Research on teaching can be concerned with teaching in general or it can be conducted with a single kind of subject matter. Which kind of research is more advantageous? This issue arises in the planning of programs of research on teaching. The generic approach looks for uniformities, concepts, and principles that apply across all or many subject matters. The curriculum-specific approach is based on the assumption that subject matter does make a difference--in the kinds of teaching behavior, methods, strategies, styles, and skills that the investigator needs to be concerned with and in the kinds of results that will be found. The arguments concerning the generic-specific issue in research on teaching can be grouped into three categories, each of which are dealt with in this paper: (1) those dealing with the generality of learning and teaching processes; (2) those dealing with the problem of multiplicity; and (3) those dealing with conceptualizations of the pedagogical domain. After examining these three areas, this paper proposes that research should be concentrated in a relatively small number of conjunctions of grade-level, subject matter, and socioeconomic level, i.e., the research approach used should be specific rather than generic. (JMK)