¹ The section of the Divided Line which is intended to represent the ‘clarity’ of the objects of mathematics has long been seen to be equal to that which represents the clarity of physical objects, but, since Plato does not draw attention to this equality, we cannot know whether it is welcomed by him. Each set of objects appears to be viewed as an im age set only once removed from theldea-archetypes, whilereflections and images in this world are twice removed.

² Here one should mention the theory of Julius A. Elias, Plato’s Defence of Poetry (London 1984), that Plato’s myths not only are akin to poetic literature but actually constitute a defence of poetry. There may be some truth in this, but the present paper is aimed at supplying a theoretical foundation for any such truth.

³ On this topic see Brisson, L., Platon: les Mots et les Mythes (Paris 1982) 81Google Scholar ff. He begins by discussing Critias 107a ff. which is full of the language of image and imitation. (He also uses the comparison in Republic 392c-393c between dramatic [as opposed to narrative] style and imitation, but this is in fact not helpful to an attempt to view myth tout-court as imitation.)The theme of Critias’ words is that he who tells of matters divine has to achieve less precision in his imitating how-things-are than he who tells stories about men, because we have far more accurate knowledge of what men are really like than of real Gods.

⁴ Smith, Janet E., ‘Plato’s Use of Myth in the Education of the Philosophic Man’, Phoenix 40 (1986) 20–34.CrossRefGoogle Scholar

⁵ For we cannot, perhaps, rely on all the dialogues having a predominantly educational purpose at all, let alone accept that each part of the dialogue must have solely educational purposes.

⁶ Cf. Brisson [1982], 93 f.

⁷ So Phaedo 108c8, el, e4, 109a7: πέπ€ΐσμαι.

⁸ See her article ‘Plato’s Myths as “Likely Accounts”, Worthy of Belief’, Apeiron 19 (1985) 24–42Google Scholar, notably 32-34.

⁹ It is worth quoting Brisson [1982], 158: “La valeur de vérité ou de fausseté d’un mythe en est une de second ordre,…”

¹⁰ Here one should mention in particular Julia Annas’ excellent article ‘Plato’s Myths of Judgement’, Phronesis 27 (1982) 119–143CrossRefGoogle Scholar.

¹¹ I do not share Smith’s view that the majority are supposed to adopt myths heard in childhood as beliefs to go unchallenged (22). It seems to me that the indoctrination referred to at Rep. 538c. is a matter of moral learning from parents etc., which is to be obeyed (ηειθάρχαν) and respected rather than believed.

¹² Cf. Rep. 345b, Prot. 314b, Phdr. 235cd.

¹³ See for example SVF 1.20.6, 2.29.37, 3.147.4.

¹⁴ I do not want to enter too deeply into the controversy surrounding the apparent attempt to confine opinion to the world of sensible particulars and to represent it as something intermediate between ignorance and knowledge in Republic V. It will suffice to say that I do not regard Plato’s Republic position as being radically inconsistent with his Meno position.

¹⁵ Compare, from the same work, 526d4,527c5, and πιστεύω at 524a8, all relating to the final myth of punishments in the afterlife.

¹⁶ Griswold, Charles, Self-Knowledge in Plato’s Phaedrus (New Haven/London 1986).Google Scholar

¹⁷ Thrasyllus (D.L. 3.58) saw it as a work on Love, others thought it to be on Rhetoric but are criticized by anon. Proleg. (22), who agrees with the manuscript tradition in seeing it as a work on Beauty.

¹⁸ Note that in the Republic the three types of soul are able to account for the five types of constitution and five types of men in Books VIII-IX.

¹⁹ 26c4-5: note the 5/) and the .

²⁰ Though Ietus not forget that the work refers to itself as an even more frequently: being a myth does not preclude its being a logos.

²¹ Apeiron 19 (1985) 36Google Scholar. A recent discussion of the meaning of εΐκώς μύθος etc. in the Timaeus (and Critias) may be found in Raoul Mortley, Désir et différence dans la tradition platonicienne (Paris 1988) 11–12Google Scholar. He holds that definitions of εΐκώς (= ‘vraisemblable’!) which emphasise closeness to the truth are one-sided; a certain distance is also implied. For myth and imitation in the Timaeus see Hadot, P., Revue de théologie et de philosophie 115 (1983), 113–133Google Scholar. If I seem to pass too readily here from giv ing a deri ν ation of εΐκώς to suggesting a meaning thatisbasedon deriv ation, it is because of Plato’s own tendency to find meanings in derivations as seen in the Cratylus and elsewhere.

²² See Sophist 236a-c for the distinction.

²³ See Meno 80c, Euthd. 275d ff., 276 ff., 295b ff.; also Antisthenes’ work Problems about Knowledge in a seemingly eristic volume.

²⁴ See Vlastos, G., “The Socratic Elenchus’, OSAP 1 (1983) 27–58.Google Scholar

²⁵ Note the emptiness of the aviary-like mental receptacle at birth, Tht. 197c. But I have no confidence that the failure of this work to explain knowledge along empirical lines can demonstrate Plato’s having abandoned non-empirical theories. We simply cannot presume that Plato was using every means at his disposal in order to arrive at a solution, especially in view of the fact that Socrates ‘ methods here mean that all the ideas must emanate from the interlocutor.

²⁶ It is essential to realise that the veibpavBd netais regularly used as a kind of passive of the verb and that it does not ordinarily cover the acts of learn¬ing by discovery or experience: e.g. Prot. 319cl and 320b6-8 . Notice also the substitution of for at Meno 70a3.

²⁷ Strictly speaking, Plato leaves open the possibility that knowledge should be acquired after birth by experience [έμπειρία) such as that acquired by the man who travels the road to Larissa (97a). From 81c6 (έωρακιΐα) it may seem that experience rather than being-taught has been the soul’s manner of acquiring knowledge in its previous existences; Plato is prepared to apply the perfect tense of μανθάνειν for the condition of having-learnt in a previous existence, but this nowhere conjures up the image of classes in the underworld.

²⁸ It is a matter of whether Simmias disbelieves the recollection-theory, or requires only to recollect it and so to positively believe it. Indeed he is already on the point of recollecting and believing (73b3-9; cf. a5-6).

²⁹ Note the continuing belief-terminology at 77a7, a9, alO.

³⁰ Note that Elias [1984] treats Recollection as if it were a Platonic myth (194-8). In my view this going too far.

³¹ Cf. Sharpies, R.W. (ed.), Plato: Meno (Warminster 1985Google Scholar) and Klein, J., A Commentary on Plato’s Meno (Chapel Hill 1965Google Scholar), ad loc.

³² I find no great difficulty in reconciling my view of inspiration with what we read in the Ion about the source of Ion’s inspiration at such times as he recites Homer (533d ff.) We have to contend with the fact that Homer is an inspired poet in the context of this work, and much of the account accords well with what is said about the better poets, seers etc. in the Meno. Of particular importance is the reappearance of the notion of a θεία μοίρα in 534c-535a and beyond. There is, however, no indication that the Ion was written at a time when Plato had already developed a Theory of Recollection.

³³ I am aware that Republic X may seem to some sufficiently uncompromising in its starce against mimetic literature and such mythical materials as itTelails to constitute a threat to the view taken here. But in fact there is no use of the term μύθος and its cognates until the Myth of Er. Whatever the attack, it is not aimed specifically at myth. Even the friends of the poets are afforded a right of reply, and it may be no accident that in handing down his judgement Plato speaks as follows: .

³⁴ There is a duty here to refer to certain works not otherwise mentioned in the notes. I am certain that some of my inquiries have been stimulated by Ferrari, G.R.F., Listening to the Cicadas (Cambridge 1987)CrossRefGoogle Scholar, though it is difficult to say exactly in what respect. One cannot overlook Zaslavsky, R., Platonic Myth and Platonic Writing (Washington 1981).Google Scholar There is also a considerable literature on the Theory of Recollection, but most has a tendency to downplay those features of the theory which are useful to my case. In particular, authors tend to exaggerate the part intended to be played by undiluted logic in the theory, thus assimilating Plato to modern tastes. It seems to me that this is justified primarily by the exploitation of two passages of the Meno (81 cd and 98 a). The former speaks of our ability to recollect all things upon remembering just one, and will suggest to some that all recollectable knowledge can be deduced from one recollectable premise. Plato scarcely forces this view upon us, however. The latter passage is thought to speak of binding one’s correct opinions by ‘calculation of the reason ‘. I have demonstrated that the text here is open to question in The Criterion of Truth (Liverpool 1989Google Scholar), a Festschrift for G.B. Kerferd. I have no objection to the notion that calculation and Socratic elenchus could be important stimuli for recollection. The role of the questioner is clearly vital in the Meno. But the appropriateness of the recollection-simile depends upon the inadequacy of argument alone to generate beliefs rather than mere conclusions.