What Have We Learned From GRB Afterglows?

The discovery by BeppoSAX and coordinated ground-based observations of persistent X-ray, visible and radio counterparts to GRB has successfully concluded a search begin in 1973. The problem of explaining the mechanisms of GRB and their persistent counterparts remains. There are two classes of models: 1) GRB continue weakly for days at all frequencies; 2) GRB emission shifts to lower frequencies as relativistic debris sweeps up surrounding gas (in an ``external shock'') and slows. In 1) the visible afterglow is accompanied by continuing gamma-ray emission, as hinted by the high energy emission of GRB940217 and the ``Gang of Four'' bursts of October 27--29, 1996; the persistent emission may fluctuate. Behavior of this sort may be found in ``internal shock'' models. Models of class 2) have been the subject of several theoretical studies which disagree in assumptions and details but which predict that at each frequency the flux should rise and then decline, with the maximum coming later at lower frequencies. Some of this behavior has been observed, but data from GRB970508 show that its afterglow cannot be simply extrapolated from its gamma-ray emission. It is likely that both classes of processes occur in most GRB. Comparisons between GRB show that they are not all scaled versions of the same event. These results suggest that most gamma-ray emission is the result of ``internal shocks'' while most afterglow is the result of ``external shocks'', and hint at the presence of collimated outflows. Self-absorption in the radio spectrum of GRB970508 permitted the size of the radiating surface to be estimated, and in future GRB it may be possible to follow the expansion of the shell in detail and to construct an energy budget.