Avoiding unrealistic priors: the case of dark energy constraints from the time variation of the fine-structure constant

Abstract

We critically assess recent claims suggesting that upper limits on the time variation of the fine-structure constant tightly constrain the coupling of a dark energy scalar field to the electromagnetic sector, and, indirectly, the violation of the weak equivalence principle. We show that such constraints depend crucially on the assumed priors, even if the dark energy was described by a dynamical scalar field with a constant equation of state parameter $w$ linearly coupled to the electromagnetic sector through a dimensionless coupling $\zeta$. We find that, although local atomic clock tests, as well as other terrestrial, astrophysical and cosmological data, put stringent bounds on $|\zeta| {\sqrt {|w+1|}}$, the time variation of the fine-structure constant cannot be used to set or to improve upper limits on $|\zeta|$ or $|w+1|$ without specifying priors, consistent but not favoured by current data, which strongly disfavour low values of $|w+1|$ or $|\zeta|$, respectively. We briefly discuss how this might change with a new generation of high-resolution ultra-stable spectrographs, such as ESPRESSO and ELT-HIRES, in combination with forthcoming missions to map the geometry of the Universe, such as Euclid, or to test the equivalence principle, such as MICROSCOPE or STEP.