Locations within the galaxy M33, the Great Spiral in Triangulum, where the beam of the Arecibo Telescope has been pointed while searching for ETI signals from any civilizations in that galaxy. The points are superimposed on a map giving the general velocities of stars and interstellar gas throughout M33. With this distribution of points and the telescope beamwidth, every star in the galaxy falls within the coverage achieved with the Arecibo Telescope. Each location was searched for signals for a minimum of 60 sec. At any given instant, about one billion stars were within the beam of the telescope.

THE SCIENCE OF SETI

SETI is a manifestation of man's drive to explore. This drive is one of the oldest and most fundamental aspects of our nature; the very origin of the hominidae as a distinct biological entity is owed, at least in part, to the boldness of our venturesome simian ancestors who abandoned their familiar forest environment to probe the savannah, there to seek fleet-footed prey. Our forebears pushed into almost every corner of the globe. They explored by climbing hills, by walking through forests, and even by crossing large bodies of water. Sometimes they may have had in mind some material purpose, but certainly they sometimes went where they did for no other purpose than to see what was there. Modern man still explores, but the arena for his exploration is now the planets and the stars beyond. But we are not limited to physical exploration. We can use the fruits of our intelligence to conduct exploration from a distance. Though some day we may wish to build space ships to travel there, to probe the stars now we need only telescopes. Yet the excitement and exhilaration that comes with this kind of exploration, as larger telescopes and more sensitive and sophisticated data acquisition techniques lead to discovery after discovery, is akin to that the Viking seamen must have felt.

Exploration has always required knowledge and understanding of the physical world. Viking boats could not have been built without knowledge of what wood to use; Viking navigation could not have been accomplished without an understanding of the winds and tides. We call this understanding of nature, which we gain from observation and experiment, scientific knowledge. To explore the stars in search of other intelligent life also requires scientific knowledge; indeed, because it can only be done using highly sophisticated technologies and methods, it requires more scientific knowledge than have man's classical explorations.

A SETI program should embrace not only a search for evidence of other civilizations, such as radio signals, but also a wide range of related scientific studies. We need knowledge of nature primarily for two purposes. One of these is to enable us to narrow the scope of the search by distinguishing promising volumes of search space; for example, we might be better able to identify promising target stars or frequency bands. The other purpose is to enable us to be able to interpret any evidence of other civilizations we obtain, and to decide what course we should follow once we are sure that other intelligent life has been discovered.

The scientific knowledge needed by a SETI program can perhaps best be illustrated in the context of the Drake equation, which relates the expected number N of intelligent, technologically advanced communicative species in the Galaxy to the product of several factors. It should be understood that the Drake equation is not a fundamental expression of the way nature behaves, as is, for example, the deceptively simple law, f= ma. Rather, the Drake equation is simply a device to enumerate the factors that influence N and hence must be considered in any attempt to estimate this number. One form of the Drake equation (there are several) is

N = R2fgfpne fifi fc L

where R is the average rate of star formation in the Galaxy, fg is the fraction of stars that are "good" stars in the sense of providing conditions thought to be necessary for life, fp is the fraction of good stars that have planets, n is the number of suitable planets in a typical planetary system, f is the fraction of suitable planets on which life starts, f¡ is the fraction of life starts that evolve intelligence, fe is the fraction of intelligent species that enter a communicative phase and L is the mean lifetime of the communicative phase. It is immediately obvious that the factors on the right-hand side of the equation have widely differing character. Some, such as R✶ and fp, involve only knowledge from one basic discipline, astrophysics. Others, such as fg, n, fi and to some extent fi involve questions spanning many disciplines, including astrophysics, prebiotic chemistry and biology. Finally, the terms fe and L involve considerations not generally found in the natural sciences, but which nevertheless lend themselves to scientific inquiry. The full panorama of the elements in the Drake equation is embodied in the concept of "cosmic evolution" (see Section II-1).

For convenience, we can take the Drake equation as a model and categorize the science of SETI under the headings of physical science, biological science, and social science. We begin with those aspects of SETI involving the physical sciences.

*

A very important question concerns the way stars are born. At present, our understanding of this stage in the evolution of matter is sketchy at best. We are able to calculate the mean rate of star formation over the age of the Galaxy (this number is generally used for R in the Drake equation), but we are not yet able to specify in detail the conditions necessary for star formation nor are we able to describe or predict the course of events involved in the evolutionary path from a relatively low density interstellar cloud to an incandescent ball of gas supplying its own energy by nuclear fusion deep in its interior. Observational progress on this question has been forthcoming over recent years through infrared and radio studies of dark clouds and newly formed stars; however, many large gaps still exist in the fabric of our knowledge. Theoretical study of star formation is in an even more rudimentary state than is its observational counterpart, due principally to the intrinsically three-dimensional, non-analytic nature of the mathematical representation of the problem, and the inability to model or parameterize the role of turbulence and plasma dynamics in dense interstellar clouds. The advent of very rapid computers is making it possible now to undertake the formidable numerical problems involved in simulating the physics of star formation, and with adequate support we soon should be in a position to compare meaningfully theory and observation. Understanding this fundamental process of star formation is important not only because it provides the means for obtaining a more accurate value of R in Drake's equation, but more significantly, it will provide the required construct within which we can understand the formation of planetary systems. An understanding of the formation of planetary systems could allow us to eliminate certain classes of stars as targets of the search.

Some stars, even though they have planetary systems, may be intrinsically hostile to life. There is some evidence, for example, that M-dwarf stars produce powerful flares with accompanying radiation levels inimical to living organisms. There is also the possibility that, as a consequence of the low luminosity of M-dwarfs, a planet in the stellar ecoshell would have to be so close to the star that its axial rotation period would become tidally locked to its orbital period about the star.