Styles of Thinking
The fashion in question concerns one of the dimensions of cognitive style--roughly speaking, the use which is made of formal logic and mathematics. Note that this is a question of the style of thought, not merely of the style of presentation: Plato and Galileo both wrote dialogues, but with very different cognitive styles. Yet thought and its expression are surely not wholly unrelated to one another, and how scientific findings are formulated for incorporation into the body of knowledge often reflects stylistic traits of the thinking behind them. . . .
1. The literary style. This cognitive style is likely to be occupied with individuals, particular sets or sets of events, case studies, clinical findings, and the like. A plot is unfolded--a behavior sequence is disclosed to have a certain significance. . . . A person, a movement, or a whole culture is interpreted in terms of the specific purposes of the actors, rather than in terms of the abstract and general theories of the scientist's own explanatory scheme.
Examples: Kenneth Burke, Sigmund Freud, Marshall McLuhan, Lester Thonssen and A. Craig Baird
2. The academic style. This is much more abstract and general than the literary style. There is some attempt to be precise, but it is verbal rather than operational. Ordinary words are used in special senses, to constitute a special vocabulary. . . . The materials dealt with tend to be ideational rather than observational data, and their treatment tends to be highly theoretical, if not, indeed, purely speculative. System is introduced by way of great "principles", applied over and over to specific cases, which illustrate the generalization rather than serve as proofs of it. . . .
Examples: Hugh Duncan, some of Burke, some of Leon Festinger, John Stewart, Irving Janis, Jack Gibb
3. The eristic style. Here there is a strong interest in proof, and of specific propositions, rather than, as in the literary and academic styles, the aim only of exhibiting the cognitive possibilities in certain broad perspectives on the subject-matter. Experimental and statistical data become important.
Examples: Gerald Miller, Martin Fishbein, Robert F. Bales, James McCroskey
4. The symbolic style. The subject-matter is conceptualized from the outset in mathematical terms. Both problems and solutions are formulated, therefore, in a more or less artificial language. . . . Measurement is important, as providing the content for the mathematical forms which are employed. Statistical data do not serve, as in the eristic style, only as a body of evidence; they are processed so as to generate new hypotheses, and even new patterns of conceptualization. In this processing, computers and other instruments, both physical and ideational, are likely to play a major role. The symbolic style is characteristic of mathematical economics, psychometrics and sociometrics, game-theoretic treatments of political problems, probabilistic approaches to learning theory, and so on.
Examples: Milton Rokeach game-theoretic models, John Thibaut and Harold Kelley
5. The postulational style. This has many of the characteristics of the symbolic style, of which, indeed, it could be regarded as a special variant. It differs from the symbolic style in general only as logic differs from mathematics. The validity of proof is at the focus of attention here, rather than the content of the propositions which occur at the various steps. Emphasis is on the system as a whole, bound together by the chains of logical derivation.
Examples: Claude Shannon, Charles Osgood (congruity hypothesis)
6. The formal style. This is very close to the postulational style, and indeed, presupposes the latter. The difference is that here the key terms are not given any interpretation; there is no reference to any specific empirical content.
Examples: geometry, algebra, calculus, statistics