Aspects of Aristotle's Logic (review)

Book Reviews Aspects o/Aristotle's Logic. By Richard Bosley. (Assen: Van Gorcum, 1975. Pp. vi + 137. Dfl. 24.00. Paper) This book is an examination of Aristotle's syllogistic from a novel and in some ways interesting standpoint. Bosley's view is that one can understand arguments properly only by considering them in all their concrete complexity: as controversial exchanges between partners in debate, each of whom has in mind a certain objective. Since he feels that modern logic is "not sensitive enough" to capture certain essential aspects of arguments, the author devotes the first section of his book to the elaboration of a "framework" for the analysis of arguments ("The Framework of the Examination"). Adherence to the model of the dialectical exchange leads him to include various distinctions of speech acts (claiming, asking, disclosing, and--rather curiously---denying) and epistemological states (knowing, being of an opinion, and----again oddly-----concluding , being justified in, being obliged to give up) in the very framework of his analytical machinery. The picture of argument that emerges is much akin to that which seems to stand behind Aristotle's Topics, especially Book VIII: an activity governed by rules (mostly syntactical in character) and directed towards goals (the proof or disproof of some proposition or other). At least, that seems to be the spirit of the following remarks: "It would be natural to wonder whether arguments are practices and, if so, whether, as with games, one might establish rules or regulations which would underwrite the necessity with which one may have to make this acknowledgment or retract that objection. One might think of formalizing the persuasiveness of an argument . But one might also think of arguments as natural processes which have a rather inevitable outcome. But on such a view one is not far from the view that one is carried along by an inner compulsion" (p. 63). Bosley does not make it clear where his sympathies lie with respect to these alternatives , but there is surely merit in the notion that Aristotle's early logical studies (in the Topics) correspond to the first of these tasks. However, such a purely .syntactical view of argument will hardly do for the Prior Analytics, let alone the entire enterprise of logic. It is not the logician's business to record the accepted practices of argumentation --the accepted rules of inference---except insofar as those practices are justified, and the only thing which can justify a rule of inference is the semantical property of validity : a justified argumentative practice is one that never leads one to draw a false conclusion from true premisses. Bosley's avoidance of semantical issues often causes his discussion of logical issues to miss the mark. Perhaps his techniques can reveal many interesting things about argumentative encounters, but they have little to say about straightforwardly logical issues like the relation between the deductive and the evaluative characteristics of a language. Given the historical origins of Aristotle's logical theorizing in the context of Academic debates, Bosley's approach would appear to have more value as a tool for investigating early logic. Indeed, I at least would incline to the view that something like his method is the only promising way to understand the argumentative practice portrayed in Plato's dialogues or described in the Topics. But he has chosen the wrong target: the Prior Analytics just is not that kind of book, as is evident from the definition of o'o~.~.o~/tal~6gas "discourse in which, certain things being posited, something else follows of necessity because o/those things being so [x(oxtxOxct~[vttL]" (24b18-20). This definition obviously concerns relations of propositions, not rules governing dialectical [361] 362 HISTORY OF PHILOSOPHY practice, and Aristotle's frequent appeal to semantical procedures in evaluating syllogistic forms (e.g., counterexamples and ecthesis) confirms this, if confirmation were needed. Several other imperfections weaken this work. Bosley sometimes deals rather oddly with the Greek text. For instance, he translates (apparently) ~.tvog ~Ttoq~ctvztxtg in De Int. 17a8 as "speech which is categorical" (p. 6), despite the definition of that phrase as "sentence to which belong truth and falsehood," i.e., "proposition," half a dozen lines earlier...