Skip to main content

Cleombrotus of Ambracia: interpretations of a suicide from Callimachus to Agathias*

  • G. D. Williams (a1)

At Phaedo 59b Echecrates asks Phaedo who was present on the day when Socrates drank the hemlock in prison. Various Athenians are named (59b 6–10), then various foreigners (59c 1–2), but when Echecrates subsequently asks if two other foreigners, Aristippus and Cleombrotus, were present, Phaedo replies that they were said to be in Aegina (59c 4). After this fleeting reference to Cleombrotus, Plato does not mention him again in the Phaedo or any other dialogue; and yet in later antiquity a certain Cleombrotus of Ambracia rose to fame in connection with the Phaedo. Callimachus is our earliest source for the anecdote which immortalized the Ambracian (A.P. 7.471):

Εἴπας ‘Ἥλιε ϰαῖρε’ Κλεμβρτος ὡμβρακιώτης

ἅλατ' φ' ὑΨηλο τεϰεος εἰς Ἀδην,

ἴζιον οὐδν ἰδὼν θαντου κακν, λλ Πλτωνος

ἕν τ περ Ψυϰς γρμμ' ναλεζμενος.

Hide All

1 That Aristippus and Cleombrotus are ζοι is not actually stated in the text; but it is plausibly inferred from the initial reference to ζοι at 59b 11.

2 Epigr. 23 Pf.; HE 1273–6 (= Callimachus LIII); Page, OCT 1378–81. Περ Ψυϰς seems to have been an alternative title for the Phaedo (see Gow and Page, 2 p. 204); Plato may have supplied it if Ep. 13 is genuine (ν τῷ περ Ψυϰς λγῳ, 363a). Cicero knew the work as eum librum qui est de animo (Tusc. 1.24).

3 For the anti-Platonic line see especially Sinko, passim with Spina, pp. 21–2 and Riginos A. S., Platonica: The Anecdotes Concerning the Life and Writings of Plato (Leiden, 1976), p. 181 (‘Callimachus here parodies the doctrine of the Phaedo’). Wilamowitz , Hellenistische Dichtung (Berlin, 1924), 1 p. 177 finds philosophical theory being derided in the epigram, but cf. his Platon (Berlin, 1920), 2 p. 57, where Cleombrotus is said to be ridiculed, Plato praised.

4 The standard response in modern scholarship is neither to affirm nor deny the identification. For suspension of judgement see on Phd. 59c the editions of D. Wyttenbach (London, 1810), p. 118, I. Bekker (London, 1826), 5 p. 131, G. Stallbaum (Gotha, 1850), p. 15, W. D. Geddes (London and Edinburgh. 1863), p. 181; also Burnet, p. 10, Hackforth, pp. 30–31, and Gow and Page, 2 p. 204. C. J. Rowe, in his new edition (Cambridge, 1993), is at first similarly tentative about the identification (see p. 116 on 59c 3) but then (on c 4) accepts as probable that the epigram is about the Platonic Cleombrotus.

5 Cf. D. L. 2.65 and 3.36, reporting that Xenophon and Plato were hostile towards Aristippus. Hence the slight in the Phaedo; but for justified scepticism about the rumour of hostility see Hackforth, p. 31.

6 On Aristippus, his hedonistic views and possible (but controversial) role in the founding of the Cyrenaic school, see Guthrie W. K. C., Socrates (Cambridge, 1970), pp. 170ff.

7 A factual point faithfully recorded? And/or one which sets up an implicit contrast between Plato's honourable absence and the less forgivable absence of Aristippus and Cleombrotus? Cf. Guthrie, op. cit., p. 169 n. 2: ‘His [Plato's] feelings may have been so intense that he could not bear the prospect of witnessing the actual death of “the best, wisest and most righteous man he ever knew”. He could have made his farewell earlier to an understanding Socrates. But since we have absolutely nothing to go on but his own four words, all such guesses at motivation are idle’. See now Most G. W., ‘A cock for Asclepius’, CQ 43 (1993), 96111, ingeniously arguing that with Socrates' last words (ὦ Κρτων, ἔφη, τῷ Ἀσκληπιῷ φελομεν λεκτρυνα, Phd. 118a 7–8) Plato has the latter thank Asclepius for Plato's anticipated recovery from illness; Plato is more present in the dialogue than his absence suggests.

8 At Phd. 58alO–c5 the origins and procedure of the Delian θεωρα are described by Phaedo in order to explain the time-lapse between Socrates' trial and death.

9 Pp. 10–11 on Phd. 59c 4.

10 P. 11.

11 Cf. Ath. 12.544d (III p. 200 Kaibel), where Aristippus is said to have wasted time living it up in Aegina (διτριβεν … τ πλλ ν Αἰγῃ τρυφφν), Athenaeus may be alluding to (the Demetrian interpretation of?) Phd. 59c.

12 Gow and Page, 2 p. 104 refer to Wilamowitz (Platon, 2 p. 57) for the theory, adding Geffken J., Griechische Literaturgeschichte (Heidelburg, 1934), 2 p. 92 (Anm. 171). But Geddes , op. cit. (n. 4), p. 181 attributes it to C. M. Wieland, who nevertheless absolves Cleombrotus from the charge of negligent absence; in a letter to Aristippus (Aristipp II 10 = Christoph Martin Wieland Werke, Bd. 4 [ed. K. Manger, Frankfurt am Main, 1988], pp. 268–9 and 446–7 n. 2), Wieland has the Ambracian explain his absence on the grounds that he could not bear to witness his beloved master's death.

13 Various Cleombroti are listed at RE XI 677–9; Fraser P. M. and Matthews E., A Lexicon of Greek Personal Names I (Oxford, 1987), pp. 261–2 also list fourteen Aegean Cleombroti from the fifth/fourth century onwards.

14 The epithet Ἀμβρακιώτης could conceivably imply that Callimachus follows an early tradition apart from Plato, though other possibilities remain. Perhaps Callimachus invents Cleombrotus’ Ambracian origins; if so, does he develop the Platonic hint that (the other) Cleombrotus was a ζνος?

15 For Ammonius see CAG–25 Busse; Elias , CAG; David , CAG; pseudo-Elias , In Porph. Isag. 12.4–5, 38–9 = pp. 14, 18 Westerink.

16 For background on these various figures see Burnet, pp. 7–10 on Phd. 59a 9ff. with Hackforth, pp. 30–31 and Field G. C., Plato and his Contemporaries (London, 1930), pp. 158–80.

17 φρουρ (62b 4) allows both meanings, ‘prison’ and ‘garrison’. ‘Prison’ fits the recurring notion in the Phaedo of the soul incarcerated in the body (e.g. 67d 1–2, 82e–83a). But for ‘garrison’ see Cic.Sen. 74 (‘vetatque Pythagoras … de praesidio et statione decedere’) with J. G. F. Powell's commentary (Cambridge, 1988), pp. 247–8 and 248 n. 1; Cicero seems to have been aware of the ambiguity in φρουρ because elsewhere he interprets the word as ‘prison’ (cf. Tusc. 1.74, Rep. 6.15 Ziegler).

18 With ν πορρτοις (62b 3) Phaedo alludes to the Orphic origins of the doctrine of the soul's immortality (see Burnet, p. 22 ad loc.). Strachan J. C. G., ‘Who did forbid suicide at Phaedo 62b?’, CQ 18 (1970), 220 argues that Plato ‘need not be taken to refer to any Pythagorean belief-theory’ here; but for fusion of Orphic/Pythagorean influences see Burnet, pp. 23–4, with Hackforth, p. 38.

19 Gallop's translation, p. 6; see also pp. 83–4 for objections to this first argument because (i) to condemn suicide as illegal is to ignore the moral issues suicide raises (‘if suicide is sinful or morally wrong, presumably it is so whether it is legal or not’), and (ii) to condemn suicide as cowardly is to invoke a term of moral reproach (‘cowardice’) which does not explain the moral status of suicide per se.

20 Cf. Gallop, p. 85: ‘Socrates is not maintaining an absolute veto upon suicide. On the contrary, with the words “until Go d sends some necessity, such as the one now before us” ([62]c 7–8), he implies that his own death will be self-inflicted. In his case, self-destruction would be not merely permissible, but a religious duty’.

21 But cf. Gallop, pp. 84–5, arguing that if everything is ordained by cosmic intelligence (cf. 97c–d), anyone who commits suicide does so according to divine will; νγκη cannot be adduced only in special cases. If this was a fully persuasive argument and Cleombrotus had seen it for himself, his suicide might have been less indefensible in Platonic terms.

22 Cf. Laws 873c–d, where Socrates discusses the punishment to be imposed on the corpse of a man who takes his life in defiance of ‘the decree of destiny’ (τν τς εἱμαρμνης … μοῖραν, c 4). Unlawful suicide is in fact distinguished from three categories of voluntary death which are apparently permissible: suicide on the order of the state (as in Socrates' case), suicide to escape intolerable pain or oppression, and suicide to avoid disgrace. On the assumption that Cleombrotus did not kill himself to escape the disgrace of Plato's alleged rebuke in the Phaedo, only the second of these three justifications for suicide could apply to him, and only then if the definition of what constitutes intolerable pain is stretched: Cleombrot' only justification could be that his soul's bodily imprisonment was a form of oppression too painful to be endured. Otherwise, and damningly, his death could only be viewed as a cowardly and illegal evasion of one of the ordinary hardships of life—having to live with the soul imprisoned in the mortal body. Given this parallel for Platonic restrictions on suicide, does Callimachus hint at the dangers implicit in Cleombrotus' limited reading when he writes ἕν … γρμμ' ναλεζμενος (4)? Cf. van Hoof A. J. L., From Autothanasia to Suicide: Self-Killing in Classical Antiquity (London, 1990), pp. 76–7: ‘he [Cleombrotus] jumped from a wall after reading the book of Plato. Had he ceased to be a “vir unius libri” by reading, apart from the Phaedo, Plato's Laws, he could have reached different conclusions’.

23 See my opening bibliography for the reference.

24 P. 3.

25 For (rare) examples of λλ in this sense see LSJ s.v. I 3.

26 Epigr. 10 Pf.; HE 1199–1202 (= Callimachus XXXIII); Page, OCT 1304–7.

27 The paraphrase is from Gow and Page, 2 p. 190.

28 Op. cit., p. 8. Sinko assumes that the words πλι ἔσεαι refer to reincarnation, but in A. W. Mair's Loeb translation (London, 1927) they are rendered differently (‘how it shall be …hereafter’); Gow and Page, 2 p. 191 ad loc. favour πλι in the Pythagorean sense. I take ‘again’ to be the more natural rendering, with ‘hereafter’ an extended sense of the term; cf. πλιν associated with rebirth and reincarnation in παλιγγενεσα See LSJ s.v. 1, 2).

29 For these three Timarchi see Gow and Page, 2 p. 190.

30 Epigr. 13 Pf.; HE 1187–92 (= Callimachus XXXI); Page, OCT 1292–7.

31 Nothing is known of this Charidas, but Gow and Page, 2 p. 188, do not doubt that he existed.

32 I take ἄνοδοι (3) to refer to reincarnation; ‘fo r if Ψεδη and μθοι are to be distinguished [cf. 4], such ἄνοδοι as those of Persephone, Heracles, Sisyphus, Orpheus, or Adonis are, like Pluto, rather μθοι than Ψεδη (Gow and Page, 2 p. 189 on 3).

33 Cf. Gow and Page, ibid.: ‘the enquirer is more likely to represent any passer-by than Callimachus himself.

34 Cf. Gow and Page, ibid.: ‘The drift of the conversation is compatible with Charidas either confirming from the grave views which he had maintained in his lifetime, or admitting the truth of those held against him’.

35 Cf. Gow and Page, ibid.: ‘it may be that, whatever views Charidas had held, this epigram expresses C'.s’. Perhaps; but surely rash to assume so.

36 For evidence of Callimachus' knowledge of Plato see Iamb. 5.26ff. and 31 (= fr. 195 Pf.) with Pfeiffer ad loc. and Fraser P. M., Ptolemaic Alexandria (Oxford, 1972), I pp. 740–41; fr. 460 with Pfeiffer ad loc. on Plato's possible role in Callimachus' Πρς Πραζιπνην; fr. 589 with Pfeiffer , History of Classical Scholarship (Oxford, 1968), pp. 93–4. Cf. Fraser , op. cit., I p. 410 on the attested familiarity with and use of Plato in Callimachus' contemporary (and pupil) Eratosthenes.

37 Cleombrotus died ‘nullam aliam ob causam nisi quod Platoni credidit. Exsecrabilisprorsus ac fugienda doctrina, si abigit homines a vita’ (Inst. 3.18). On Lactantius' hostility to pagan Greek philosophy see Ellspermann G. L., The Attitude of the Early Christian Latin Writers toward Pagan Literature and Learning, The Catholic University of America Patristic Studies LXXXII (Washington, DC, 1949), pp. 79ff. Whether Lactantius knew the Callimachean epigram at first hand is doubtful, for his allusions to the Greek poets seem limited to Homer, Hesiod, Euripides and Musaeus; see Ogilvie R. M., The Library of Lactantius (Oxford, 1978), pp. 2027.

38 Jerome , Ep. ad Paulam 39.3 (= P.L. 22.468): ‘tales stulta philosophia habeat martyres, habeat Zenonem, Cleobrotum [sic] vel Catonem’ (Cato allegedly read the Phaedo before committing suicide after defeat at Thapsus in 46 B.C.; cf. Cic. Tusc. 1.74, Lact. Inst. 3.18). At Or. 4.70 (= P.G. 35.588) Gregory includes Cleombrotus in a list of pagan martyrs who are contrasted unfavourably with Christian counterparts (τ Κλεομβρτου πδημα το Ἀüμβρακιώτου, τῷ περ Ψυϰς λγῳ πιλοσοπηθν).

39 For discussion of Augustine's chapters on suicide as sin (CD. 1.17–27) see van der Horst P. W., ‘A pagan Platonist and a Christian Platonist on suicide’, Vigiliae Christianae 25 (1971), 282–8. Augustine's source at 1.22 may have been Cicero, for the ‘Theombrotus’ reported in certain of the Augustinian mss. is paralleled in the Ciceronian mss. at Tusc. 1.84 and Scaur. 4. On the improbable assumption that Cicero himself was in error, misquotation from memory offers one explanation, dependence on an unreliable text another; for discussion see Lundstrdm S., ‘Falsche Eigennamen in den Tuskulanen?’, Eranos 58 (1960), 6679, with Spina, pp. 22ff.

40 Fo r which see Conf. 7.13 (9), 7.26 (20), 8.3 (2). Augustine might also have seen the Christian relevance of Plato's argument that man, as a possession of the gods, is not free to dispose of his life as he pleases (Phd. 62c–d); see Bels J., ’La mort volontaire dans l'oeuvre de saint Augustin’, Revue de I'Histoire des Religions 187 (1975), 170. 4 1 CAG Busse.

42 Cf. Gow and Page, 2 p. 204: ‘Ammonius …calls Cleombrotus μειρκιον without warrant from Callimachus’. The innovation arises naturally out of Ammonius' interpretation of the epigram: Cleombrotus' folly is attributed to his immaturity. Despite the hint of boyishness in τῷ τῳ, μειρκιον could denote a student of 20 or so years (see LSJ s.v.). The term γκΨας certainly suggests a student busily poring over his books (cf. προσωμεν τονυν κα α∨γωμεν τ γραμματεῖα κα γκπτωμεν τοῖς γγεγραμμνοις κα ἴδωμεν …, J. Chrys. (Expos. in Psal. 5.1 [= P.G. 55.61). I take it that Bentley, in proposing γκροας stumbling upon’) for γκΨας (see Bromfield C. J., Callimachi quae supersunt … [London, 1815], p. 150), sought to stress Cleombrotus' amateurish approach to the Phaedo.

48 For Elias see CAG Busse; David, CAG Busse; pseudo-Elias, In Porph. Isag. 12.22–6 = pp. 16–17 Westerink, Cf. Olymp . In Ale. 5. Iff. Westerink (summarizing Damascius): through his suicide Cleombrotus rudely severs the bonds connecting the body and soul rather than loosening them, as the true philosopher should, δι το συμπαθος (i.e. the intermediate stage of μετριοπθεια which the philosopher goes through on his way to πθεια).

44 CAG–2 Busse (= App. Anth. 3.177, p. 320 Cougny); also in Elias (CAG–10), pseudo-Elia (In Porph. hag. 12.38–9 = p. 18 Westerink) and schol. to Dion. Thrax (Gram. Gr. l.[3].160.19–23 Hilgard = Anec. Gr. 725–6 Bekker). Perhaps, as Cougny suggests (op. cit., p. 375), Auypo's should be read in the second line to avoid two unconnected adjectives with δσμον Elias claims the couplet as his own, but Busse argues (CAG 18.1 p. vii) that Elias' lectures were set down by a pupil who was responsible for the careless and false attribution.

45 The difference between Olympiodorus and Cleombrotus lies ultimately in the difference between πδησεν and πδησεν a quaint coincidence, or is Olympiodorus making a playful point?

46 On the alleged (but disputed) suicides of the two ‘heroic’ Crassi distinguished at Scaur. 1–2 see Marshall B. A., ‘Some Crassi in Cicero's Pro Scauro’, Latomus 35 (1976), 92–6 with idem, A Historical Commentary on Asconius (Columbia, 1985), pp. 140–44. Qui tamen ipsi [Pythagoras et Plato] mortem ita laudant ut fugere vitam vetent atque id contra foedus fieri dicant legemque naturae. Aliam quidem causam mortis voluntariae nullam profecto iustam reperietis.

47 TRF 3 fr. xxxiv (in incertis) Ribbeck, who conjectures that the lines belong to Accius' Philocteta.

48 Cf. Aristoph. Eg. 83–4, Diod. 11.58.3, Plut. Them. 36.1.

49 See also An. 9.10.3 (Themistocles did better to commit suicide than make war on his homeland) with Bailey D. R. Shackleton, Epistulae ad Atticum (Cambridge, 1967), IV p. 378ad loc.

50 For examples of Graeculus used contemptuously see OLD s.v. la.

51 See T. W. Dougan's commentary (Cambridge, 1905), p. 106 ad loc. For the anecdote that Hegesias influenced his students to commit suicide see Plut. Mor. 497d, Val. Max. 8.9.3.ext., D.L. 2.86 (Hegesias was known as Πεισιθνανος) with Murray J. Clark, ‘An ancient pessimist’, The Philosophical Review 2 (1893), 2434; testimonia are assembled by Giannantoni G., Socratis et Socraticorum Reliquiae 1 (Naples, 1990), 2, p. 63 (fr.160) and pp. 113–15 (frr.1–7).

52 This is the suggestive line taken by White S. in an article entitled ‘Callimachus on Plato and Cleombrotus’ (TAPhA 124 [1994], 135–61). But cf. Griffin M., ‘Philosophy, Cato and Roman suicide’, G&R 33 (1986), 71: ‘Despite the predominantly negative view of Plato and his pupil…there is a kernel of truth in the story of Cleombrotus: that is the impression Plato's dialogue could make on the unwary reader, overwhelmed by its message of the immortality of the the soul and its incarceration in the body’. Could Callimachus be warning of the possible ill-effects of philosophy on impressionable readers/listeners? But even if so, can Plato be held responsible for Cleombrotus' decision to commit suicide? For the charge that Plato exerted an unhealthy influence by inducing many to study philosophy superficially, cf. Dicaearch. ap.

53 Sinko , op. cit., p. 9 is inclined to think so on the (questionable) assumption that the 'ιβις was one of the few books which Ovid took into exile (cf. Tr. 3.14.37–8).

54 Cf. (e.g.) Am. 2.19.35–6, where Ovid alludes to Epigr. 31.5–6 Pf.; and he clearly knew Epigr. 2 Pf. (see my Conversing after sunset: a Callimachean echo in Ovid's exile poetry’, CQ 41 [1991], 169–77).

55 So Gow and Page, 2 p. 204.

56 Sinko , op. cit., p. 2 and n. 1 reports scholarly claims that Cicero may have read εἰς ἅλαδε forv εἰς Ἀδην.

57 For this latter connotation see Virg.Aen. 6.543, 551, 577.

58 For Tartarus termed ‘scelerata sedes’ or the like see Cic. Clu. 71, Tib. 1.3.67–8, Virg.Aen. 6.563, Ov. M 4.456.

59 See Scaur. 5, Tusc. 1.74, Rep. 6.15 Ziegler, Sen. 73.

60 Given his malevolent intent in the Ibis, Ovid creates a suitably black irony by abandoning the Phaedo's Greek title and portraying the dialogue ‘on the soul’/‘on life’ (Περ Ψυϰς) as a dialogue ‘on death’ (‘de nece’, 494; but cf. Cicero's description of the Phaedo as ‘liber de morte’ [Scaur. 4]). Alternatively, ‘Socraticum’ could be construed as a transferred epithet (sc. ‘… ut qui Socratica de nece legit opus’).

61 For succinct commentary see Viansino G., Agazia Scolastico: Epigrammi (Milan, 1967), pp. 152–3 (= Epigr. 95).

62 The τρβων is the philosopher's common garb (see LSJ s.v. for examples); the diminutive τριβώνιον makes Nicostratus the most scantily clad of philosophers. For the beard as part of the philosopher's image see Viansino , op. cit., p. 153 on 11.

63 For σκινδαλαμς used of the hair/straw-splitting philosopher par excellence see Aristoph.Ra. 819 and M. 130 with K. J. Dover (Oxford, 1968), p. 110 ad loc.

64 Nicostratus invents στεγνοπυς for his erudite purpose; ἄϋλος is the rarer an d post-Aristotelian alternative to σώματος (cf. 5).

65 So Viansino , op. cit., p. 150.

66 See Cameron Averil, Agathias (Oxford, 1970), pp. 89111, countering previous claims that Agathias was a pagan.

67 Cited by Viansino, p. 153 on 16. Cf. Cameron , op. cit., p. 105, linking the epigram with Agathias' belief ‘that human knowledge can never reach the whole truth’.

68 For this dating see Reinach S., ‘La question du Philopatris’, RA 40 (1902), 79110 with R. Anastasis’ edition (Messina, 1968), pp. 7–34; but for an earlier post-classical date see Baldwin B., ‘The date and purpose of the Philopatris’, YCS 27 (1982), 321–44 with stylistic analysis of the piece.

69 The anonymous referee suggests that the joke in pseudo-Lucian may lie not in Critias' alluding to Cleombrotus' proverbial reputation as a madman, but in his deriding something ‘unquestionably true, beautiful, good’. An intriguing possibility; but does it not go against the weight of classical opinion on Cleombrotus, at least as I have presented it?

* I am grateful to Alan Cameron and Stephen White for commenting on an earlier draft of this paper, and to Byron Harries for invaluable assistance at every stage. My thanks also to an anonymous referee and to the editors for various improvements. The following works are cited by author's name only: R. S. Bluck, Plato's Phaedo (London, 1955); J. Burnet, Plato's Phaedo (Oxford, 1911); D. Gallop, Plato: Phaedo (Oxford, 1975); A. S. F. Gow and D. L. Page, The Greek Anthology: Hellenistic Epigrams (Cambridge, 1965); R. Hackforth, Plato's Phaedo (Cambridge, 1955); T. Sinko, ‘De Callimachi epigr. XXIII. W, Eos 11 (1905), 1–10; L. Spina, ‘Cleombroto, la fortuna di un suicidio (Callimaco, ep. 23)’, Vichiana 18 (1989), 12–39.

Recommend this journal

Email your librarian or administrator to recommend adding this journal to your organisation's collection.

The Classical Quarterly
  • ISSN: 0009-8388
  • EISSN: 1471-6844
  • URL: /core/journals/classical-quarterly
Please enter your name
Please enter a valid email address
Who would you like to send this to? *


Full text views

Total number of HTML views: 0
Total number of PDF views: 6 *
Loading metrics...

Abstract views

Total abstract views: 100 *
Loading metrics...

* Views captured on Cambridge Core between September 2016 - 17th January 2018. This data will be updated every 24 hours.