the essential factor in working definitions of community. George Murdock (1949) gave us one of the earliest definitions of community. Social land is combined with social identity as ''groups of people who normally reside together in face-to-face association." Community is both place and identity. A third descriptor of community is social identity through temporal periods. In this third definition, community consists of a social identity unfolding through time (Arensberg and Kimball, 1965). Examples are secret societies, age cohorts with distinct life courses, or people who share a decisive historic event such as war, the Great Depression, the "Sixties." There are two variations of social identity defined in time. The first is with a shared place and the second is without. A fourth descriptor defining community is function. Community is having a specific basis for a shared identity such as an occupation, a profession, common mission or common craft.
In sum, the classic definitions of community are social identities bound (a) by place; (b) by time; (c) by time and place; (d) by function; (e) by function and time; and (f) by function, time, and place (Arensberg and Kimball, 1965). These types of community are not mutually exclusive, nor are they single dimensions. Different kinds of communities can coexist simultaneously, and of course, individuals can be members of multiple communities at the same time. We are all in some community—our clients come from community, our programs regardless of sponsorship are set up in community, and drug treatment researchers are a "community" as well. At first, the definitions of community appear to have little to do with this committee's deliberations. But in fact they have a lot to do with our mission and highlight the major challenge to bidirectional communication between drug treatment practitioners and researchers. Professionals who work in institutional settings are more apt to define community by function as did our full committee at our first meeting. We defined eight descriptors of community with regard to drug treatment. "Community-based" treatment programs were defined by (1) treatment modality, (2) setting, (3) service units in large organizations, (4) accountability, (5) profit-nonprofit, (6) residential or outpatient treatment, (7) source of funding, and (8) client catchment area. Furthermore, treatment programs are sponsored by hospitals, universities, health departments, corporations, and prisons. These descriptors corresponded with only two classical definitions of community—place (setting) and function. We have not considered temporal period nor have we addressed the most common of all descriptor of community, social identity. Essentially, we have taken the legal and formal organization of treatment programs as bases for community. By focusing on only setting and function, we miss what is "community" for most drug treatment programs. Given our review of community definitions, what then does "community-based" mean?