The Sciences of Education experience an "epistemological dispersion" (Tardy, 1989), in which they are not alone, for the same happens to the Sciences of Man, both being attached to the program of scientificity that removes people from the place of production of knowledge, be it under the modern paradigm, be it under the postmodern. What is proposed here is a humanist theory of knowledge in which the three kinds of syllogism - enthymeme, dialectical, and demonstrative - are regarded as complementary in the process of establishing knowledge, based on criteria proper to the methodologies or rules of doing, or to algorithms recognized by the social groups that give support to them. Such rules of doing or algorithms do not constitute a natural logic, since any logic is a calculation. The algorithms for the production of knowledge are found in the indissoluble relation between the orator (ethos), the audience (pathos) and the discourse (logos), where the arguments are validated. Hence, a humanist theory of knowledge allows the sublimation of the above-mentioned "epistemological dispersion", for it makes explicit the situations in which a piece of knowledge is validated. Thus, the methodologies are reasonable guarantees, but do not determine reliable knowledge, because its guarantors are the social groups that admit and support it.
Sciences of Man; Sciences of Education; Humanist theory of knowledge; Rhetoric; Rhetoric