It has been argued that certain broad absorption line quasars are viewed within 35 degrees of the axis of a relativistic radio jet, based on two-epoch radio flux density variability. It is true if the surface brightness of a radio source is observed
to change by a sufficiently large amount, the inferred brightness temperature will exceed 10^12 K and Doppler beaming in our direction must be invoked to avoid a Compton cooling catastrophe. However, flux density changes cannot be linked to surface brightness changes without knowledge of the size of the source. If an optically thick source changes in projected area but not surface brightness, its brightness temperature is constant and its flux variability yields no constraint on its orientation. Moreover, as pointed out by Rees, spherical expansion of an emission source at relativistic speeds yields an apparently superluminal increase in its projected area, which can explain short-timescale flux density variability without requiring a relativistic jet oriented near to our line of sight. Therefore, two-epoch radio flux density variability by itself cannot unambiguously identify sources with jets directed towards us. Only VLBI imaging can robustly determine the fraction of broad absorption line quasars which are polar.
