Franciscan Studies

BURIDAN AND OCKHAM: THE LOGIC OF KNOWING I. A Medieval Disagreement on a Matter of Logic In De Sophisticis Elenchis XXIV (179a26-179b33), Aristotle addresses the following puzzle. It seems that one might find oneself in the position that one does not know the one approaching, although, at the same time, one knows Coriscus, and Coriscus is in fact the one approaching . Yet, according to some, ifone knows Coriscus, and Coriscus is indeed the one approaching, it follows that one does know the one approaching. If all ofthis is granted, however, one in the position described would both know and not know the same thing—viz., the one approaching; but this is impossible. Ofcourse Aristotle has his own solution to the puzzle; but I shall not tackle that topic here.1 Instead, I want to focus on a disagreement between two fourteenth-century philosophers —Buridan and Ockham—over one ofthe steps involved in the generation ofthe puzzle . Briefly, Buridan regards the argument form Al: s knows a a is b .". s knows b 1 Aristotle's treatment ofpuzzles ofthis sort has received a considerable amount of attention in recent years. See, for example, S. L. Peterson's, "The Masker Paradox" (Ph.D. Dissertation, Princeton University, 1969), Nicholas White's "Aristotle on Sameness and Oneness" (The Philosophical Review LXXX 2 (1971): 177-197), Gareth B. Matthews' "Accidental Unities" (Language and Logos: Studies in Ancient Greek Philosophy presented to G. E. L. Owen. Ed. Malcolm Schofield and Martha Craven Nussbaum (Cambridge: Cambridge University Press, 1982) 223-240) and my own "Aristotle on Paradoxes of Accidence " (Ph.D. Dissertation, University of Massachusetts, 1975). 48ANTHONY WILLING as invalid.2 Thus he rejects the view that if one knows Coriscus, and Coriscus is the one approaching, it follows that one knows the one approaching. But Ockham, in contrast, takes arguments of the form Al to be valid. That two such influential nominalist logicians of the same period should differ on a matter such as this is noteworthy. It remains, however, to account for the difference. I shall argue that this logical difference is a reflection of a fundamental epistemological difference. II. Ockham's Position Ockham takes up Aristotle's puzzle in his Expositio super libros Elenchorum .3 Aristotle himself, in De Sophisticis Elenchis, classifies the puzzle as one that involves the "fallacy of accident"; more specifically, he claims to find this fallacy in the inference from 'You know Coriscus' and 'Coriscus is the one approaching* to 'You know the one approaching .' But, as already indicated, when Ockham, in his own discussion of the fallacy of accident, considers this inference, he maintains that it is not an instance of that fallacy; in fact, he contends that it is not an instance of any fallacy—it is a perfectly acceptable argument. Ockham defends his position in this way: For it follows 'You know Coriscus; therefore Coriscus is known by you'; now it follows 'Coriscus is known by you; Coriscus is the one approaching; therefore the one approaching is known by you,' and further, 'therefore you know the one approaching'; therefore, from the first to the last: you know Coriscus; Coriscus is the one approaching ; therefore you know the one approaching. The second argument is clear, since it is an expository syllogism from two singulars in the third figure—just as this is: Socrates is a man; Socrates is an animal; therefore an [at least one] animal is a man—; and it can be proven, since from the opposite of the conclusion and the minor the opposite ofthe major follows, thus: no one approaching is known by you; Coriscus is the one approaching; therefore Coriscus is not known by you. That that syllogism is good is clear, since it is regulated by dici de nul2 The letters 'a' and 'b' in this argument form (and others following) are place-holders for singular terms. I use the verb 'to know' loosely. 3 Ockham, Expos. Elench., lib. II, c. 9 (OPh III, 229 ff.). Buridan and Ockham: The Logic of Knowing49 lo, nor is it in need of anything else for the purpose of appearing necessary.4 This defence ofthe acceptability ofAl rests on three argument forms: A 2 s...