Timing, ``Accidentals'' and Other Artifacts in EPR Experiments

Subtraction of ``accidentals'' in Einstein-Podolsky-Rosen experiments frequently changes results compatible with local realism into ones that appear to demonstrate non-locality. The validity of the procedure depends on the unproven assumption of the independence of emission events. Other possible sources of bias include enhancement, imperfect synchronisation, over-reliance on rotational invariance, and the well-known detection loophole. Investigation of existing results may be more fruitful than attempts at loophole-free Bell tests, improving our understanding of light.