International Encyclopedia of Systems and Cybernetics

J. MILLER, who has actively promoted cross-level research as a general systems methodology, proposes three degrees of generality for it:

"The first predicts that a formal identity and/ or disidentity is present across individual systems of the same species or type. A research that tests the hypothesis that individual rats differ among themselves in the rates at which they characteristically learn to run mazes is of that sort.

"Cross-type or cross species generalization is the second degree of generality of hypothesis. Such generalizations are more powerful than the first sort. Also they involve more variance. Research in comparative anatomy, comparative physiology, comparative psychology, and coomparative sociology is of that sort.

"The relative size of skulls of rats, cats, dogs, dolphins, apes, and human beings, for instance, can be compared with the average measures of intelligence of each species to confirm the hypothesis that animals with larger skulls – and larger brains- have greater cognitive capacity. Of course there is also variance among individual members of a single species.

"The third type of scientific generalization is from one level to another. These cross-level generalizations will, ordinarily, have greater variance than the other sorts since they include variance among types and among individuals. They can, however, be made. They can enable insights and discoveries made at one level to give rise to general principles which apply to multiple levels of living systems and, in some cases, to nonliving systems and technologies as well. As understanding of systems in general increases, fragments of knowledge may coalesce into patterns and science as a whole may becomes more coherent. Indeed, evaluating the validity of cross-level hypotheses may become a very significant innovation in the study of living systems.

"Scientists undertake cross-level research less frequently than cross-individual or cross-type. There are a number of reasons for this. Scientists are trained to be specialists in one or another of the many disciplines and subdisciplines into which the subject-matter of science is traditionally divided. They limit the scope of their expertise on the grounds that one person cannot be informed in depth and expert on a broader range of phenomena. The usual patterns of rewards to scientists, including academic promotion and tenure, recognition in national scientific societies, grant awards, prizes, and so forth reinforce specialism. What is more, the traditional cautiousness of scientists makes them skeptical of any generalization from, say, a cell, to a system as different as an organism or a society. Scientists working at one level of living systems are rarely current with the literature at other levels. This is demonstrated by the paucity of references to researchers at other levels in the texts or bibliographies of scientific papers. As a result, when a scientist discovers a principle at one level, he typically does not think it likely enough to be relevant to another level to take the trouble to find whether similar principles have been reported in the literature at other levels.

"General living systems theory in no way denies the importance of specialists' knowledge or the advancement of understanding of phenomena in limited scientific areas. Specialized research will properly continue to be the chief means by which science advances. At the same time, however, there should be studies of generalities or similarities, as well as differences, among different classes of phenomena" (1986, p. 74-76).