BAYESIAN-ANALYSIS OF THE POLARIZATION OF DISTANT RADIO-SOURCES - LIMITS ON COSMOLOGICAL BIREFRINGENCE

A recent study of the rotation of the plane of polarization of light f rom 160 cosmological sources claims to find significant evidence for c osmological anisotropy. We point out methodological weaknesses of that study, and reanalyze the same data using Bayesian methods that overco me these problems. We find that the data always favor isotropic models for die distribution of observed polarizations over counterparts that have a cosmological anisotropy of the type advocated in the earlier s tudy. Although anisotropic models are not completely ruled out, the da ta put strong lower limits on the length scale lambda (in units of the Hubble length) associated with the anisotropy; the lower limits of 95 % credible regions for lambda lie between 0.43 and 0.62 in all anisotr opic models we studied, values several times larger than the best-fit value of lambda approximate to 0.1 found in the earlier study. The len gth scale is not constrained from above. The vast majority of sources in the data are at distances closer than 0.4 Hubble lengths (correspon ding to a redshift of approximate to 0.8); the results are thus consis tent with there being no significant anisotropy on the length scale pr obed by these data. [S0556-2821(97)02824-5].