1. One example is in the translation by Tredennick that is reprinted in the widely circulated The Collected Dialogues of Plato edited by E. Hamilton and H. Cairns. 2. A standard account of the deductive conception of explanation may be found in Hempel and Oppenheim 1953.
3. There is a loosely structured argument at Phaedo 72E-75B by which Socrates recommends acceptance of the existence of the Forms. There are arguments in other dialogues as well for the existence of one or another relevant Form, such as the argument for a unique Form of Bed at Republic 597C. My point is that none of these arguments is deductive in a manner even approximating that required by Cherniss' account.
4. Many of Plato's deductive arguments have the force of refutations, and there is a sense in which a refutation is an "explanation" of error. But this, fairly obviously, is not the sense of explanation at issue in the discussion above.
5. Morrow 1962 reviews the evidence from Cicero up to mid-20th century, and attests that a "great majority of modem students" (p. 16) accept the 7th and 8th letters as genuine, and expresses his own view that the 7th constitutes a stan- dard by which the status of the remainder might be tested (p. 15). The results of a recent stylometric study in Ledger 1989 based on computerized techniques, reinforces Morrow's confidence. As Ledger puts it, "Epistle 7 is nowadays accepted by the majority of scholars and its thought and content has been widely discussed. It hardly seems necessary to rehearse once more the arguments pro and contra. The importance of having its authenticity confirmed is that it will now be possible to rely on the long excursus on the nature of reality (342A-344C) with confidence as a guide to Plato's later thought" (p. 150). Scholars who per- sist in rejecting the authenticity of this letter (e.g., Vlastos 1973, 202-203) tend to do so on the basis of conflict with their interpretation of various passages in the dialogues. At this point, it seems fair to say, the priorities ought to work in the opposite direction, ruling out interpretations that are at odds with the Seventh Letter.
6. The widely accepted claim that these doctrines do not appear in the dia- logues is countered in Sayre 1983. 7. See Kramer 1990. This book is reviewed at length by the present author in Sayre 1993.
8. The term �lnr6v can also mean 'capable of being spoken of' specifically. Read with this sense, 341C6 runs directly counter to the notion of Plato's having passed on "esoteric doctrines" by oral communication.
9. At Symposium 211A2-4, the Beautiful itself is said not to admit ugliness in any respect. At Phaedo 103B4, C6, opposite Forms generally are said never to admit each other.
10. With regard to the instability of names (as at 343A9), see Cratylus, 384D2-8.
11. The expression nepi avuav at 341 C5 echoes the nepi 'toû 1tPåylID'tOÇ at 341C4, which I take to cover philosophic knowledge gained by either means noted at 341C2-3-namely, either from his (Plato's) instruction or from personal discov- ery.
12. See footnote 6. 13. Examples are the opposition between the doctrine of the incomposite soul in the Phaedo (e.g. at 80B) and that of the tripartite soul in the Republic, and Socrates' conclusion that virtue is not teachable in the Meno (99E) in opposition to hisconclusion that it is teachable in the Protagoras (361B).
14. A claim on the author's part never to have written philosophy also appears in the Second Letter (314C), which however is generally considered to be one of the more dubious of the thirteen. 15. The term vo"v5 (for intelligence) at Timaeus 51D3 is the very term used both at Republic 534B5 and at 342C5 of the Seventh Letter, which supports the rele- vance of the Timaeus passage to the present issue.
16. My understanding of the Phaedrus has benefited considerably from Griswold 1986.
17. There are obvious parallels here with Diotima's message in the Symposium, esp.- 212A.
18. This conception of knowledge as mental discourse is extended to other late dialogues in Sayre 1992. 19. The written version of my paper incorporates two substantial passages not present in the version on which Professor Griswold's comments were prepared. Both additions were intended to correct oversights which his commentary made evident, and both are identified in the text of this Reply. Other passages I might have wanted to change as a result of his perceptive observations were left in original form, in order not to deprive certain aspects of his commentary of their target. I want to thank Professor Griswold wholeheartedly for his thoughtful criticisms of my arguments, and for his sympathy with the contents of my paper overall. I am grateful to the Editors also for the present opportunity to address these criticisms more fully.