In response to:
The Virtues of Plato from the September 27, 1979 issue
To the Editors:
I am puzzled by Myles Burnyeat’s claim that Terence Irwin leaves the theory of recollection out of his account of Plato, and that he ignores Plato’s belief in the pre-existence of the soul [NYR, September 27]. Since the theory of recollection is so conspicuous in the index, the table of contents, and the text of Irwin’s book, this can’t be an oversight, it must be rhetorical excess. But Burnyeat could have found a better way to express his disagreement with Irwin over which parts of the theory are philosophically fundamental.
The theory has two aspects: (1) We can call up from within ourselves mathematical, moral, and other universal knowledge that could not possibly be derived from sense experience; (2) this knowledge was originally acquired by the soul prior to its incarnation, Irwin explains why he thinks (1) can stand independently of (2), and accords to (1) a central place in the theory of Forms and in Plato’s reaction to Socratic ethics and the Socratic method of inquiry. Burnyeat apparently believes that without (2), (1) does not help to account for Plato’s belief in the separateness of the Forms from the world of sense: that (2) is the essential epistemological premise. I do not agree: if there is innate knowledge of universals which could not be derived from sense, and which lifts us out of our ordinary embodied lives, it is natural to suspect that the objects of that knowledge are not sensible objects. (Plato had other reasons for rejecting the alternative that they are not objects at all.) The theory of pre-existence is a further conclusion of this line of argument rather than a necessary premise. That’s quite clear in the passage at Phaedo 76 to which Burnyeat refers for support.
Department of Philosophy
Princeton, New Jersey
To the Editors:
Myles Burnyeat’s review “The Virtues of Plato” considers the relation between vision and argument. He suggests that since there is both argument and vision in Plato, a book about Plato should be both an argumentative and a visionary book. He says that J.N. Findlay’s book is “all vision without argument,” and that mine is “all argument with no vision.” (See also his last, ambiguous, sentence, which might be interpreted this way too.) Visionary books may be good things; but (like Doctor Johnson) I wonder why a book should be visionary just because it is about a visionary, any more than it ought to be far just because it is about a fat man.
On my view Socrates is a philosopher whose vision outruns his arguments. Socrates thinks it is always good for him to be just, because he wrongly thinks he can show that justice is an effective instrumental means to his own good; no doubt his vision persuades him that he can find the argument. Burnyeat rejects this interpretation because he wants Socrates to have better arguments for his vision of the supreme good of justice. Here Burnyeat is the one who refuses to recognize a vision outrunning argument. I don’t say this to recommend my view of Socrates against his, but only to show that his maxims about vision and argument sound good until they are applied to actual cases; then they are too vague to be helpful.
I see why Burnyeat thinks that my book does not say much about some of the Platonic vision (or some Platonic visions), since it does not say much about some aspects of Plato’s transcendent metaphysics that are not justified by arguments. It does not follow that I present a Plato with no vision. I believe that Plato’s views on morality and its epistemological and psychological implications can be defended by plausible arguments, and that they constitute an impressive and considerable contribution to moral theory apart from the transcendent metaphysics. Such an impressive and considerable contribution seems to me to qualify as a “vision.” Some students of Plato have thought that if you take away the transcendent metaphysics, you have no vision left. I don’t know if Burnyeat thinks this; I don’t.
Burnyeat’s lively rhetoric may raise some reader’s expectations when he says that my account is “dramatically different” from the usual one; “Mature Platonism is root and branch opposed to the Socratic philosophy. The Theory of Forms was an act of parricide.” That sounds good, “heady stuff,” as Burnyeat would say; and it would be a dramatic result. I’m afraid my view is too dull for these vivid and violent metaphors. Plato believes that a defense of Socrates’ most central and important doctrines requires the rejection of some less central and less important doctrines, and the radical revision of others—naturally we may want to dispute Plato’s judgments of relative centrality and importance (see Plato’s Moral Theory, pp. 174, 247f). Admittedly, many randomly chosen pairs of philosophers agree about something. Plato’s agreement with Socrates is rather more significant. He can plausibly argue for his view that he presents the best defense of Socrates’ central doctrines. Mill’s revision of Bentham is radical. It is neither parricide nor root and branch opposition. Nor is Plato’s revision of Socrates.
I cannot discuss all Burnyeat’s searching and suggestive questions about the Theory of Forms. I would like to mention just one. He remarks: “Many philosophers have learned to live with the idea that moral predicates like ‘good’ and ‘just’ are not definable in neutral, empirically decidable terms. Why didn’t Plato? According to Irwin….” The context might suggest to a reader that I think Plato did not learn to live with this idea, and that I try to explain why he did not. But I think Plato did live with it, and that the Theory of Forms shows that he did, not that he did not. Burnyeat’s remark would be understandable if he agreed with a common view of Platonic Forms, that they are meant to be objects of some intuitive acquaintance, a self-certifying quasi-seeing that removes the doubts caused by ordinary empirical objects. I am not persuaded by this view of the Theory of Forms, and doubt whether Plato ever set out on this “quest for certainty” (see Plato’s Moral Theory, p. 159f). I am not sure if Burnyeat means to suggest that the idea about the definability of moral predicates is a rather obvious, even boring, truth. I don’t think it is. Many philosophers have rejected it for plausible reasons; Plato’s position is deservedly controversial.
Burnyeat thinks “Plato’s moral theory is less Aristotelian than Irwin makes it appear, because Irwin is uncomfortable with the transcendent otherworldly dimension which Plato, unlike Socrates, has constantly in view.” I am not sure why I am thought to be “uncomfortable” with this. I do think that some of Plato’s transcendent interests conflict with some of his other aims in the Republic; I don’t know if Burnyeat disagrees. I doubt whether it is fair to contrast an “Aristotelian” and an “otherworldly” theory. Aristotle’s view of the transcendent is different from Plato’s; but the last book of the Nicomachean Ethics seems to me to express a transcendent point of view. Perhaps Burnyeat means to stress the “constantly” in his claim. If so, I doubt if the claim is true for the Republic.
T. H. Irwin
Ithaca, New York
M.F Burnyeat replies:
Professor Nagel is perfectly correct that Terence Irwin’s book has a good deal to say about the theory of recollection. But in every case but one what is spoken of under that name is quasi-recollection (QR in Irwin’s code), which means: arriving at knowledge without being taught, by using internal resources such as one’s own beliefs and reasoning. That which makes Plato’s theory of recollection a theory of recollection, viz. preexistence and prenatal knowledge, is left out. QR does not even imply innateness, let alone preexistence.
The one exception is the footnote on pages 315-316 [Plato’s Moral Theory] where Irwin defends his position that “QR occurs” (a proposition weaker than Nagel’s proposition (1)) can stand and is worth discussing independently of preexistence. With one point in that note I heartily agree: in the Meno the argument to preexistence is an argument to the best explanation of how QR is possible. The Phaedo argument goes similarly: preexistence is arrived at as the only reasonable explanation of (1). Although expounded last, in the order of explanation it comes first. My suggestion, accordingly, was that Plato originally reached the Theory of Forms from (1)—this alone, by the way, I called the essential epistemological premise—plus the thought that preexistence is the only reasonable explanation of it.
For suppose it established that there is innate knowledge of universals which could not be derived from sense (a proposition, be it noted, already stronger than (1)). Modern rationalist philosophers have believed this much without finding any reason to adopt the Platonic Theory of Forms, the theory that Justice and Equality exist on their own separately from their sensible instances. So there must be something wrong with Nagel’s suggestion that innateness is enough, even if I am wrong about what, in Plato’s case, the extra considerations might be. Nagel’s proposition (1) leaves open, for example, the possibility that justice and equality exist as universals present in their instances to which our minds are innately adapted by the grace of God or evolution. Irwin was right to see a problem here, even if he makes things even harder for Plato by having him start, not from (1), but from considerations about the coincidence of opposites.
Now of course (1) can stand independently of (2). So can QR. The question, however—and this brings me to Irwin’s letter—is whether Plato’s view of the moral life isn’t so bound up with his transcendent metaphysics that if you leave out the metaphysics you take away a large part of what is characteristically Platonic in the moral theory. Certainly, what Irwin gives us is an impressive and considerable moral theory, rather like Aristotle’s theory, and it was an impressive and considerable achievement to set it out in the way he has done. But is it Plato?
To stick with recollection for the moment, consider Irwin’s account of the Phaedrus. The really virtuous man, he says (pp. 172-175, 240), appears mad to others because he has come through recollection (= QR) to understand what is really admirable and so he chooses virtue for its own sake, regardless of the consequences. Read the Phaedrus and you find that the man under discussion is mad. He is mad because and in the sense that he is possessed by a divine inspiration, like the poet or seer. But unlike poet or seer, his inspiration comes from something divine within himself, his own powers of reason. To start recollection (in the full Platonic sense) is to start removing oneself from this world and its concerns because it is a recovery from within of that knowledge of true values the full possession of which makes a god divine, and the loss of which is the cause of the pattern of desires characteristic of humanly embodied life. There is a world of difference between QR and Platonic recollection, and that difference is all-important for the Phaedrus‘ brilliant phenomenology of love with and love without recollection of separated Forms.
Irwin is entitled to reply that the vision so strikingly present in the Phaedrus is less obtrusive elsewhere: that will be a matter of emphasis, and my emphasis was admittedly in part a reaction to the very thoroughness with which Irwin sets about what I see as an Aristotelianization of Plato. But all I asked for was some response to the vision, not a visionary book. The analogy is not a thin book about a fat man but a study of Laurel and Hardy which makes no mention of their girth. Suppose someone claimed that their brand of humor can stand, and deserves critical scrutiny, without reference to the contribution of physical size and stature. To my mind, for large areas of Plato’s moral philosophy—not all, I gladly concede—the act of abstraction which Irwin summons us to perform is hardly less strenuous than that.
I accept, however, that I expressed myself misleadingly on the point about Plato learning to live with the idea that moral predicates are not definable in neutral terms. I do happen to think that the claim can only be made good in a version so strong as to be an obvious, boring truth, but other philosophers have thought differently, so let us suppose them right and ask, “What does that show?” Of the many answers one could imagine, one is: “Nothing much, you can’t find a neutral empirical way of deciding questions of goodness and justice, and that’s that.” Plato’s very different answer, on Irwin’s story, is: transcendent Forms of Goodness and Justice. I did not mean to imply that these do after all supply a neutral method of decision, perhaps by some mysteriously intuitive acquaintance; like Irwin, I think this notion of acquaintance should have no place in our account of Platonism. My question was, and is, what makes Plato respond with the second answer less drastic than the metaphysics of transcendent Forms? It seems to me a very large jump that Plato makes, and one that needs more to explain it than the confusion posited by Irwin or the inference which Nagel now offers in its place.
This brings me, finally, to the question how far away from Socrates Plato, having jumped, ends up. I spoke of parricide: Irwin protests that his interpretation is “too dull” for that—and refers us to Plato’s Moral Theory page 174 where his (brilliantly original) interpretation has Plato denounce Socratic virtue as slavish virtue. In Greek terms you cannot get much more savage than that, so if Irwin is right about the target of “slavish virtue,” how substantial is the agreement with Socrates (p. 248, Irwin’s second reference) on the point that virtue benefits the virtuous man? Socrates without the craft analogy (which for Irwin, remember, is at the heart of Socratic ethics), without the paradoxical denial of moral weakness, his account of human motivation radically reshaped by Plato—this Socrates is not alive and well and living in the Academy. He was gradually dismembered in Plato’s middle dialogues and, like it or not, it was Irwin’s considerable achievement to show us in painful detail how little life remains.