Journal of the History of Philosophy

BOOK REVIEWS 391 languages, commentaries, and secondary literature are quite different from the skillsof a philosopher. They should not be put off from benefiting from Guthrie's mind at home in the text by his apparent ignorance of or indifference to what philosophers of the last fifty years have been up to. This volume will replace the Aristotleof W. D. Ross (London: Methuen. 19~~) for anyone making a systematic study of Aristotle. And the chapter on "The Mind of Aristotle" will find its way into anthologies. WILLIAM H. HAY University of Wisconsin, Madison Henry Teloh. The Developmentof Plato'sMetaphysics. University Park: The Pennsylvania State University Press, 1981. Pp. xiii + ~56. $18.75. In view of the growing interest today in Plato's late metaphysical works, one might expect a book by this title to deal extensively with the Sophist, the Parmenides,and the Philebus. Readers who expect this will be disappointed, for Teloh's treatment of these dialogues is spotty and brief. Detailed discussion of the later works is limited to the first part of the Parmenides, and to a few selected passages from the Philebus, the Sophist, the Theaetetus, and the Timaeus (in that order). The second part of the Parmenides is almost completely ignored, on the authority of Ross who concluded it is mere "mental gymnastics." The lion's share of the book is divided in roughly equal portions between the early dialogues (including the Hippias Major) and the middle (primarily the Phaedo and some sections of the Republic). What we find as a result is a series of arguments on interrelated issues (ethical and epistemological, as well as metaphysical) loosely organized by chronological period, rather than a sequential account of Plato's changing metaphysical views. Despite its somewhat haphazard organization, the main themes of the book are boldly delineated. One is that the virtues in the early dialogues (piety, courage, temperance, etc.) are not Forms but rather psychological states. Being identical with the knowledge of good and evil, moreover, each is identical with all the others. Virtues are causes of virtuous behavior, according to the principle that "a cause must have what it transmits or produces" (42). And virtues explain virtuous behavior, insofar as they are "self-predicated" paradigms which one can view inwardly and "see what sort of thing produces just or pious action." An attractive feature of this view is that it makes self-knowledge out literally to be knowledge of the conditions of one's own psyche. By Teloh's reading, the bridge between the early dialogues (without Forms) and the middle (where Forms come to be stressed) is revealed in the contrast between the treatment of the beautiful in the HippiasMajor and that of the Beautiful in the Symposium . In the former dialogue, as in the Gorgias,beauty is the harmony or organization of an object's elements. As the result of a Parmenidean '~jolt" (99; why administered at this point is never made clear), however, Plato in the Symposium begins to treat the beautiful in particular objects as inferior to another Beautiful which is unchanging and separate and wholly pure. On the basis of his examination of this shift, Teloh con- 392 HISTORY OF PHILOSOPHY dudes that there was a "change in kind" (98) not only between the ontologies of the early and middle dialogues but also between their epistemologies. The second major theme is that Plato's epistemology of the middle period incorporates two different models of knowledge, which Teloh labels the "visual" and the "discursive." Although both models portray knowledge as directed toward Forms, "they imply incompatible accounts of what a Form is" ((5). The visual model is based on the metaphor of the "eye of the soul" (lOl), so that knowledge becomes a sort of mental vision. The objects of this vision are uniform (resisting contrary predicates), incomposite (hence indissoluble), and completely isolated (even from other Forms). It is the visual model, accordingly, that generates the standard view of the Forms in the middle dialogues. An interesting variation on this standard view defended by Teloh is (1) that in the middle dialogues there are Forms only for "attributive, relational, and incomplete predicates" (129), but not for sortal predicates, and...