Is the cosmic microwave background really non-Gaussian?

Two recent papers have claimed detection of non-Gaussian features in the CO BE Differential Microwave Radiometer sky maps of the cosmic microwave backg round. We confirm these results, but argue that Gaussianity is still not co nvincingly ruled out. Since a score of non-Gaussianity tests have now been published, one might expect some mildly significant results even by chance. Moreover, in the case of one measure that yields a detection, a bispectrum statistic, we find that if the non-Gaussian feature is real, it may well b e due to detector noise rather than a non-Gaussian sky signal, since a sign al-to-noise analysis localizes it to angular scales smaller than the beam. We study its spatial origin in case it is nonetheless due to a sky signal ( e.g., a cosmic string wake or flat-spectrum foreground contaminant). It app ears highly localized in the direction b = 39.degrees 5 l = 257 degrees, si nce removing a mere five pixels inside a single COBE beam area centered the re makes the effect statistically insignificant. We also test Gaussianity w ith an eigenmode analysis which allows a sky map to be treated as a random number generator. A battery of tests of this generator yields results consi stent with Gaussianity.