1 Guthrie, W.K.C., A History of Greek Philosophy, Vol. IV (Cambridge 1975), 489.Google Scholar

2 See Guthrie, IV, 489-93. Cf. Bluck, R.S., Plato's Meno (Cambridge 1961), 35Google Scholarff., and Cushman, R.E., Therapeia: Plato's Conception of Philosophy (Chapel Hill 1958)Google Scholar for similar views. Guthrie refers to the controversy regarding the continuity and discontinuity between δόζα and knowledge in the Meno and Republic at IV, 489, n. 3. See also Sprute, Jürgen, Der Begriff der DOXA in der platonischen Philosophie (Gottingen 1962)CrossRefGoogle Scholar and Ebert, Theodor, Meinung und Wissen in der Philosophie Platons (Berlin 1974)CrossRefGoogle Scholar, Chs. 3 and 4.

3 See, for example, the use of οἰέται at 82el0; ἔμοιγϵ δοχϵῖ οὕτω at 83dl; δοζάζων at 97bl. Also Plato frequently uses cognate forms of the verb ‘to agree’ to express belief, e.g., ὡμοƛογήχαμϵν at 97al.

4 This is clear from an examination of the texts cited in note 3. E.g., the grammatical object of δοζάζων at 97bl is the clause ἥτιζ ἐστὶν ἡ ὁδὀζ.

5 It is a matter of some controversy whether Plato, prior to the Sophist, distinguished precisely and for logical and epistemological purposes the roles of subject and predicate. For defense of such a distinction earlier than the Sophist, see Fine, G., ‘Plato on Naming,’ Philosophical Quarterly, 27 (1977) 289–301CrossRefGoogle Scholar. For our current purposes, however, I think that this issue can be avoided. If the agent of an act of believing takes his belief to be about something outside of that belief, then this something can be named or pointed to either by the belief-statement as a whole or by its subject term.

6 Cf. 86bl-4.

7 Cf. ἐνέσονται αὐτῷ ἀƛηθϵῖζ δόζαι, 86a6-7.

9 In contrast to Rep. V, where δόζα is explicitly called a δύναμιζ, a power or capacity. On dispositional and occurrent senses of belief, see DeSousa, Ronald B., ‘How to Give a Piece of Your Mind: or, The Logic of Belief and Assent,' Review of Metaphysics, 25 (1971-72) 52–79Google Scholar.

8 For an interesting discussion of the role of questions in understanding belief in the Meno and the paradox of inquiry, see Lawrence Powers, H., ‘Knowledge by Deduction; Philosophical Review 87 (1978) 343-52.CrossRefGoogle Scholar

10 For discussion, see Hintikka, J., ‘Knowledge and Its Objects in Plato,’ Ajatus, (1971) 168–200Google Scholar, especially 177-80.

11 It is Just this feature or set of features that Plato returns to when, at Tim. 51e, he sets out reasons for distinguishing νοῦζ from δόζα ἀƛηθήζ.

12 Most likely, then, the diversified questioning at different times, mentioned by Plato at 85c10-d1, is intended by him as a pragmatic device for moving someone from true belief to knowledge. On such a reading, this questioning is not identical with the ‘bond’ or αἰτίαζ ƛογισμῷ that conceptually distinguishes one from the other.

13 I have in mind, of course, the well-known treatments of falsity in the Theaetetus and the Sophist.

14 Cherniss, Guthrie, and Gulley are strong advocates of Forms as the objects of recollection in the Meno. See Cherniss, Harold, ‘Review of Klara Buchmann, Die Stellung des Meno in der platonischen Philosophie,’ American Journal of Philology, 58 (1937) 498Google Scholar; ‘Plato as Mathematician,’ Review of Metaphysics, 4 (1950-51) 421; Guthrie, IV, 249 n. 3, 253-4 n. 3; Gulley, , ‘Plato's Theory of Recollection,’ Classical Quarterly, 3-4 (1954) 195-7Google Scholar. Neither the textual nor the philosophical arguments for the Forms, however, are secure. Meno 86a-b, when carefully examined, turns out to be a dialectical argument the premises of which do not unambiguously point to Forms as the objects of only disembodied knowing but indicate as well that the objects of knowledge are also those of true belief and that they are in the soul. Furthermore, as White, (Plato on Knowledge and Reality [hereafter PKR) [Indianapolis 1976], 59Google Scholar n.13, 54 n.5, 59 n.35) and Irwin, (Plato's Moral Theory [hereafter PMT) [Oxford 1977], 316Google Scholar n.15, 315 n.12) have shown, belief and knowledge have a very extensive scope in the Meno. Nor is the paradox of inquiry a problem about a priori learning or about a regress that might develop if all learning were thought to be empirical; rather it is a paradox about self-initiated or self-appropriated inquiry that is restricted not by the character of the subject matter but only by the resources available to the student when the inquiry begins. With Ross, I find it an impressive fact that nowhere during the discussion of the paradox of inquiry, the doctrine of recollection, and thereafter are the transcendent Forms mentioned. The reason for their absence is neither neglect nor design; it is simply that Plato had not discovered them yet - or at least no dialogue prior to the Phaedo decisively testifies to that discovery.

15 White has a helpful note on this distinction in the dialogues, PKR, 54-5 n.8.

16 On the broad view of the objects of belief and knowledge in the Meno and on Plato's neutrality regarding the specification of such objects, see White, PKR, 48-9, 59 n.35; Irwin, PMT, 315 n.12, 316 n.15. The narrower, a priori view is advocated by Vlastos, Gregory, ‘Anamnesis in the Meno,’ Dialogue, 4 (1965) 154-7Google Scholar; Gulley, , ‘Plato's Theory of Recollection,’ 194-5, and Plato's Theory of Knowledge (London 1962), 14–15Google Scholar; Guthrie, IV, 261.

17 At Meno 77-78.

18 See Phaedo 65bff., 74aff., 78bff., and 100bff. Cf. Gallop, David (trans. & notes). Plato Phaedo (Oxford 1975), 93-7Google Scholar.

19 See Harold Cherniss, The Philosophical Economy of the Theory of Ideas,' reprinted in Allen, R.E. (ed.), Studies in Plato's Metaphysics (London 1965)Google Scholar; Nehamas, Alexander, ‘Plato on the Imperfection of the Sensible World,' American Philosophical Quarterly, 12 (1975) 105-17Google Scholar; White, PKR, 5-10; Vlastos, Gregory, ‘A Metaphysical Paradox,’ Proceedings and Addresses of the American Philosophical Association, 39 (1966) 5–19CrossRefGoogle Scholar; ‘Degrees of Reality in Plato,’ in Bambrough, Renford, ed., New Essays in Plato and Aristotle (London 1965)Google Scholar, both reprinted in Vlastos, , Platonic Studies (Princeton 1973)Google Scholar.

20 In the terms we introduced earlier, the two cannot have the same external object. Since Plato has not yet clarified the precise role of the external object as the object of the agent's reference, the separation of objects seems to open up a chasm between belief and knowledge.

21 Cf. Phaedo 78b4-84b8, especially 79a1-d8.

22 See Gallop, Plato Phaedo, 146; Burnet, John, ed., Plato's Phaedo (Oxford 1911), 78,CrossRefGoogle Scholar ad loc.

23 Cf. Rep. VII, 519b and X, 611e-612a. The philosopher avoids the dictates of pleasure and pain, i.e., he avoids treating pleasure as the good, because to do so would lead to treating those objects that bring pleasure as the most real. In the Phaedo, then, hedonism is rejected at least in part because it leads away from knowledge and wisdom.

24 On ἀƛηθήζ, see Gallop, Plato Phaedo, 144.

25 Later, Forms are introduced as αἰτίαι (l00bff.). But never in the Phaedo does Plato seem to appreciate the difficulties surrounding the notions of participation, presence-in, and copying, notions that are intended to characterize the relation between Forms and physical objects.

26 Education also insures the continuance of Justice and the other virtues from one generation to the next. That is, it insures that Justice is a permanent possibility.

27 While it is a primary credential, it is of course not the only one; other features of the philosopher's character and desires are relevant as well. Similarly, while it is a necessary condition for the highest kind of well-being, knowledge is not necessary for lesser forms of well-being, or derivative forms (cf. IX, 590c2-d6).

28 Similarly, the first task of the Socratic elenchos is to awaken in the interlocutor a realization of his own ignorance. Cf. Meno 84a3-d2.

29 Strictly speaking, belief and knowledge in Rep. V-VII have as their scope what F is. See Mills, K. W., ‘Plato's “Non-hypothetical Starting Point”,’ Durham University Journal, (date not known) 152-9Google Scholar.

30 A recent, meticulous account can be found in Fine, G., ‘Knowledge and Belief in Republic V,’ Archiv für Geschichte der Philosophie, 60 (1978) 121-39CrossRefGoogle Scholar. Other valuable treatments include: Allen, R.E., ‘The Argument from Opposites in Rep. V,’ Review of Metaphysics, 15 (1961) 325-35Google Scholar; Gosling, J.C., ‘Doxa and Dunamis in Plato's Republic,’ Phronesis, 13 (1968) 119-30CrossRefGoogle Scholar; idem, ‘Republic, Book V: ta polla kala, etc.; Phronesis, 5 (1960) 117-28; J. Hintikka, 'Knowledge and Its Objects in Plato'; Vlastos, ‘Degrees of Reality in Plato.'

31 The non-philosophers are the ‘lovers of sights and sounds.’ See Rep. V, 475dl-476d6. Cf. Phaedo 65e-67b, 78b-84b; Hippias Major 298aff.

32 It may also seem natural to seek such an account in the image of the Cave. But the Cave, as Plato explicitly says (VII, 514al-2), is a likeness or simile for παιδϵία. It is better, I think, to turn first to the curriculum and the very substance of that παιδϵία and only later to return to the Cave for confirmation.

33 Cf. V, 476d5; VI, 508e4; VI, 510d6; 5llal; VII, 533d-e.

34 Some will object that in the Line διάνια is clearly situated in the upper section of the whole and so must be a kind of νόησιζ or ἐπιστήμη and not of δόζα. Technically speaking, this is correct. Perhaps it would be more precise to say that διάνοια is a kind of thinking but one that is akin to belief in some ways and to knowledge in others. I choose to call it a kind of belief, however, because it is then clear that διάνοια involves some kind of assent or commitment, while mere thinking does not.

35 For samples of such discussion, see the articles cited in notes 19 and 30.

36 Exactly how this works is not our present concern. According to Plato's description, it involves the ‘joint presence of contraries’ in a single perception. What this means and how it leads to aporia (VII, 524e5) are matters of dispute.

37 In this paper I do not develop any direct argument concerning whether for Plato knowledge of the Forms is a direct mental grasping or propositional or both. For discussion, see the papers cited in note 30. Indirectly, some of things I say do give support to the view that in some sense knowledge of the Forms is both propositional and non-propositional.

38 Plato discusses, but does not really distinguish carefully, both an act of mathematical thinking and the process of mathematical thinking. My concern here is only with the former. For that reason, we can avoid any detailed consideration of the hypothetical character of mathematics and its relation to the upward and downward paths of dialectic.

39 This may very well be a direct allusion to the Meno and the dialogue with the slave-boy, the topic of which is a theorem about squares and diagonals.

40 There is a vast literature on the Cave, together with the Sun and the Line. I have found the following to be especially helpful: Neil Cooper, The Importance of Διάνοια in Plato's Theory of Forms,’ Classical Quarterly N.S. 16 (1966) 65-9; Ferguson, A.S., ‘Plato's Simile of Light,’ Classical Quarterly, 15 (1921) 131-52CrossRefGoogle Scholar and 16 (1922) 15-28; Fogelin, Robert, Three Platonic Analogies,' Philosophical Review, 80 (1971) 371-82CrossRefGoogle Scholar; Hackforth, R., ‘Plato's Divided Line and Dialectic,’ Classical Quarterly, 36 (1942) 1–9CrossRefGoogle Scholar; Malcolm, John,The Line and the Cave,’ Phronesis, 7 (1962) 38–45CrossRefGoogle Scholar; Murphy, N.R., The “Simile of Light” in Plato's Republic,’ Classical Quarterly 26 (1932) 93–102CrossRefGoogle Scholar; idem, , The Interpretation of Plato's Republic (Oxford 1951)Google Scholar, Chs. VI-IX; Raven, J.E., ‘Sun, Divided Line, and Cave,’ Classical Quarterly, N.S. 3 (1953) 22–32CrossRefGoogle Scholar; Robinson, Richard, Plato's Earlier Dialectic 2nd ed. (hereafter PED) (Oxford 1953)Google Scholar, Chs. X-XI; Stocks, J.L., The Divided Line of Plato Rep. VI,’ Classical Quarterly, 5 (1911) 73–88CrossRefGoogle Scholar; Tanner, R.G., ‘Διάνοια and Plato's Cave,’ Classical Quarterly, 20 (1970) 81–91CrossRefGoogle Scholar.

41 VII, 515a4-5; cf. the excellent accounts of Robinson, PED, Chs. X-XI and Murphy, The Interpretation of Plato's Republic, Chs. VI-IX.

42 One might take this to be contrary to Plato's explicit wishes, expressed at VII, 517a-b. For a critique of the parallelism between the Line and the Cave, together with an alternative reading of VII, 517a-b, see Robinson, PED, Ch. XI, especially 185-90.

43 Rep. VI, 515c1-2; VII, 538c-539a; cf. passages listed in note 31.

44 At VII, 532e1 Plato refers to the stages (ϵἴδη) of dialectic.

45 VI, 511b3-d2; cf. VII, 532d8-535a2.

47 At Met. A, 987b14ff., for example. For recent discussion of the Aristotelian evidence, see Julia Annas, ‘On the ‘1ntermediates”,’ Archiv für Geschichte der Philosophie, 57 (1975) 148-56.

46 See citations in note 40 and Morrison, J.S., ‘Two Unresolved Difficulties in the Line and Cave,’ Phronesis, 22 (1977) 212-31CrossRefGoogle Scholar.

48 White, PKR; Gulley, PTK; Guthrie, IV; Robinson, PED; Brentlinger, John A., 'The Divided Line and Plato's ‘'Theory of Intermediates”,’ Phronesis, 8 (1963) 146-66CrossRefGoogle Scholar, all deny the need for mathematicals in the Republic. Even Wedberg, a serious advocate of a Platonic theory of intermediates in the tradition of Adam and Hardie, is extremely cautious about the presence of that theory in the Republic (Wedberg, A., Plato's Philosophy of Mathematics [hereafter PPM] [Stockholm 1955]) 109Google Scholar; cf. Adam, James, The Republic of Plato [Cambridge 1902; repr. 1965). II, 159-63Google Scholar; Hardie, W.F.R., A Study in Plato [Oxford 1936]Google Scholar, Ch.VI).

49 For discussion, see Hardie, Study, 50-52; Brentlinger, ‘Plato's Divided Line ….,' 156; White, PKR, 110 n.32; Robinson, PED, 193; Stocks, The Divided Line ….'

50 The expression ἤ ὅσων οἱ παρϵƛηƛυθότϵζ at VII, 534a8 refers back to VI, 510c -d.

51 Brentlinger's claim, that 534a by itself indicates an ontological difference between the objects of διάνοια and those of νόησιζ (156), is simply a non sequitur.

52 Wedberg, PPM, 106. For a similar sentiment, see Hardie, Study, 51 and Brentlinger, 'Plato's Divided Line … ; 158-9.

53 For decisive evidence, see Brentlinger, ‘Plato's Divided Line …,’ 151 n.2, contra Robinson, PED, 199-200.

54 In setting out these beliefs about δυνάμϵιζ (cf. V, 477c6), I am indebted to Fine's extremely careful and provocative discussion. See Fine, ‘Knowledge and Belief in Republic V,’ 127.

55 See Fine, ‘Knowledge and Belief … ,’ 128-9.

56 Nor can it enable us to hear them.

57 See Adam, II, 157 (cf. Rep. VII. 532b-c); Wedberg, PPM, 105-6; Brentlinger, 'Plato's Divided Line … : 156-7.

58 Hardie, Study, 58-60.

59 On this reference to Forms as indirect or for-the-sake-of objects of διάνοια, see the inadequate remarks of Adam, II, 68-9. Cf. Robinson, PED, 197 and Brentlinger, 'Plato's Divided Line …; 149.

60 See Robinson, PED, 190-2.

61 See especially Wedberg, PPM, Chs. IV-V; Brentlinger, ‘Plato's Divided Line …; 159-62; Hardie, Study, 49-50. For a contrary view, see Cherniss, Harold, The Riddle of the Early Academy (California 1945). 75-8Google Scholar and Aristotle's Criticism of Plato and the Academy (Maryland 1944), Appendix VI.

62 Wedberg (PPM, 123-4) takes the plurality of units, referred to at 526a, to be the units that compose the mathematical numbers; Hardie (Study, 53) holds that these units are the mathematical numbers themselves.

63 See Shorey II, 164-5 ad loc.; Cherniss, Aristotle's Criticism …, 497 and especially 518.

64 For helpful comments on earlier drafts and discussion of issues in the paper, let me thank John Ackrill, Richard Kraut, Romane Clark, and the editors of the Canadian Journal of Philosophy. In thinking about the role of agent-reference in Plato's epistemology, I have been greatly helped by the work of Hector-Neri Castañeda. Almost singlehandedly he has produced an exciting literature on the problems of reference within propositional attitudes with special attention to the roles of, what Castañeda calls, indicators and quasi-indicators. Plato's thinking about matters such as agent-reference is very rudimentary, but it does play a role, as I have tried to show, in his central epistemological project, the inquiry into what is real. i.e., into what F is, and the acquisition of knowledge of the Forms.