¹ Greimas, A.J., Structural semantics (1983) 173Google Scholar.

² This paper was begun during a research project at King's College London on Greek images for language and thought. I am grateful for the award of a British Academy Postdoctoral Fellowship, which made this research possible. During my fellowship at King's I was fortunate to benefit from discussion with M.S. Silk and M.M. McCabe. I would like to thank the audience at the Cambridge Philological Society for their very useful corrections and insights on this paper, and my colleagues. Dr D.L. Cairns, Dr M.F. Heath and Dr R.W. Brock, for their invaluable comments, criticisms and encouragement.

³ On the power of words in Athens see Vernant, J.P., The origins of Greek thought (1982) 49–50Google Scholar; Buxton, R.G., Persuasion in Greek tragedy (1982) 4Google Scholar, and Entralgo, P. Laín, The therapy of the word in classical antiquity (trans. Rather, L.J. and Sharp, J.M.) (1970) 240–1Google Scholar.

⁴ Aeschylus, , Seven against Thebes 563Google Scholar (‘the word pierces my breast’) and Suppliants 466 (‘I hear a threat that is a lash to my heart’). For Aeschylus' use of metaphors for logos see the excellent account of Sansone, David (Aeschylean metaphors for intellectual activity, Hermes Einzelschrift 35, 1975)Google Scholar. Particularly illuminating are his comments on logos as ‘something organic’ (34–9 and 77–92).

⁵ Gorgias, , Encomium of Helen 8Google Scholar.

⁶ For argument as a journey see Critias 106a (‘How delighted I am … – like a person at the end of a long journey – now I have gladly left off from the journey of discourse’) and Laws 812a (‘It seems to me that we have not travelled beyond the original discussion’). See also Lys. 213e; Rep. 484a. 532e; Laws 688e and 779d. The image of the journey in Greek thought is discussed by Becker, O., Das Bild des Weges, Hermes Einzelschrift 4 (1937)Google Scholar.

⁷ See e.g. Phil. 15d, Soph. 246c and Tht. 179d.

⁸ See e.g. Euthyd. 277d; Hipp. Maj. 286d, 287a; Phil. 22d; Soph. 246c, 261a; Rep. 348a and 501c.

⁹ Tht. 154e ἐϰϱούομεν (‘after coming to grips in such a sophistical battle, we are striking each other's words with words’); Euthyd. 294d , (‘but they bravely went to meet the enemyquestions, in agreement that they knew the answers, like wild boar thrusting themselves to meet the blows’); Symp. 219b (‘But indeed after hearing and after saying these things, I thought that after letting loose my arrows, as it were, I had wounded him’); Symp. 189b (‘But, Aristophanes, you intend to throw your weapon and then to run away’) and Tht. 180a (‘But if you ask someone something, drawing out – as it were from a quiver – little enigmatic phrases, they shoot them at you like arrows, and if you seek to grasp the meaning of one man's account, suddenly you are struck by another newly-made metaphor’).

¹⁰ At Hipp. Min. 369c Hippias charges Socrates with ‘getting hold’ of the details (ἐφαπτόμενος) and failing to ‘grapple with the matter as a whole’ ().

¹¹ At Phil. 22e–23a in the debate on the relative benefits of pleasure and reason, Socrates' argument against the benefits of pleasure is characterized as a boxing-match between him and his logoi on one side and pleasure on the other. Notice how in this passage both the speaker (Socrates) and the logoi are jointly responsible for the knock-out blow: (‘Well, Socrates, it does now seem to me that pleasure has fallen to you, knocked out, as it were, by your arguments just now’). Compare Gorg. 462a where a logos has been knocked off its feet (the speakers are presented as wrestlers at 461c–d).

¹² The boxing metaphor here is part of Plato's wider image of argument as a combat sport which is common in the dialogues. It is well established that Greek authors made great use of combat imagery (see e.g. Poliakoff, M., Combat sports in the ancient world (1987) 32 and 52)Google Scholar and critics have further pointed out that combat imagery was particularly associated with the sophists. O'Regan, Daphne (Rhetoric, comedy and the violence of language in Aristophanes' Clouds (1992))Google Scholar discusses the sophistic use of combat terminology for debate and Protagoras' punning book title (‘Overthrowing arguments’), viewing them as the inspiration for Aristophanes' presentations of violent language (see e.g. 11 and 124). Likewise, Hermann, F.G. (‘Wrestling metaphors in Plato's Theaetetus’. Nikephoros 8 (1995) 77–109)Google Scholar has observed that ‘wrestling metaphors in the context of discussion and argument seem to be particularly connected with Protagoras’ (106) and has highlighted the references to Protagorean terminology at Euthyd. 286c, where Socrates refers to the argument of the Protagoreans which not only ‘trips up’ (ἀνατϱέπων) the other arguments but also itself, and at 287e4, where the argument ‘in throwing down falls itself’. What has not been especially noted is that Plato follows Protagoras (and indeed Aristophanes' satire) in making the logoi themselves combatants rather than simply weapons in the hands of the interlocutors. What is of interest to me in the Euthyd. passages is how the logos is an active force with its own power of movement (286b–c) αὑτόν; (288a) .

¹³ That honour is at stake in the elenchos has been noted by Cairns, D.L. (Aidôs (1993) 371 n.82)Google Scholar, who cites passages in Plato where dialectic is shown to require bravery and perseverance (e.g. Euthyd. 294d and Charm. 160d–e) and where the elenchos is viewed as potentially humiliating (e.g. Euthyd. 295b and Charm. 169c).

¹⁴ The alternative scenario of a logos escaping from a speaker is presented at Phdo. 89c (‘if I were you and the argument (the point) were to escape me/get away from me’). Cf. Parm. 135d6 (, ‘Otherwise the truth will escape you’).

¹⁵ The phrase ‘discursive space’ is used by Adi Ophir in Plato's invisible cities, discourse and power in the Republic (1991) 124–31Google Scholar, and is explained thus (124): ‘The dramatic space is the one unfolded and organized by and through the movement of the dialogue's interlocutors. The discursive space is the one unfolded and organized by and through the movement of the dialogue's arguments.’ In chapter 5 ‘The space of discourse’ (132–67), Ophir discusses spatial and journey metaphors for argument in the Rep.

¹⁶ Paul Shorey's translation of Rep. appears in Plato, collected dialogues, Hamilton, E. and Cairns, H. (edd.), Bollingen Series LXXI (1961)Google Scholar. Other translations used from this collection are: Lane Cooper (Euthyphro); W.K.C. Guthrie (Meno) and F.M. Cornford (Theaetetus).

¹⁷ See Cairns, , Aidôs 15Google Scholar n. 15 on ‘averting the gaze or seeking to hide oneself’ in the phenomenology of shame and aidôs, and 292–3 on the act of veiling: ‘The veiling of one's head is a typical aidôs-reaction. a consequence of the fear of being seen and part of the general complex of associations between aidôs and the eyes.’ Also on veiling as an act of modesty and shame see Cairns, ‘“Off with her αἰδώς”: Herodotus 1.8.3–4’, Classical Quarterly n.s. 46 (1996) 78–83, esp. 79–80CrossRefGoogle Scholar.

¹⁸ The verb ϰινεῖν is used similarly at Polit. 277d5 and 297c8. See Rowe's note on Polit. 272d5 and the idea of ‘moving’ or ‘waking up’ (ἐγείϱω) an account (Rowe, C.J., Plato: Statesman (1995)Google Scholar).

¹⁹ Rep. 503a echoes Pindar Nem. 5.14–19. The passages share both the context of risk and the idea of a thought or argument choosing to remain unseen. In Nem. 5 the poet declares: ‘I shrink from telling of a mighty deed, | one ventured not in accord with justice’ and elaborates: (‘I will halt, for not every exact truth | is better for showing its face’; tr. Race, W.H., Pindar II (1997))Google Scholar.

²⁰ Compare Phaedrus 229d.

²¹ Rowe, C.J., Plato: Phaedrus (1986)Google Scholar.

²² In The shape of Athenian law (1993), S.C. Todd explains the procedure of a diamarturia (136): ‘Diamarturia is a formal assertion of fact by means of a witness. The witness is produced before a magistrate during the preliminary stages of a trial, and the fact to which he testifies is one which has a binding legal impact on the course of that trial … The ideal use of diamarturia is clear: its effect is to bar the original hearing, because the witness is deemed to be telling the truth until the opposite is proved; if the plaintiff wishes to proceed further, he must sue the witness in a dikê pseudomarturiôn.’ In Plato's image the question, ‘does the man who knows need the science of rhetoric?’, constitutes a court case involving personified rhetoric. Before the formal hearing of this case certain arguments present themselves as witnesses and offer an obstructive plea (‘Rhetoric is lying when she says that she is a science’), designed to prevent this particular case coming to trial. Phaedrus, as the magistrate of the court, will have to decide whether or not the trial goes ahead. This legal image presents the situation where a subsidiary point in debate is shown to have an important bearing on the main issue. If the evidence of the logoi is accepted and it is decided that rhetoric is not a science, the original question will have to be modified before progress can continue.

²³ In a similar vein, at Tht. 202a certain demonstrative pronouns and accompanying adjectives are said to ‘run around’ everywhere as they attach themselves to different nouns: (‘We ought not even to add “just” or “it” or “each” or “alone” or “this”, or any other of a host of such terms. These terms, running loose about the place, are attached to everything, and they are distinct from the things to which they are applied’; tr. Cornford).

²⁴ See e.g. the use of ἔμπεδος in Homer, which denotes steadfastness in the battle-line and in personal conduct (Il. 5.254. 527; 16.520; 18.158; and Od. 11.152, 628; 17.464; 19.493; 21.426 and 22.226), and the positive association between the absence of movement and peace at Bacchylides 5.199–200 (τοὺς ὁ ). Steiner, Deborah (The crown of song (1986))Google Scholar discusses how Pindar uses symbols of motion to represent impermanence and instability in human life (ch. 6 ‘Winds and waves’. 66–70) and how storm imagery is contrasted with that of good weather. Further, she explains how at Pyth. 5.10–11 and Isth. 7.23 Pindar develops the contrast between motion and fixity through supporting imagery of light and darkness: ‘Pindar thus turns his metaphors towards his encomiastic ends, celebrating the man who may transcend the motion-filled condition in which he lives, and achieve one fixed moment of calm and sunshine’ (72).

²⁵ Tr. Campbell, D.A., Greek lyric I, Sappho and Alcaeus (1990)Google Scholar.

²⁶ King, H., ‘Bound to bleed: Artemis and Greek women’ in Cameron, A. and Kuhrt, A. (edd.). Images of women in antiquity (1983) 109–41Google Scholar.

²⁷ Dean-Jones, Lesley in Women's bodies in classical Greek science (1994)Google Scholar discusses the Hippocratic view that the womb could relocate to other parts of a woman's body (‘The “wandering womb”’ 69–77) and sets out the various verbs used for the womb's movement (e.g. , σεύομαι). She comments on Plato's use of the idea of the mobile womb (Tim. 91b–d) and makes the important observation (70): ‘The Hippocratics never describe the womb explicitly as an individual animal wandering at will within the body of a woman, in fact the gynaecologists never use the verb “to wander” (πλάνω) to describe the womb's movements’. Why Plato should have added this term to the vocabulary of the womb's movements is not considered by Dean-Jones, but is important for my discussion on wandering in Plato (below). Plato introduces the idea of ‘wandering’ to bring the concept of the mobile womb into line with his distinction between orderly and disorderly motion whereby ‘wandering’ becomes a sign of irrationality.

²⁸ For Alcmaeon's views, as reported in Theophrastus' Fragment on sensation, ch. 26, see Longrigg, James, Greek rational medicine: philosophy and medicine from Alcmaeon to the Alexandrians (1993) 47–81, esp. 58CrossRefGoogle Scholar.

²⁹ Warner, R., Thucydides, the Peloponnesian war (1954)Google Scholar.

³⁰ Padel, Ruth, In and out of the mind (1992)CrossRefGoogle Scholar.

³¹ While Padel allows that ‘movement is also the source of being moved, of interest, intensity, and excitement’ (In and out 68), her emphasis remains on movement as destructive and painful.

³² See Longrigg, Greek rational medicine, e.g. 37 (air), 62 (blood), 68 (pneuma), 79–80 (semen). In view of these necessary movements, Padel's claim that ‘Greeks do not have our sense that perpetual movement within is normal’ (In and out 67) would seem to be too sweeping.

³³ See Longrigg, , Greek rational medicine 42–3Google Scholar; King, ‘Bound to bleed’, and Jones, Dean, Women's bodies 50–1 and 123–35Google Scholar.

³⁴ On the image of the chariot of the Muses and of the poem as a journey see Harriott, R., Poetry and criticism before Plato (1969) 63–8Google Scholar, and Steiner, Crown, ch. 7 ‘Pindar's paths’ (76–86).

³⁵ Pindar, Nem. 5.1–3 , … (‘I am not a sculptor, so as to fashion stationary | statues that stand on their same base. | Rather on board every ship | and in every boat, sweet song, | go forth from Aigina and spread the news …’ (tr. Race). Steiner notes the positive aspects of this mobility (66–7): ‘the ode's power to travel to the far corners of the earth, through space and time, and so to spread the glory of its patron’. For further comments on how Pindar attributes the power of movement to his poetry and how this is a positive feature, see Steiner, 39, 75 and 85.

³⁶ See Padel, In and out (31,83, 121, 176), and Whom gods destroy: elements of Greek and tragic madness (1995) ch.10 ‘Madness as “wandering”’ and ch. 11 ‘Resonances of wandering’.

³⁷ In and out 121. For the wandering motif, see also PV 572, 576. 585 etc.; Or. 466; Aj. 886 and OT 3, 122–3, etc.

³⁸ Dowden, Ken, Death and the maiden (1989)Google Scholar.

³⁹ See Padel, , Whom gods 108–9Google Scholar.

⁴⁰ The external movement of wandering is closely linked with inner movement in these lines, as Agamemnon goes on to say that his heart is not ἔμπεδον (94) (see note 24 above). For a negative view of wandering in general, see Od. 15.343 where Odysseus informs Eumaeus (‘than wandering nothing is more evil for mortals’). See also Archilochus 130 (West) lines 4–5 (‘then comes much trouble, and a man wanders in need of food and distraught in mind’; tr. Edmonds, J.M. (adapted), Elegy and iambus II (1931))Google Scholar.

⁴¹ On storm imagery in Greek poetry in general see Padel, In and out 81–8 (‘Flow and storm’), and for Pindar's use of storm imagery see Steiner, , Crown 66–72Google Scholar.

⁴² The verb ϰυλίνδεται (‘rolls’) will be examined in the Platonic context (below).

⁴³ Il. 11.307; 14.18; Od. 5.296. When waves are described as rolling in Homer, the usual context is that of a storm at sea, where the language of rolling conveys the destructive force of the waves. On one occasion only (Od. 9.147) rolling is used as a neutral description of the movement of large waves at sea.

⁴⁴ See also Il. 17.99, 17.688 and Od. 8.81. Hainsworth, J.B. (A commentary on Homer's Odyssey I (1988))Google Scholar, identifies the use of ϰυλίνδετο at Od. 8.81 as a ‘traditional metaphor’ (352).

⁴⁵ See also Il. 22.414, 24.165 and Nicholas Richardson's comments on this act of mourning (The Iliad: a commentary V (1991) 150)Google Scholar.

⁴⁶ Fagles, R., Homer. The Iliad (1991)Google Scholar.

⁴⁷ At Il. 13.578–9 Helenus strikes Deipyrus and shears off his helmet; the helmet falls to the ground and one of the soldiers picks it up as it rolls among the feet of the fighters (, ). In the space of one line Deipyrus is dead. Similarly at 16.794–6, when Apollo knocks off Patroclus' helmet (which he has borrowed from Achilles), attention is drawn to the fate of the helmet itself: . Since Patroclus will soon be dead, the helmet rolling on the ground prefigures the fate of the man. Janko, R. (The Iliad: a commentary IV (1992) 411)Google Scholar, observes on Il. 16.794: ‘The poet dwells on this moving detail as if it stands for the hero's own head lying in the dust.’ Cf. 112 on Il. 13.526–30: ‘falling objects rolling noisily on the battlefield are a topos’.

⁴⁸ This association of rolling with defeat and dishonour appears again at Il. 14.411, where Aias strikes Hector on the chest with a boulder, making him spin like a top and then fall to the ground. The boulder is one of many that had been used to prop up the ships and ‘had rolled among the feet of the combatants’ (παϱ’ ). The stone is thus ordinary and common, a lowly object which prefigures the lowly state of the defeated Hector.

⁴⁹ See Janko, R., Il. comm. IV 62Google Scholar, on the boulder comparison and on the force behind Hector's motion.

⁵⁰ Od. 5.296 describes waves rolling in a stormy sea and the related image of trouble rolling upon people like a great wave is used at 2.163 and 8.81. ϰυλίνδομαι denotes writhing around in grief at 4.541 (repeated at 10.499): , and the verb is used to convey Sisyphus' loss of control as he seeks to complete his hopeless task (11.598, ).

⁵¹ The intensive verb occurs only twice in Homer, in the passage quoted and at Il. 22.221, where Athene tells Achilles it would now be futile for Apollo to grovel before Zeus on behalf of Hector. On Il. 22.221 Richardson, (Il. comm. V 131)Google Scholar comments: ‘This is remarkably contemptuous towards Apollo, especially means ‘grovelling in front of as a suppliant, like Priam when supplicating the Trojans, , at 414a; … This vivid compound recurs at Od. 17.525, and nowhere else in later literature.’ Russo, Joseph (A commentary on Homer's Odyssey III (1992) 44)Google Scholar explains the verb's sound effect: onomatopoeically recreates the forward rolling it signifies, as it takes us all the way to the penthemimeral caesura in completely dactylic movement. It occurs only here and at Il. 22.221, and nowhere else in Greek literature.’

⁵² Lattimore, R., The Odyssey of Homer (1965)Google Scholar.

⁵³ See also Hipp. Min. 372e; Hipp. Maj. 304c; Prot. 356d; Alc. 117a–118b; Phdr. 263b; Soph. 230b; Phdo. 81d; and Rep. 479d.

⁵⁴ Cornford, F.M., Plato's cosmology (1937) 165Google Scholar.

⁵⁵ E.g. Rep. 472a, 457b, 457c, 473c.

⁵⁶ E.g. Euthyd. 277d; 293a; Lach. 194c; Tht. 177b, 184a; Phil. 14a. Compare the image of argument as a strong river at Laws 892c–93b.

⁵⁷ See e.g. Phdo. 78d () and Symp. 211b ().

⁵⁸ Compare Gorg. 527b where a proven argument ‘stands steadfast’ ( and 508e–509a where established points are described as ‘held fast and bound by arguments of steel and adamant’ ().

⁵⁹ The text is disputed. F.M. Cornford (Plato's cosmology) follows Burnet's text (ἀνιϰήτοις) and concludes (ad loc): ‘The uncertainty of the reading does not affect the sense.’ This may be true, but it does affect the image. Taylor, A.E. (A commentary on Plato's Timaeus (1928))Google Scholar also favours ἀνιϰήτοις: ‘The ἀνιϰήτοις of A in b8 is shown to be more probable than the ἀνιϰήτοις of F by the ancient versions. The inexpugnabilis of Chalcidius might represent either word … but Cicero's neque convinci potest is unambiguous. Here again we may suspect that the common archetype of A and F probably had a marginal variant.’ Archer-Hind, R.D. (The Timaeus of Plato (1888))Google Scholar opts for ἀνιϰήτοις and translates: ‘so far as it lies in words to be incontrovertible and immovable’. Because of the image established by μονίμους, I also favour ἀνιϰήτοις. In support of this reading, one may observe the same play on ἀνίϰητος/ἀϰίνητος in Bacchylides 5.57 and 200, while against the reading one may note the double pairings in this Tim. passage: μονίμους with ἀμεταπτώτους (abiding/unchangeable) and ἀνελέγϰτοις with ἀνιϰήτοις (incontrovertible/irrefutable).

⁶⁰ Cornford, Plato's cosmology.

⁶¹ In contrast, no account of the changing, visible world can ever be ‘a final statement of exact truth’ (Cornford, , Plato's cosmology 23–4)Google Scholar.

⁶² Osborne, Catherine, ‘Space, time, shape, and direction: creative discourse in the Timaeus’. 179–211 in Gill, C. and McCabe, M.M. (edd.), Form and argument in late Plato (1996)Google Scholar.

⁶³ Baxter, T., The Cratylus: Plato's critique of naming (1992)Google Scholar.

⁶⁴ She reiterates the point in her later section on ‘Time’ (196): ‘These problems with our ordinary language indicate that normal discourse is not well suited to dealing with what is timeless; and one possible explanation of this is that discourse, like the world, is inherently temporal’.

⁶⁵ Taylor, Commentary; Cornford, Plato's cosmology; Zeyl, Donald J., ‘Plato and talk of a world in flux: Timaeus 49a6–50b5’, Harvard Studies in Classical Philology 19 (1975) 125–48CrossRefGoogle Scholar.

⁶⁶ For soul as the cause of motion see Phaedrus 245c–d.

⁶⁷ For fathers and sons see 275e4 and 278a6; for brothers see 276a1 and 278b1; and for seeds and plants see 276c8 and 276d–277a.

⁶⁸ Mackenzie, M.M., ‘Paradox in Plato's Phaedrus', Proceedings of the Cambridge Philological Society n.s. 28 (1982) 64–76. See 73 n. 14CrossRefGoogle Scholar.

⁶⁹ The verb appears on only two other occasions in the corpus: Rep. 388b6 and Tim. 59d5. Rep. 388b6 is a quotation of Il. 22.414–15 where ‘rolling’ is an expression of Priam's grief. Therefore, the only passage in Plato where ϰυλίνδομαι does not carry an epistemological reference is Tim. 59d5 where it is used in the context of an etymology for ὕδωϱ – the name ‘liquid’ derives from its motion and its rolling course over the ground.

⁷⁰ This passage will be discussed below, in conjunction with Phdr. 257a.

⁷¹ Without the services of the human body, the head would roll about on the ground in the same way as the lowly stones and the dishonoured heroes in Homer (see above).

⁷² Lebeck, Anne (‘The central myth of Plato's Phaedrus’, Greek, Roman and Byzantine Studies 13 (1972) 267–90)Google Scholar notes the use of ϰυλινδεῖται at both Phdr. 257a 1–2 and 275el, but translates the verb as ‘flit’, which obscures its connotations and significance in the passages (286–7).

⁷³ On ‘rolling’ in Pindar see Bulman, P., Phthonos in Pindar (1992)Google Scholar, Referring to the usages at Nem. 4.40 and Pyth. 2.23, she observes: ‘For Pindar, as for Plato, ϰυλινδεῖν is pejorative, designating the aimlessness and uselessness of objects rolling around in space and time’ (85). This tantalizing note unfortunately adduces none of the Platonic evidence. Verdenius, W.J. (Commentaries on Pindar I, Mnemosyne Suppl. 97 (1987) 95)Google Scholar comments on rolling as an unsteady movement in this passage and at Pyth. 4. 209.

⁷⁴ Race, , Pindar I (1997)Google Scholar. For Pindar's views on ἐλπίς see Verdenius, , Commentaries on Pindar 95Google Scholar. On the theme of human stability and mutability see Fränkel, H., ‘Man's “ephemeras” nature according to Pindar and others’, Transactions of the American Philological Association 77 (1946) 131–45CrossRefGoogle Scholar. On the phrase ἄνω … ϰάτω here see Verdenius, , Commentaries on Pindar 93–4Google Scholar, who argues that the common translation ‘up and down’ is incorrect and that ‘to and fro’ is preferable, since the movement indicated is horizontal, not vertical. Verdenius cites Plato's usage of the phrase at Phil. 43a3 in support of his argument. Verdenius has also discussed the phrase in his earlier note ‘ΑΝΩ ΚΑΙ ΚΑΤΩ’, Mnemosyne 17 (1964) 387Google Scholar. where he rightly concludes that the meaning of the phrase at Rep. 344b, as at 508d and Phdo. 96b. has to be ‘to and fro’ or ‘moving about’. Plato uses the phrase to convey disorderly motion, especially as a metaphor in the context of changing opinion, see Prot. 356d; Gorg. 481 e: Hipp. Min. 376c and Tht. 195c. (For instances where ‘up’ and ‘down’ are used by Plato as a vertical metaphor for progressions in thought see Robinson, R., ‘Up and down in Plato's logic’, American Journal of Philology 84 (1963) 300–3CrossRefGoogle Scholar.)

⁷⁵ The ideas of wandering and rolling are fused together, as noted above, in the use of the verb at Od. 17.525 and are also brought into close conjunction at Il. 11.307–8 in a description of the wild motion of wind and waves (). Plato directly associates wandering with rolling on two memorable occasions: first, at Phdo. 81d, where the lost unphilosophical soul is ‘rolled around’ graveyards (ϰυλινδουμένη) and where such souls are compelled to ‘wander’ about these places until they have ‘paid their penalty’ (); second at Rep. 479d, where Socrates sets out the three levels of existence and knowledge and locates the objects of the masses' opinions in the mid-region between non-existence and full reality: . (‘We would seem to have found, then, that the many conventions of the many about the fair and honourable and other things are tumbled [rolled] about in the mid-region between that which is not and that which is in the true and absolute sense’; tr. Shorey). In the subsequent paragraph Socrates refers to the object caught by the faculty of opinion as ‘the wanderer’ (τὸ πλανητόν), thus making the objects of opinion both roll and wander in this mid-region.

⁷⁶ Sommerstein, A.H., Aristophanes, Wasps (1983)Google Scholar.

⁷⁷ Rowe, C.J., Plato: Statesman (1995)Google Scholar.

⁷⁸ Waterfield's, R. translation (Plato, Republic (1994)Google Scholar) of the phrase as ‘it's been curled up at our feet’ entirely misses the force of the verb, since ‘curled up’ not only denotes a lack of movement but also suggests a fairly comfortable state. Although he later (479d) translates the verb more appropriately (‘mill around’), the lack of consistency obscures the connection. In contrast, Shorey's consistent translation of ϰυλίνδομαι allows Plato's sly point to emerge.

⁷⁹ Mackenzie, , ‘Paradox’ 69–70Google Scholar.

⁸⁰ Mackenzie, , ‘Paradox’ 72Google Scholar.

⁸¹ History of Greek philosophy I (1962) 351–7Google Scholar, on circularity in Greek thought and Plato's debt to Alcmaeon (on which see also Mugler, Ch., ‘Alcméon et les cycles physiologiques de Platon’, Revue des études grecques 71 (1958) 42–50CrossRefGoogle Scholar). For poetic expression of the theme of circularity see Sophocles, . Aj. 670–3Google Scholar and Trach. 129–30.

⁸² Skemp, J.B., The theory of motion in Plato's later dialogues (1942; enlarged ed. 1967)Google Scholar.

⁸³ In ‘Straight and circular in Parmenides and the Timaeus’, Phronesis 19 (1974) 189–209CrossRefGoogle Scholar, Lynne Ballew has shown that this opposition in the context of knowledge and being goes back to Parmenides.

⁸⁴ On the implications of sphericity see Mortley, R.J., ‘Plato's choice of the sphere’, Revue des études grecques 82 (1969) 342–5CrossRefGoogle Scholar.

⁸⁵ Note also the storm and wave imagery in this passage: .

⁸⁶ Compare the illustration of intelligence at Laws 898a–b. Edward N. Lee has offered a detailed analysis of Plato's use of the idea of circular thought in ‘Reason and rotation: circular movement as the model of mind (nous) in later Plato’ (in Werkmeister, W.H. (ed.), Facets of Plato's philosophy, Phronesis supplementary volume 11 (1976))Google Scholar. While his review of the relevant passages in Timaeus and Laws is useful (esp. section two on the nature of axial rotation), his subsequent exegesis is obscure. For example, Lee explains the notion of circular thought (which he takes to be metaphorical. 88) by means of the idea of perspective and vision (e.g. 81: ‘it is this “perspectiveless” character of its all-pervading orientation towards the center that seems to me the decisive feature of Plato's rotation analogy’). But Lee's vision metaphor seems only to complicate matters further. When his account concludes with a distinction between two different perspectives (the ‘threshold’ and the ‘dweller's’) and two types of mysticism (‘quietist or static’ and ‘a more activistic mysticism’, 92), Plato's texts seem to have been left behind.

⁸⁷ Sedley, David, ‘“Becoming like god” in the Timaeus and Aristotle’ in Calvo, T. and Brisson, L. (edd.). Interpreting the Timaeus–Critias, International Plato Studies 9 (1997) 327–39Google Scholar.

⁸⁸ Skemp, J.B., ‘Plants in Plato's Timaeus’, Classical Quarterly 41 (1947) 53–60CrossRefGoogle Scholar.

⁸⁹ Saunders, T.J., Plato, The laws (1970)Google Scholar.

⁹⁰ J.B. Skemp's translation of this key section (893c–e) is as follows (Theory of motion 97–8): ‘“Is it of moving things which have the power of being stationary at their centre that you speak”, we shall reply, “when you say that they move in one place, like the revolution of so-called ‘humming tops’?” “Yes.” … “Moreover, when you speak of things moved in a succession of places, you appear to be describing objects moving on continually from place to place, in some cases pivoting themselves at one point gliding along, in other cases rolling with changing pivotal points.”’ Thus Skemp terms the two types of motion in several locations as (a) ‘gliding’ (with one point of contact) and (b) ‘rolling’ (with several points of contact). On the idea of axial rotation at 893c4–7, Skemp comments (100): ‘The reason for the pride of place accorded to this kind of motion and the detailed exposition of it is, of course, because it is the motion of the οὐϱανός itself, perfectly combining motion and rest.’ Skemp goes on to observe that ‘We have here in a summary form the language of Tim. 39a and of Laws 7.822a, where the reference is to the double-motion astronomy’ (101).

⁹¹ On 893d6 and the ‘passing from rotation to rectilinear motion’ Skemp comments (Theory of motion 101): ‘Why Plato distinguishes gliding and rolling is not clear. The heavenly bodies rotate and glide but do not roll, for their axis is fixed. There may be a reference to lost theories of planetary motions or Plato may simply wish to give an exhaustive classification. This does not mean that rolling is not a “physical” motion: it is a physical possibility, even if not actualised in the planets.’ On my analysis of ϰυλίνδομαι Plato uses πεϱιϰυλινδεῖσθαι to mark the transition from orderly to disorderly motion.

⁹² Remember that in the Republic Plato's way of demonstrating that something can be in motion and at rest at the same time is the illustration of the spinning tops, which are still in respect of their axes but move in respect of their circumferences (436d).

⁹³ For the same image, see also Tht. 200c (); Cleit. 410a () and Phil. 19a (). See also Euthyd. 291 b; Euthyph. 15b; Polit. 286e; Laws 659d and 688b. Circular motion is also used to express confusion in Plato's images of intellectual dizziness and whirlpools; see Phdo. 79c; Crat. 411b, 439c; Prot. 339e; Lys. 216c; Tht. 155c and Laws 663b, 892e.

⁹⁴ Irwin, T., Plato, Gorgias (1979)Google Scholar.

⁹⁵ ‘The role of the interlocutor in Plato – theory and practice’ in Pelling, C.B.R. (ed.), Characterization and individuality in Greek literature (1990) 174–96Google Scholar.

⁹⁶ The striking phrase ‘’ occurs remarkably few times in the Platonic corpus; apart from the three occurrences in the Gorg. (490e11, 491b8, 527d7) and this echo at Tim. 40a8, the only other use of the phrase is at Minos 316d1, where again the theme is of intellectual consistency ( δοϰοῦσι). Other passages where τεϱὶ τῶν αὐτῶν appears in the context of intellectual consistency are: Soph. 230b8; Meno 96a3 and Rep. 603d2. Although all these passages deal with the theme of consistency in thought, the only ones that share the phrase ‘the same … about the same’ are the three in the Gorg. and that in the Tim. Further, since Gorg. 491b8 and 527d7 present the negative formulation, ‘never the same about the same’, only two passages share the positive formulation, ‘always the same about the same’, i.e. Gorg. 490e11 ()and Tim. 40a8 ().

⁹⁷ Cornford, (Plato's cosmology 118–19Google Scholar) explains how it is that the stars move ‘forward’ while keeping their relative positions, and comments on the notion of axial rotation: ‘Every star has also, we are now told, a second motion, rotation on its own axis. The reason is that “each always thinks the same thoughts about the same things”. Here, for the first time in the Timaeus, it is explained why axial rotation is regarded as “that one of the seven motions which above all belongs to reason and intelligence” (34a).’ For Alcmaeon's influence on the formulation of this passage see Guthrie, , History of Greek philosophy I 356Google Scholar.

⁹⁸ In a similar fashion Anne Lebeck (‘Central myth’) observes Plato's idea of moving discourse at work in the Phdr. and further notes the contrast between ‘ordered’ and ‘disordered’ motion in Plato (284): ‘Important for this dialogue and for Plato's philosophy as a whole is the image of motion and a way by which to go. On the one hand, movement symbolizes impermanence, flux, deranged perception. On the other, it is creation, life, soul. And intelligence is understanding things in their motion … The first is disordered movement, aimless wandering, zig-zag. The second is ordered motion, turning in place, ascent … Every dialogue, by virtue of form alone, is a symbol of such movement and the search for a way.’

⁹⁹ Discourse attains stability through its relationship with the wholly immovable Forms, and thus whatever fixity discourse has is dependent on, and necessarily inferior to, the perfect stability of Being. In terms of value, then, even the best discourse is inferior to the Forms to which it relates. Osborne (‘Creative discourse’ 187–9) discusses the hierarchy of discourse in Plato and comments on the different truth-status of discourse and the Forms (187): ‘Discourse itself seems to be a candidate for reliability but not for truth, which is a feature attributable to the permanent entities to which reliable discourse relates, or, by extension perhaps, to the cognitive state associated with them.’