The Thomist: A Speculative Quarterly Review

DUMMETT ON FREGE: A REVIEW DISCUSSION* IT IS BY HIS EXPOSITION and development of Frege's thought that Michael Dummett may well turn out to have won a permanent place in the history of philosophy. For Frege is unquestionably the greatest logician and philosopher of logic since the days of Aristotle and the Stoics; and as an interpreter of Frege's thought Dummett stands absolutely without a rival. A proper appreciation of Frege did not come till many years after his death. Many of his technical achievements soon entered into the common stock of logicians; but few writers in the field gave due credit to Frege-Quine in his Mathematical Logic is an honorable exception. It took much longer for people to study Frege's philosophy of logic, which underlay his technical work, and to appreciate Frege in his own right and not just as a forerunner. During the last thirty years there have been many articles and not a few books devoted to Frege's philosophy; but a good deal of this material has been written by logically and philosophically incompetent authors, or else by people (like Church and Carnap) who have been so possessed by certain ideas of their own that their image of Frege's thought was quite distorted. Dummett's exposition of Frege is a mighty achievement; he must know Frege's works almost by heart, and he is almost invariably faithful to Frege's letter and spirit. When he purports to be developing Frege's thought rather than expounding it, what he writes bears the mark of an authentic development. As we shall see, he has his faults, but as an expositor of Frege nobody can compete with him in scope and accuracy and • acumen. * Michael Dummett, Frege: Philosophy of Language (Cambridge: Ilarvard University Press, 1981) ; Truth and Other Enigmas (Cambridge: Harvard University Press, 1979); The Interpretation of Freglfs Philosophy (Cambridge : Harvard University Press, 1981). 116 DUMMETT ON FREGE: A REVIEW DISCUSSION 117' The Interpretation of Frege's Philosophy is largely devoted to polemic against other interpreters. We may well regret this diversion of Dummett's energies; he has postponed the completion of a volume on Frege's philosophy of mathematics in order to refilte misconceptions about Frege's general philosophy of language. When Aquinas had finished one Aristotelian commentary , he began another, without turning aside to castigate other commentators; Dummett might well have followed this example. In particular, Dummett had no need to expound and refute in detail the work of men who plainly despise Frege; their books will soon pass to the limbo of the cheap bookstore, where I have seen other books of this genre, e.g. one 'proving' that Plato was a charlatan. Dummett's work on Frege has one great negative fault; he virtually ignores the light thrown on Frege by Wittgenstein's Tractatus. Some of Frege's deep insights were to be presented anew in the Tractatus, more sharply and clearly than Frege had presented them, and without certain errors (e.g. about the relation of sentences and names) that affected Frege's mature work. On some points, indeed, where Frege and Wittgenstein disagree, one might well prefer Frege's views; even so, the criticism of Frege in the Tractatus can never safely be ignored. Dummett hardly ever alludes to the Tractatus, and when he does so he gives us weak objections based on manifest misunderstandings . I have space only to discuss one particular example: negation . Wittgenstein pointed out the possibility of passing froni a given language, say English, to an alternative language, let us call it Unglish, by the following rule: any English sentence S has as its Unglish translation a sentence spelled the same way as an English sentence contradictory to S. We need not content ourselves with this rule for whole sentences: we could easily construct an English-Unglish/Unglish-English dictionary comformable to this rule, so that the translations always worked out the right way; it would be a matter of what modern logicians call duality. Now without considering further how such a dictionary would work, a very little thought shows that 118 P:. T. GEACH the Unglish for·' not' would be 'not'; for in...