predicted by 4.0 . The experimenters took this "problematic" result to mean that their detector had been misaligned by half a degree. Statistical analysis then included this (presumed) half-degree error to report a successful verification of special relativity.

The Hasselkamp paper demonstrates:

... that "4.0-style" results have already been appearing, unrecognised, in at least some experiments,
... that current test theory allows a "4.0" result that disagrees with SR to be corrected and presented as supporting SR to an accuracy of a few percent – even if the collected data disagreed with the SR prediction by a factor of two,
... that expert peer review has failed to highlight the potential significance of these results and alert the community, and,
... that the strength of accumulated experimental evidence apparently supporting special relativity is not necessarily significant evidence against 4.0. The Hasselkamp test might not be the only experiment where a successful "SR" result reflects data that makes a better match with 4.0.

To find if 4.0 is more or less accurate than special relativity requires further testing, performed in the context of a test theory that does not allow "excess" redshifts to be discarded or calibrated away.

A firm answer should be achievable with current technology.