An Analysis of the Gamma‐Ray Burst No‐Host Problem

Abstract

The lack of galaxies inside gamma-ray burst (GRB) error boxes has often been used to argue against the cosmological interpretation of this phenomenon. The fundamental problem with host galaxy tests is that our understanding of GRBs is too limited for constructing a sensitive measure of the presence of suitable hosts. In this paper the origin of the no-host problem is explained, and the tests used to make these inferences are evaluated. It is shown that these analyses are too dependent upon factors unrelated to bursts or to their host galaxies and that normal galaxies have not been eliminated as possible hosts. New data from a near-infrared sample of third Interplanetary Network error boxes, which reached levels deep enough to detect galaxies at every position, show the brightest galaxies to be at the distances (z ~ 0.2) and luminosities (L ~ 0.6L*) that are predicted by the simplest assumptions about cosmological GRBs.