Although we do not fully understand AGNs in the way in which we understand stars, many of the main ideas of their nature seem to be coming into focus. Perhaps our knowledge of AGNs in 1990 is roughly equivalent to our predecessors' knowledge of the nature of stars in 1935 or 1940. They had been known since antiquity, and had been observed astrophysically with telescopes since the latter part of the last century; Seyfert galaxies were `discovered' in 1943 and QSOs twenty years later. Our rate of learning, though not as fast as we might wish, is not as slow as we might fear either.
The general ideas of photoionization by a hard spectrum, powered by an accretion disk about a black hole and by non-thermal radiation generated in and near it seem well established. We have many clues and insights into the structure and velocity field, but no complete picture from which detailed predictions can be calculated from first principles. Dust and star formation are clearly important parts of the AGN process, but we do not understand the differences between the situations in which they do and do not lead to it. Much observational evidence suggests that interactions between galaxies can lead to the formation of black holes, fuel them, and refuel existing black holes. But we cannot calculate the evolution of a massive black hole and its associated AGN as we can calculate the evolution of a star.
The Hubble space telescope, if it performs up to its specifications, will surely increase our understanding of AGNs. So will large ground-based telescopes, optimized for seeing. Significantly better angular resolution should reveal much of the structure of the NLR The new x-ray observatory ROSAT and the planned Gamma Ray Observatory should make possible much better measurements of the high-energy spectra of AGNs. Infrared imaging cameras and millimeter-wave interferometers will enable us to make much more complete studies of the dust and molecular gas in these objects. Supercomputers will provide more detailed simulations of many more cases of interactions between galaxies, and the perturbations of the velocity fields of the gas within them. I expect rapid progress in understanding the physical nature of AGNs in the coming decade.
This review was written during my sabbatical leave at the Institute for Advanced Study in 1989-90. I am most grateful to the Institute, its director, Marvin L. Goldberger and its professor in astronomy, John N. Bahcall, for their hospitality and support. I am also grateful to my home institution, the University of California, Santa Cruz, for its continued support of my research, and also to the National Science Foundation for its support, most recently under grant AST 86-11457.