Plato, Republic, ed. John Burnet, 5.469

Current edition Perseus

Πολιτεία (Greek) (Platonis Opera Tomus IV: Tetralogia VIII. Burnet, John, editor. Oxford: Clarendon Press, 1905.)

Commentary

Commentary on Plato, Republic (Plato. The Republic, Vol 1-2. Adam, James, editor. Cambridge: Cambridge University Press, 1902-1903.) focus

Πλάτωνος Πολιτεία . On the name, characters, and date of action of the dialogue, see Introd. §§ 1, 2, 3.

κατέβην κτλ. Dionys. Hal. de comp. verb. p. 208 (Reiske) ὁ δὲ Πλάτων, τοὺς ἑαυτοῦ διαλόγους κτενίζων καὶ βοστρυχίζων, καὶ πάντα τρόπον ἀναπλέκων, οὐ διέλιπεν ὀγδοήκοντα γεγονὼς ἔτη. πᾶσι γὰρ δή που τοῖς φιλολόγοις γνώριμα τὰ περὶ τῆς φιλοπονίας τἀνδρὸς ἱστορούμενα, τά τ’ ἄλλα, καὶ δὴ καὶ τὰ περὶ τὴν δέλτον ἣν τελευτήσαντος αὐτοῦ λέγουσιν εὑρεθῆναι ποικίλως μετακειμένην τὴν ἀρχὴν τῆς πολιτείας ἔχουσαν τήνδε “κατέβην χθὲς εἰς Πειραιᾶ μετὰ Γλαύκωνος τοῦ Ἀρίστωνος.” See also Quint. VIII 6. 64, and Diog. Laert. III 37. The latter gives as his authorities Euphorion and Panaetius. As Cicero was tolerably familiar with the writings of Panaetius, it is possible that he too has the same story in view in de Sen. V 13, where he says of Plato

“scribens est mortuus.”

The anecdote may well be true, but does not of course justify any inference as to the date of composition of the Republic. See Introd. § 4.

τῇ θεῷ . What goddess? Bendis or Athena? The festival is the Bendideia (354 A) and it is perhaps safest to acquiesce in the usual view that Bendis is here meant. “Alii Minervam intelligunt, quae vulgo ἡ θεὸς appellabatur; neque mihi videtur Socrates in ista Panathenaeorum propinquitate de Minerva veneranda cogitare non potuisse: sed quod simpliciter τὴν ἑορτὴν dicit, numina diversa statuere non sinit” (Schneider). We hear of a temple of Bendis in the Piraeus in 403 B.C. ( τὴν ὁδὸν ἡ φέρει πρός τε τὸ ἱερὸν τῆς Μουνυχίας Ἀρτεμίδος καὶ τὸ Βενδίδειον Xen. Hell. II 4. 11 ). See also Introd. § 3 and App. I.

νῦν πρῶτον . Perhaps 410 B.C. Introd. § 3.

οἱ Θρᾷκες . Probably resident aliens (as opposed to the ἐπιχώριοι or natives), living for commercial purposes in the Piraeus, which at all times contained a large admixture of foreign population. It was part of Athenian policy to encourage commercial settlers by allowing them to exercise their own cults (Foucart des assoc. relig. chez les Grecs p. 131). Foucart holds that the worship of the Thracian goddess Bendis was brought to the Piraeus by Thracian merchants (p. 84). Others have supposed that οἱ Θρᾷκες refers to envoys from Thrace, or Thracian mercenaries, the survivors of those who came to Athens in 414 B.C. (Thuc. VII 27); but the other view is more probable.

τὸ ἄστυ or

ἄστυ 327 C is regular for Athens itself as opposed to the Piraeus. Hartman would omit the article (cf. Lys. 13. 88

τοὺς ἐν ἄστει οἱ ἐν τῷ Πειραιεῖ

): but it occurs infra 328 C, Phaedr. 230 C , Arist. Pol. Ath. 38. 1 and elsewhere.

αὐτός : ‘ ipse’ ‘ erus’ ‘ the master’ as often: cf. e.g. Prot. 314 D οὐ σχολὴ αὐτῷ and the Pythagorean αὐτὸς ἔφα. With the deictic

οὗτος cf. Symp. 175 A Σωκράτης οὗτος—ἕστηκεν, ‘there goes Socrates—standing.’

ἓν λείπεται . See cr. n. ἐλλείπεται (which Hermann and others retain) is less pointed, in view of the two alternatives ἢ—κρείττους γένεσθε ἢ μένετ’ αὐτοῦ. For λείπεται said of the μεταξύ τι ( Symp. 202 A ) or third alternative, cf. Theaet. 188 A ἄλλο γ’ οὐδὲν λείπεται περὶ ἕκαστον πλὴν εἰδέναι ἢ μὴ εἰδέναι.

ὡς ‐‐ διανοεῖσθε : ‘well, you may make up your mind that we shall refuse to listen.’ Cf. (with Stallbaum) Crat. 439 C διανοηθέντες—ὡς ἰόντων ἁπάντων ἀεὶ καὶ ῥεόντων.

μὴ is owing to the imperative: cf. Soph. O. C. 1154 and Jebb's note.

λαμπὰς κτλ . λαμπάς was the official name for a torch-race: see Mommsen Heortologie pp. 170 note, 282.

τῇ θεῷ : see on 327 A and App. I.

λαμπάδια : Harpocratio remarks ἣν νῦν ἡμεῖς λαμπάδα καλοῦμεν, οὕτως ὠνόμαζον. But λαμπάς was used for ‘torch’ even in classical Greek. Plato chooses λαμπάδιον because he has just used λαμπάς in a different sense.

διαδωσουσιν κτλ. shews that—except for the novel substitution of mounted competitors for runners—the torch-race in question was of the kind alluded to in Hdt. VIII 98 and elsewhere as held in honour of Hephaestus. The competition was not between one individual and another, but between different lines of competitors, the torch being passed on from man to man. Victory fell to the chain whose torch, still burning, first reached the goal. The well-known figure in Laws 776 B καθάπερ λαμπάδα τὸν βίον παραδιδόντας ἄλλοις ἐξ ἄλλων refers to the same form of race. Plato nowhere mentions the simpler form described by Pausanias (1 30. 2), in which individuals contended against each other: see Baumeister Denkmäler d. kl. Altert. p. 522.

ἄξιον θεάσασθαι . Songs and dances were the leading features in a παννυχίς. See Soph. Ant. 1146 —1152 and Eur. Heracl. 781 —783 ἀνεμόεντι δὲ γᾶς ἐπ’ ὄχθῳ | (the Acropolis) ὀλολύγματα παννυχίοις ὑπὸ παρθένων ἰακχεῖ ποδῶν κρότοισιν (in honour of Athena at the Panathenaea).

ἐξαναστήσομεθα κτλ. The promise is nowhere fulfilled.

μὴ ἄλλως ποιεῖτε . Schanz ( Novae Comm. Plat. p. 25) shews that this phrase, which is tolerably frequent in Plato, always occurs in combination with a positive command (here μένετε) except in II 369 B.

εἰς τοῦ Πολεμάρχου . Polemarchus was older than Lysias (infra 331 D), and we are to infer that at this time Cephalus lived with him. There is no reason why we should (with Blass Att. Ber. p. 338) reject Plato's statement that Polemarchus had a house in the Piraeus: the words of Lysias (12. 16), which Blass relies upon as shewing that Polemarchus lived not in the Piraeus, but in Athens, refer to 404 B.C. and do not prove it even for that year. Lysias probably lived at this time in a house of his own in the Piraeus, as in 404 B.C. ( Lys. 12. 8 ): it is to be noted that he is mentioned along with the visitors, in contrast with Cephalus ( ἦν δ’ ἔνδον κτλ. —τεθυκὼς γὰρ ἐτύγχανεν ἐν τῇ αὐλῇ infra C). Cf. Boeckh Kl. Schr. IV p. 475 note 1 and Shuckburgh Lys. Orat. ed. 2 p. xii.

διὰ χρόνου ‐‐ αὐτόν . καί ‘indeed’ goes with the whole clause: cf. Soph. Ant. 1253 ἀλλ’ εἰσόμεσθα μή τι καὶ κατάσχετον | κρυφῇ καλύπτει καρδίᾳ θυμουμένη with Jebb's note. Tucker translates ‘for it was some time since I had so much as seen him’—throwing, I think, too much emphasis on καί.

προσκεφαλαίου τε καὶ δίφρου : virtually a hendiadys, as Hartman remarks, comparing Homer Il. IX 200 εἷσεν δ’ ἐν κλισμοῖσι τάπησί τε πορφυρέοισιν. It is somewhat fanciful to suppose (with Hartman) that Plato throughout this picture was thinking of the aged Nestor seated among his sons ( Od. III 32 ff.). τινος adds a touch of vagueness: ‘a sort of combination of cushion and chair’ (Tucker).

τεθυκὼς γάρ explains ἐστεφανωμένος: “coronati sacrificabant, ut satis constat” Stallbaum. The God to whom Cephalus had been sacrificing was doubtless Ζεὺς ἑρκεῖος, whose altar stood in the αὐλή.

οὐδὲ ‐‐ Πειραιᾶ . A negative must be supplied, “ut amice expostulabundus cum Socrate senex hoc dicere videatur: tu neque alia facis, quae debebas, neque nostram domum frequentas. Simili ellipsi nostrates: Du kommst auch nicht oft zu uns” (Schneider). οὐδέ is ‘also not’: for exx. see Riddell Digest of Platonic Idioms § 141 and Jebb on Soph. O. C. 590 f. οὐδέ in οὐδὲ πάνυ ῥᾴδιον IX 587 C is another instance, in which, as here, the idiom has a kind of colloquial effect. Stallbaum takes οὐδέ with θαμίζεις “ne ventitas quidem ad nos, h. e. raro sane domum nostram frequentas”; but his equation hardly holds good, and is not justified by Xen. Symp. 4. 23. where οὐδέ coheres closely with the emphatic σοῦ. Others have suspected corruption, proposing οὔ τι (Ast, cf. Od. V 88 πάρος γε μὲν οὔ τι θαμίζεις), οὐ δέ (Nitzsch), or οὐ δή (Hartman). οὔ τι is very unlikely; for θαμίζω is not exclusively a poetic word (cf. Laws 843 B), and we need not suppose that Plato is thinking of Homer. I agree with Hartman that οὐ δέ is improbable: δέ is not sufficiently explained by saying that it is “adversative to the idea contained in ἠσπάζετο” (J. and C., with Schneider Additamenta p. 2). None of the cases quoted by Sauppe Ep. Crit. ad G. Hermannum p. 77 ( Ar. Knights 1302 , Hdt. IX 108, Theogn. 659, 887, 1070 and Callinus I 2) seem to me to justify the change of οὐδέ to οὐ δέ. Hartman's correction is better: but I believe the text is sound.

μὴ οὖν κτλ. To this sentence Lach. 181 B C furnishes a near parallel. νεανίαις refers to Socrates' companions who had come from Athens, as opposed to Cephalus, Polemarchus and the others; the emphasis, as often, being on the καί clause: ‘associate with these young men, but come and visit us also.’ So also Boeckh Kl. Schr. IV p. 475. There is no sufficient reason for reading νεανίσκοις (with II and other MSS): see Introd. § 3.

καὶ μὴν κτλ. : ‘Indeed, Cephalus,’ etc. γε need not be added (with II and other MSS) after χαίρω: cf. Phaed. 84 D καὶ μήν, ὦ Σώκρατες, τἀληθῆ σοι ἐρῶ, Euthyd. 275 E 304 C al., with Jebb on Soph. O. T. 749 , 1005.

τραχεῖα καὶ χαλεπή κτλ. The language (as Ast observes) is perhaps suggested by Hesiod OD. 290 ff. μακρὸς δὲ καὶ ὄρθιος οἶμος ἐς αὐτὴν | καὶ τρηχὺς τὸ πρῶτον· ἐπὴν δ’ εἰς ἄκρον ἵκηται, | ῥηιδίη δὴ ἔπειτα πέλει, χαλεπή περ ἐοῦσα. Cf. II 364 D note

ἐπὶ γήραος οὐδῷ . The phrase occurs first in the Iliad (XXII 60, XXIV 487) to denote the natural limit of the life of man. Cephalus is μάλα πρεσβύτης 328 B. The same meaning suits also in Od. XV 246 ( οὐδ’ ἵκετο γήραος οὐδόν) 348 and XXIII 212, Hymn. Aphr. 106, Hes. OD. 331, Hdt. III 14 and elsewhere. Leaf can hardly (I think) be right in explaining οὐδῷ as= ὁδῷ in Il. XXII 60. γήραος is a descriptive genitive (like τέλος γήραος ἀργαλέου Mimn. Fr. 2. 6, τοῦ λογου in δόλιχον—not δολιχὸν—τοῦ λόγου Prot. 329 A ), old age being itself the threshold by which we leave the House of Life. We enter as it were by one door and pass out by another. The idea underlying the phrase may be compared with Democritus' ὁ κόσμος σκηνή, ὁ βίος πάροδος· ἠλθες, εἶδες, ἀπῆλθες (Mullach Fr. Phil. Gr. I p. 356).

χαλεπὸν κτλ . χαλεπόν is neuter on account of τοῦτο in ὅ τί σοι φαίνεται τοῦτο, and τοῦ βίου is a simple partitive genitive: cf. Xen. Mem. I 6. 4 ἐπισκεψώμεθα τί χαλεπὸν ᾔσθησαι τοὐμοῦ βίου. I cannot agree with Tucker in rendering ‘disagreeable in respect of the sort of life.’ Ast takes χαλεπόν as masc. (comparing cases like III 416 B τὴν μεγίστην τῆς εὐλαβείας), but αὐτό shews that he is wrong. Translate simply ‘whether it is a painful period of life.’ It is needless to insert (with Hartman) τι after χαλεπόν: still worse is Liebhold's addition of τέλος.

ἐξαγγέλλεις : like the ἐξάγγελος in tragedy, Cephalus is the bearer of news from behind the scenes.

παροιμίαν . ἥλιξ ἥλικα τέρπει ( Phaedr. 240 C ).

ξυνιόντες : i.q. ὅταν ξυνίωσιν ‘whenever they come together.’ Such a use of the participle is admissible when the main verb is in the present of habitual action. ξυνόντες is a needless conjecture.

οὐδὲ ζῶντες . Soph. Ant. 1165 —1167 τὰς γὰρ ἡδονὰς | ὅταν προδῶσιν ἄνδρες, οὐ τίθημ’ ἐγὼ | ζῆν τοῦτον, ἀλλ’ ἔμψυχον ἡγοῦμαι νεκρόν. Cf. also Mimn. Fr. I. 1 ff.: Sim. Fr. 71 τίς γὰρ ἁδονᾶς ἄτερ<*> θνατῶν βίος ποθεινός: Eur. Fr. 1065 . Similar sentiments are very common throughout Greek literature, especially in poetry.

ἔτι ‐‐ συγγίγνεσθαι . These words are rejected by Hirschig, Cobet, and Hartman, but their genuineness is supported by the singular αὐτό in αὐτὸ ἀπέφυγον and by Plut. περὶ φιλοπλουτίας 5. 525 A ὁ Σοφοκλῆς ἐρωτηθεὶς εἰ δύναται γυναικὶ πλησιάζειν, Εὐφήμει, ἄνθρωπε, εἶπεν κτλ. In such matters Greek realism called a spade a spade. In spite of the anecdote here told, few writers have painted sadder pictures of old age than Sophocles: see for example O. C. 1235— 1238 and Fr. 684. More in keeping with the present passage is Fr. 688 οὐκ ἔστι γῆρας τῶν σοφῶν, ἐν οἷς ὁ νοῦς | θείᾳ ξύνεστιν ἡμέρᾳ τεθραμμένος.

ἀπέφυγον ‐‐ ἀποφυγών . The repetition adds a certain impressiveness to the sentence. Herwerden is in error when he ejects ἀποφυγών, which seems to have been read also by Plutarch (referred to in last note).

κατατείνουσαι is intransitive. If the meaning were (as Ast holds) transitive —man being conceived as the puppet of the desires cf. Laws 644 E—we should expect ἐπι‐ or συν‐ rather than κατατείνουσαι: see Phaed. 94 C and 98 D.

παντάπασιν κτλ. The impressive iteration is in keeping with the age and earnestness of the speaker: cf. 331 A, B.

ἐστι . Stallbaum and others eject this word, but it is not easy to see why a scribe should have inserted it, particularly in such an idiomatic position. The asyndeton before δεσποτων is regular in explanatory clauses. I read ἐστι (with A) in preference to ἔστι: the meaning ‘is possible’ does not suit, and would require ἀπαλλαγῆναι rather than ἀπηλλάχθαι. Translate ‘it is the deliverance once and for all from tyrants full many and furious.’ The grammatical subject, as in English, remains vague; it is involved in ἐπειδὰν —χαλάσωσιν. For the use of ἐστι cf. Euthyphr. 2 D φαίνεταί μοι—ἄρχεσθαι ὀρθῶς· ὀρθῶς γάρ ἐστι τῶν νέων πρότερον ἐπιμεληθῆναι. The sentence-accent falls on πολλῶν and μαινομένων and not on ἐστι. The view of old age presented here recalls the μελέτη θανάτου of the Phaedo.

εὔκολοι . Like Sophocles himself: ὁ δ’ εὔκολος μὲν ἐνθάδ’ , εὔκολος δ’ ἐκεῖ ( Ar. Frogs 82 ).

ἐκίνουν . κινεῖν ‘rouse’ is technical in the Socratic dialect for the stimulating of the intellect by interrogation: cf. (with Stallbaum) Lys. 223 A , Xen. Mem. IV 2. 2. See also Ar. Clouds 745 .

οὐ μέντοι γε . The collocation μέντοι γε, which rarely occurs in good Greek, is condemned by Porson (on Eur. Med. 675 ) and others. In Plato it is found only here and in Crat. 424 C , [ Sisyph.] 388 A. Here some inferior MSS omit γε. It would be easy (with Hoefer de particulis Plat. p. 38, Cobet, and Blaydes) to write οὐ μέντοι ὅσον γε, but “notanda talia potius quam mutanda.” The idiom, though exceptional, is (in my judgment) sufficiently supported (see the instances cited by Blaydes on Ar. Thesm. 709 ). It should also perhaps be remembered that the speaker, Cephalus, was not a native Athenian. Cf. 331 B E notes

τὸ τοῦ Θεμιστοκλέους . The story as told by Herodotus VIII 125 is probably more true, if less pointed: ὡς δὲ ἐκ τῆς Λακεδαίμονος ἀπίκετο (sc. Θεμιστοκλῆς) ἐς τὰς Ἀθήνας, ἐνθαῦτα Τιμόδημος Ἀφι<*>ναῖος—φθόνῳ καταμαργέων ἐνείκεε τὸν Θεμιστοκλέα—ὡς διὰ τὰς Ἀθήνας ἔχοι τὰ γέρεα τὰ παρὰ Λακεδαιμονίων, ἀλλ) οὐ δἰ ἑωυτόν. ὁ δὲ—εἶπε· οὕτω ἔχει τοι· οὔτ) ἂν ἐγὼ ἐὼν Βελβινίτης (Belbina was a small island about 2 miles south of Sunium) ἐτιμήθην οὕτω πρὸς Σπαρτιητέων, οὔτ’ ἂν σὺ ὤνθρωπε ἐὼν Ἀθηναῖος. The changes are not due to Plato: for τῷ in τῷ Σεριφίῳ—for which Heindorf on Charm. 155 D wrongly suggests τῳ, like Cicero's Seriphio cuidam ( Cato Mai. 8)—shews that Plato's form of the story was also familiar. The Platonic version, in which Belbina has become Seriphus, and Themistocles' detractor a Seriphian, afterwards held the field.

καὶ τοῖς δή . καί is ‘also’ and δή illative.

εὔκολος ‐‐ ἑαυτῷ . The dative is used as with εὐμενής: cf. Ar. Frogs 359 μηδ’ εὔκολός ἐστι πολίταις (v.l. πολίτης). To suit the application precisely to the story we should require (1) neither would the ἐπιεικής easily endure old age with poverty, (2) nor the μὴ ἐπιεικής easily endure old age with riches. For (2) Plato substitutes ‘nor would the bad man ever attain to peace with himself by becoming rich’; thereby conveying the further idea that the bad man is not εὔκολος ἑαυτῷ under any circumstances or at any time. Richards' suggestion ἐν αὐτῷ (i.e. γήρᾳ) for ἑαυτῷ is neat, but loses sight of this additional point. The allusion to old age in the second clause, so far as it is necessary to allude to it, is contained in ποτε.

ποἶ ἐπεκτησάμην κτλ. : ‘do you want to know what I acquired, Socrates?’ ποῖα is simply ‘what’ as in Men. 87 E σκεψώμεθα δὴ καθ’ ἕκαστον ἀναλαμβάνοντες, π οῖά ἐστιν ἃ ἡμᾶς ὠφελεῖ. ὑγίειά φαμεν καὶ ἰσχὺς καὶ κάλλος καὶ πλοῦτος δή· ταῦτα λέγομεν καὶ τὰ τοιαῦτα ὠφέλιμα, and in the usual τὰ ποῖα ταῦτα; There is no derision implied, as in ποῖος Κτήσιππος ( Euthyd. 291 A ) and the like: had Cephalus desired to pour scorn on the suggestion, he would have said πόθεν ἐπεκτησάμην; (cf. Crat. 398 E ): and it would be absurd to deride a charge to which you at once plead guilty ( γέγονα χρηματιστής κτλ.). If Socrates' question had been not πότερον—τὰ πλείω παρέλαβες ἢ ἐπεκτήσω, but ποῖα ἐπεκτήσω, Cephalus would have said ὁποῖα ἐπεκτησάμην: but this idiom is inadmissible, except where the same interrogative occurs in its direct form in the original question. In view of the answer ( μέσος τις κτλ.) which Cephalus gives, πόσα for ποῖα would be too precise. Of the various emendations which have been suggested, the only plausible one (in point of sense) is Richards' πότερον for ποἶ or ποῖ: this would assimilate the original and the repeated question, but is less well adapted to Cephalus' reply. Cephalus in point of fact uses an old man's privilege and accommodates his interrogator's question to his own reply. See also V 465 E note

Λυσανίας δέ . Groen van Prinsterer's suggestion ( Platon. Prosopogr. p. 111) Λυσίας for Λυσανίας is at first sight plausible, since it is in harmony with the well-known Greek custom of calling grandsons after their grandfathers: but the fashion was by no means invariable: see Blümner, Gr. Privatalterth. p. 284. [Plut.] vit. Lys. 835 C also calls Cephalus son of Lysanias.

τούτοισιν . Bekker and others read τουτοισί, but there is no reason for deserting the MSS. The archaic dative in ‐οισι is tolerably often used by Plato. In the Republic alone it recurs in 345 E, 388 D, 389 B, 468 D (Homer), 560 E, 564 C, 607 B ( ‐αισι) (poetic): see also Schneider on III 389 B, and for the usage of inscriptions Meisterhans^{3} p. 126. In this particular passage the archaic ending suits the age of the speaker; but it should be remembered that Plato's style (at least in his more mature dialogues) is not a mere reproduction of the vernacular Attic, but also in no small measure a literary language or ‘Kunstsprache,’ in which Ionisms and poetic and archaic forms are occasionally employed: see especially Hirzel Der Dialog I pp. 246—250 notes Hirzel (ib. p. 34 note 1) gives reasons for holding that a sort of κοινὴ διάλεκτος, resembling the dialect of Herodotus, was actually spoken in certain cultivated circles at Athens in the Periclean age, e.g. by Anaxagoras and his group, by the Ionian sophists and their followers etc., and some of Plato's Ionisms may be inherited from this source. Cf. VII 533 B note

οὗ τοι ἕνεκα ‐‐ ὅτι . The reading τούτου for οὗ, though supported by Stobaeus ( Flor. 94. 22), is a correction made by some one unacquainted with the idiom, which is common enough in conversational style: cf. infra 491 B ὃ μὲν πάντων θαυμαστότατον ἀκοῦσαι, ὅτι κτλ. and Ar. Frogs 108 . Hartman's τοῦ τοι (interrogative) is ingenious, but unnecessary.

διπλῇ ἢ οἱ ἄλλοι . The meaning is simply ‘twice as much as the others’: cf. e.g. Laws 868 A διπλῇ τὸ βλάβος ἐκτεισάτω and 928 B ζημιούτω— διπλῇ. The ἤ is like ἤ after διπλάσιος, πολλαπλάσιος etc. If διπλῇ meant simply ‘on two grounds,’ it could not be followed by ἤ, and we should have to regard ἢ οἱ ἄλλοι as an interpolation. Cephalus expresses himself somewhat loosely, as if loving a thing on two grounds, or in two ways, were equivalent to loving it twice as much. ταύτῃ below is defined by the ὥσπερ clause, and is preferred to ὥσπερ, partly in order to correspond to διπλῇ but stiil more to suit κατὰ τὴν χρείαν. The present passage is through Aristotle ( Eth. Nic. IV 2. 1120^{b} 14, cf. ib. IX 7. 1168^{a} 1—3) the source of the proverb about ‘parents and poets.’

ξυγγενέσθαι : ‘to meet’ in social intercourse, as in Ap. 41 A . ξυγγίγνεσθαι (suggested by Richards) would express habitual intercourse, which is not what Plato means to say. With the sentiment cf. Symp. 173 C ὅταν μέν τινας περὶ φιλοσοφίας λόγους ἢ αὐτὸς ποιῶμαι ἢ ἄλλων ἀκούω—ὑπερφυῶς ὡς χαίρω· ὅταν δὲ ἀλλους τινάς, ἄλλως τε καὶ τοὺς ὑμετέρους τοὺς τῶν πλουσίων καὶ χρηματιστικῶν, αὐτός τε ἄχθομαι ὑμᾶς τε τοὺς ἑταίρους ἐλεῶ, ὅτι οἴεσθε τὶ ποιεῖν οὐδὲν ποιοῦντες.

ἐπειδὰν ‐‐ τελευτήσειν : ‘when a man faces the thought that he must die,’ not (with Jowett) ‘when a man thinks himself to be near death,’ which would be ἐπειδάν τις ἐγγὺς εἶναι οἴηται τοῦ τελευτῆσαι, as Herwerden proposes to read (cf. Laws 922 C ὅταν ἤδη μέλλειν ἡγώμεθα τελευτᾶν). “Senum, non iuvenum τὸ οἴεσθαι τελευτήσειν est” (Hartman): the weakness of old age convinces us at last that we too must die. Cf. Simon. 85. 7—10 θνητῶν δ’ ὄφρα τις ἄνθος ἔχῃ πολυήρατον ἥβης | κοῦφον ἔχων θυμόν, πόλλ’ ἀτέλεστα νοεῖ· | οὔτε γὰρ ἐλπίδ’ ἔχει γηρασέμεν οὔτε θανεῖσθαι, | οὐδ’ ὑγιὴς ὅταν ᾖ, φροντίδ’ ἔχει καμάτου .

ἀδικήσαντα ‐‐ διδόναι δίκην . Plato is fond of this verbal play: cf Euthyph. 8 B and 8 E τῷ γε ἀδικοῦντι δοτέον δίκην. He who does not render justice in deeds must render justice in punishment: for the tale of justice must be made up. Note that we have here in ἀδικία and δίκη the first casual allusion to the subject of the Republic.

αὐτὸς κτλ . αὐτός= ipse s. ultro as opposed to οἱ λεγόμενοι μῦθοι. The verb is to be supplied by a kind of zeugma from μᾶλλόν τι καθορᾷ αὐτά (i.e. τὰ ἐκεῖ; or rather the predicate is accommodated to the second alternative. Cf. 344 B infra and VIII 553 C. To regard the bodily weakness of old age as in itself the cause of clearer vision of the world beyond may be in harmony with the doctrine of the Phaedo, but Cephalus is not represented as a Platonist. Tucker needlessly doubts the text.

καὶ ἐκ τῶν ὕπνων κτλ . καί is ‘both,’ not ‘and,’ and balances καὶ ζῇ: ‘many a time, like children, awakes out of sleep in terror and lives in the expectation of ill.’ For ὥσπερ οἱ παῖδες compare Phaed. 77 D , E, and for the general sentiment Arist. Eth. Nic. I 13. 1102^{b} 8—11 ἀργία γάρ ἐστιν ὁ ὕπνος τῆς φυχῆς ᾗ λέγεται σπουδαία καὶ φαύλη, πλὴν ε<*> πῃ κατὰ μικρὸν διικνοῦνταί τινες τῶν κινήσεων, καὶ ταύτῃ βελτίω γίνεται τὰ φαντάσματα τῶν ἐπιεικῶν ἢ τῶν τυχόντων.

ἡδεῖα ‐‐ γηροτρόφος . ἡδεῖα is suggested by Pindar's γλυκεῖα, and καὶ ἀγαθή, as presently appears, is not part of the quotation, but goes with ἐλπίς and is added by Plato in contrast to μετὰ κακῆς ἐλπίδος.

γηροτρόφος κτλ. : ‘to nurse him in old age, as Pindar also says.’ γηροτρόφος is best taken by itself and not with ἀγαθή .

ἀτάλλοισα κτλ . ἀτάλλω is used of rearing children, and helps out the idea of γηροτρόφος: δὶς παῖδες οἱ γέροντες. It is not clear how the fragment is to be arranged, nor to what class of Pindar's poems it belongs. See Bergk Poet. Lyr. Gr.^{4} I p. 452.

εὖ οὖν ‐‐ σφόδρα . The emphasis is quite in keeping with Cephalus' age and character; and Hartman is certainly wrong in condemning the clause: cf. 329 C, 331 B.

ὀφείλοντα ‐‐ θεῷ θυσίας τινάς . Phaed. 118 A εἶπεν, ὃ δὴ τελευταῖον ἐφθέγξατο, ὦ Κρίτων, ἔφη, τῷ Ἀσκληπιῷ ὀφείλομεν ἀλεκτρυόνα· ἀλλὰ ἀπόδοτε καὶ μὴ ἀμελήσητε. Wealth is in Cephalus' view the indispensable χορηγία ἀρετῆς.

ἀλλά γε ἓν ἀνθ) ἑνός . ἀλλά γε is extremely rare in Attic prose: in the Platonic corpus it occurs—according to the best manuscript authority—here and in Rep. VIII 543 C, Phaed. 86 E , Hipp. Maior 287 B, Phaedr. 262 A ( ἀλλά γε δή), Phaed. 116 D (id.). In some of these passages ἀλλ’ ἄγε has been conjectured — wrongly, as I think (with Schneider), at all events in the passage from the Republic:—but ἀλλ’ ἄγε cannot be read in the Phaedrus and Hippias Maior. There is no a priori objection to the collocation, which is also implied in ἀλλὰ γάρ ( γ’ ἄρα); and in later Greek ἀλλά γε aroused no objection. The meaning is ‘but still,’ originally ‘yes, but’: as Schneider says, “ γε in his dictionibus concedit aliquatenus praecedentia, sed magis urget sequentia.” There is perhaps also a dramatic motive for putting ἀλλά γε into the mouth of Cephalus: see on οὐ μέντοι γε in 329 E. Against the reading of Stobaeus ( Flor. 94. 22) ἀλλὰ ἕν γε ἀνθ’ ἑνός, we may urge the further objection that the idiomatic phrase ἓν ἀνθ’ ἑνός (‘setting one thing against another,’ as Jowett correctly translates it) seems to depend for its peculiar force (like μόνος μόνῳ and the like) on the juxtaposition of its two parts: cf. Phil. 63 B ( ἓν ἀνθ’ ἑνός) and Laws 705 B ( ἀνθ’ ἑνὸς ἕν). The passage quoted by Stallbaum from Euripides Orest. 651 ἓν μὲν τόδ’ ἡμῖν ἀνθ’ ἑνὸς δοῦναί σε χρή is quite different and does not mean ‘hoc praecipue,’ but ‘one thing in return for one thing,’ as is clear from lines 646 f.

οὐκ ἐλάχιστον is not adverbial (as Hartman and others suppose), but belongs to τοῦτο: ‘setting one thing against another, I should regard this as not the least important object for which wealth is most useful to a man of sense.’ The emphasis is characteristic: cf. 329 C, 331 A.

τὴν ἀλήθειαν κτλ. This theory of justice or righteousness is deduced from the words of Cephalus: τὸ γὰρ μηδὲ ἄκοντά τινα ἐξαπατῆσαι ἢ ψεύσασθαι being generalised into ἀλήθειαν (truthfulness, cf. τἀληθῆ λέγειν below), and μηδ’ αὖ ὀφείλοντα ἢ θεῷ θυσίας τινὰς ἢ ἀνθρώπῳ χρήματα into ἀποδιδόναι ἄν τίς τι παρά του λάβῃ. Cf. (with Wohlrab) Mimn. Fr. 8 ἀληθείη δὲ παρέστω | σοὶ καὶ ἐμοί, πάντων χρῆμα δικαιότατον. It is simply Truth and Honesty, the two chief ingredients in the popular conception of morality.

ἁπλῶς οὕτως : ‘quite without qualification.’ For this idiomatic οὕτως cf. ῥᾳδιως οὕτω II 377 B note

οἷον τοιόνδε λέγω . Similar points of casuistry are raised in Socrates' conversation with Euthydemus ap. Xen. Mem. IV 2 12 ff.

οὐδ’ αὖ κτλ. I have removed the comma before οὐδέ, because the ὁ in ὁ ἀποδιδούς covers both participles, the person in both cases being the same.

ἔφη . There is not sufficient reason for changing the best supported reading ἔφη, ἐγω to ἔφην ἐγώ. Polemarchus is throughout the introduction represented as a vivacious person: e.g. in ὁρᾷς οὖν ἡμᾶς—ὅσοι ἐσμέν (327 C), and in the lively emphasis with which he breaks in just above: πάνυ μὲν οὖν—εἴπερ γέ τι χρὴ Σιμωνίδῃ πείθεσθαι. True to his name, he is first to mingle in the fray. It is this φιλολογία on the part of his son which draws a smile from Cephalus: over-much προθυμία always struck the Greeks as laughable: cf. e.g. Eur. Ion 1172 ff. The words in which Socrates addresses Polemarchus σὺ ὁ τοῦ λόγου κληρονόμος are also somewhat more appropriate if the title was self-chosen. Cephalus leaves the argument to be carried on by the assembled company (for ὑμῖν does not mean Polemarchus and Socrates alone): whereupon Polemarchus, seizing hold on the word παραδίδωμι in its sense of ‘transmit,’ ‘bequeath,’ playfully claims the right to inherit his λόγος as Cephalus' eldest son and heir. It may be added that ἔφη ἐγώ was much more likely to be changed to ἔφην ἐγώ than vice versa. With the Greek compare Phaed. 89 C ἀλλὰ καὶ ἐμέ, ἔφη, τὸν Ἰόλεων παρακάλει.

ἅμα ᾔει πρὸς τὰ ἱερά . Soph. Fr. 206 γήρᾳ πρεπόντως σῷζε την εὐφημίαν. The editors quote Cicero Epp. ad Att. IV 16. 3 “credo Platonem vix putasse satis consonum fore, si hominem id aetatis in tam longo sermone diutius retinuisset.” Cf. the words of Theodorus in Theaet. 162 B οἶμαι ὑμᾶς πείσειν ἐμὲ μὲν <*>ᾶν θεᾶσθαι καὶ μὴ ἕλκειν πρὸς τὸ γυμνάσιον, σκληρὸν ἤδη ὄντα, τῷ δὲ δὴ νεωτέρῳ τε<*>καὶ ὑγροτέρῳ ὄντι προσπαλαίειν. It is worthy of note that the entrance and exit of Cephalus are alike associated with the services of religion: see 328 C and Introd. § 2.

ὁ τοῦ λόγου κληρονόμος . See on παῖδες ἐκείνου τοῦ ἀνδρός II 368 A.

τὰ ὀφειλόμενα ‐‐ ἐστι . Probably some current saying attributed to Simonides: there is nothing like it in his fragments. The words do not profess to be a definition of justice: if they did, τό would appear before δίκαιον. It is not likely that Simonides himself explained this particular saying as Polemarchus does, although he would not have disapproved of the explanation. In Xen. Hier. II 2 he is represented as saying that tyrants are ἱκανώτατοι—κακῶσαι μὲν ἐχθρούς, ὀνῆσαι δὲ φίλους. The words of Socrates σὺ μέν, ὦ Πολέμαρχε, ἴσως γιγνώσκεις, ἐγὼ δὲ ἀγνοῶ tend to fix the responsibility of the explanation on Polemarchus alone. Probably Simonides (if the saying is his) meant no more than that we should ‘render unto Caesar the things which are Caesar's.’ Plato virtually confesses in 332 B that his interpretation is forced.

ἔμοιγε : said with confidence, as Σιμωνίδῃ γε with emphasis and some mockery: with you one might disagree, but not with Simonides.

σοφὸς ‐‐ θεῖος . Cf. Prot. 315 E . σοφός and θεῖος were fashionable words of praise: in the mouth of Socrates they are generally ironical. Plato's own connotation of the word θεῖος is given in Men. 99 C οὐκοῦν, ὦ Μένων, ἄξιον τούτους θείους καλεῖν τοὺς ἄνδρας, οἵτινες νοῦν μὴ ἔχοντες πολλὰ καὶ μεγάλα κατορθοῦσιν ὧν πράττουσι καὶ λέγουσιν; Ὀρθῶς ἂν καλοῖμεν θείους τε, οὓς νῦν δὴ ἐλέγομεν χρησμῳδοὺς καὶ μάντεις καὶ τοὺς ποιητικοὺς ἅπαντας· καὶ τοὺς πολιτικοὺς οὐχ ἥκιστα τούτων φαῖμεν ἂν θείους τε εἶναι καὶ ἐνθουσιάζειν, ἐπίπνους ὄντας καὶ κατεχομένους ἐκ τοῦ θεοῦ, ὅταν κατορθῶσι λέγοντες πολλὰ καὶ μεγάλα πράγματα, μηδὲν εἰδότες ὦν λέγουσι.

ἀνήρ . I formerly read ἁνήρ, but ἀνὴρ (in the predicate) is satisfactory enough: cf. Men. 99 D θεῖος ἀνήρ, φασίν, οὗτος.

παρακαταθεμένου κτλ. Xen. Cyr. I 6. 31 ff. καὶ ἔτι προβὰς (sc. ἐπὶ τῶν ἡμετέρων προγόνων γενόμενός ποτε ἀνὴρ διδάσκαλος τῶν παίδων ) ταῦτα ἐδίδασκεν ὡς καὶ τοὺς φίλους δίκαιον εἴη ἐξαπατᾶν, ἐπί γε ἀγαθῷ, καὶ κλέπτειν τὰ τῶν φίλων, ἐπί γε ἀγαθῷ: Mem. IV 2. 17 ff.

ὁτῳοῦν is to be taken with παρακαταθεμένου and not with ἀπαιτοῦντι.

καίτοι γε ὀφειλόμενον . There is the same dispute about καίτοι γε as about μέντοι γε and ἀλλά γε (see on 329 E, 331 B). καίτοι γε has the best MS authority in its favour here and in IV 440D: elsewhere in Plato it is not well-attested except in the νοθευόμενοι, where it occurs Min. 318 E , Axioch. 364 B , 368 E. καίτοι γε is also found occasionally in Aristophanes, Xenophon, Aristotle, and the orators: see Blaydes on Ar. Ach. 611 , and the Lex. Arist. Many distinguished critics would emend the idiom everywhere; but the instances are far too numerous for such a drastic policy. The difference between καίτοι ὀφειλόμενόν γε που (which Hoefer de part. Plat. p. 38 would read) and καίτοι γε ὀφειλόμενον would seem to be that in the former more stress is thrown on the word ὀφειλόμενον, in the latter on τοι. καίτοι γε is ‘and surely’ rather than ‘quamquam’ (as Kugler holds de part. τοι eiusque comp. ap. Pl. usu p. 20), cf. IV 440 D note The periphrasis ὀφειλόμενον—ἐστι is used of course to correspond to τὰ ὀφειλόμενα in E above: such periphrases (the principle of which is explained in Euthyph. 9 E ff.) are extremely common in Plato. See W. J. Alexander in A. J. Ph. IV pp. 299 ff.

ἀποδοτέον ‐‐ ἀπαιτοῖ : ‘well, but we were not on any account to make restoration at the time when the claimant is’—according to the Greek idiom ‘was’ —‘mad.’ Socrates, as in ὅπερ ἄρτι ἐλέγομεν, is appealing to the admissions made by the πατὴρ τοῦ λόγου (in 331 C), as he is justified in doing when addressing his heir. ὁπότε is not—as τότε shews —the particle of ‘indefinite frequency,’ but stands for ὅτε of the direct: the whole clause τότε ὁπότε τις μὴ σωφρόνως ἀπαιτοῖ is thus in the oratio obliqua of self-quotation and exactly corresponds to εἰ μανεὶς ἀπαιτοῖ in 331 C. Madvig's ἀπαιτεῖ for ἀπαιτοῖ is therefore unnecessary. Goodwin MT. p. 213 explains the optative otherwise, but not (I think) rightly.

ἀγαθὸν μέν τι δρᾶν sc. αὐτούς, for τοῖς φίλοις depends on ὀφείλειν, to which τοὺς φίλους is the subject.

μανθάνω ‐‐ ὅτι . ὅτι is ‘because,’ not ‘that,’ as always (I believe) in Plato's use of this phrase: cf. Euthyph. 3 B , 9 B and infra III 402 E, VIII 568 E. For the sentiment cf. (with J. and C.) Xen. Mem. IV 2 17 ff.

ὀφείλεται δέ . See cr. n. In explanatory clauses of this kind δέ and not δέ γε is the correct usage: cf. infra 337 D, 344 A. I therefore follow Bekker in reading δέ.

προσήκει . ὀφειλόμενον has thus been equated with προσῆκον by means of the special cases τὸ τοῖς φίλοις ὀφειλόμενον and τὸ τοῖς ἐχθροῖς ὀφειλόμενον. τὸ προσῆκον is a more general term and is the regular word in classical Greek for ‘proper conduct’ or ‘duty’ (as the Greeks conceived it), the Stoic καθῆκον being very rarely used in this sense by good authors.

ᾐνίξατο ‐‐ ποιητικῶς . Theaet. 194 C τὸ τῆς ψυχῆς κέαρ, ὃ ἔφη Ὅμηρος αἰνιττόμενος τὴν τοῦ κηροῦ ὁμοιότητα. The present passage is no more serious than that in the Theaetetus: Plato knew that Simonides merely meant to say ‘it is just to render what you owe.’

ἀλλὰ τί ὄει ; is a rhetorical question, which needs and receives no answer, like τί μήν; and τί μὴν δοκεῖς; ( Theaet. 162 B ). It is equivalent to ‘of course.’ For the use of τί Stallbaum compares Gorg. 480 B τί γὰρ δὴ φῶμεν; to which there is also no reply. This explanation is preferable to that of Madvig, who gives ἀλλὰ τί οἴει to Socrates, and takes ἔφη as equivalent to συνέφη— a harsh usage in a narrated dialogue, and not likely to have been intended by Plato, because sure to be misunderstood. Liebhold's ἄλλο τι οἴει; <οὐκ> ἔφη has everything against it.

ὦ πρὸς Διὸς κτλ. ‘In the name of heaven, said I, if any one then had asked him’ etc. ‘what reply do you think he would have made to us?’ ὦ before πρὸς Διός is (as Schanz holds) an interjection, and does not require a vocative to follow it: cf. Euthyd. 287 A , 290 E. It is tempting (with Tucker) to take ὦ πρὸς Διός as part of the address to Simonides (cf. Euthyd. 294 B ὦ πρὸς τῶν θεῶν, ἦν δ’ ἐγώ, ὦ Διονυσόδωρε—αὐτὼ τῷ ὄντι πάντα ἐπίστασθον). But on this view the presence of εἰ οὖν—ἤρετο forms a difficulty, and ὦ πρὸς Διός may very well go with τί ἂν οἴει— ἀποκρίνασθαι.

ὀφειλόμενον καὶ προσῆκον . It is characteristic of Plato to combine the thing explained and the explanation itself in this way: see my note on Prot. 314 A . Here ὀφειλόμενον is necessary to enable Simonides to recognise his own saying.

ἰατρικὴ ‐‐ μαγειρική . In Gorg. 463 A ff. Plato refuses the name of ‘art’ to ὀψοποιική: it is but an ἐμπειρία or τριβή, a sort of bastard adjunct to ἰατρική, as κομμωτική is to γυμναστική. Here, where less precision is required, both are regarded as τέχναι.

εἶεν according to Timaeus ( Lexicon s. v.) expresses συγκατάθεσις μὲν τῶν εἰρημένων, συναφὴ δὲ πρὸς τὰ μέλλοντα. It rarely expresses συγκατάθεσις (‘assent’) and no more: see on IV 436 C. The word was pronounced εἶἑν with intervocalic aspiration (Uhlig in Fl. Jahrb. 1880 pp. 790 ff.) and may possibly be a compound of εἷα and ἕν (used as in ἓν μὲν τόδ’ ἤδη τῶν τριῶν παλαισμάτων Aesch. Eum. 589 ). εἶἑν is the usual orthography in Paris A, and has left some traces also in the Bodleian MS e.g. Gorg. 466 C .

τέχνη δικαιοσύνη . The Socratic view that Justice is an art—a view that dominates the whole of the conversation w<*> Polemarchus—is thus introduced quite incidentally.

τὸ ‐‐ λέγει . Cf. Xen. Hiero II 2 (cited above on 331 E).

τί δέ; ὁ δίκαιος κτλ. This punctuation throws more emphasis on ὁ δίκαιος than τί δὲ ὁ δίκαιος; which appears in some editions. It is therefore to be preferred in introducing the application of the two illustrations. So also below in 333 A τί δὲ δή; τὴν δικαιοσύνην κτλ.

προσπολεμεῖν explains ἐχθροὺς βλάπτειν as ξυμμαχεῖν explains φίλους ὠφελεῖν. Ast's προπολεμεῖν (a conjecture of Stephanus) would leave ἐχθ<*>οὺς βλάπτειν unrepresented. Stephanus' conjecture was natural enough with the wrong reading καὶ ξυμμαχεῖν, which Ast also followed. For

ἔμοιγε δοκεῖ Hartman demands ἔμοιγε δοκεῖν; but cf. 333 B, Crito 43 D, Phaed. 108 D , Menex. 236 B . These cases shew that δοκεῖ can be used without ὡς: and ἐμοὶ ( ἔμοιγε) δοκεῖν does not occur in the Republic (Grünenwald in Schanz's Beitr. <*>ur hist. Synt. d. gr. Spr. II 3 p. 12).

ξυμβολαια are contracts where money is involved. Polemarchus (as in εἰς ἀργυρίου in B below), in harmony with the natural meaning of Simonides' saying, thinks first of pecuniary dealings as the sphere in which δικαιοσύνη acts. Socrates substitutes for ξυμβόλαια the more general term κοινωνήματα, in order once more to introduce the analogy of the arts.

εἰς τίνα δὴ κοινωνίαν is idiomatic for εἰς τίνος δὴ κοινωνίαν. Compare VIII 556 C ἢ ἐν ὁδῶν πορείαις ἢ ἐν ἄλλαις τισὶ κοινωνίαις and τὴν τιμὴν ταύτην (where the English idiom would expect τὴν τιμὴν ταύτης) in II 371 E. In spite of εἰς κρουμάτων and εἰς ἀργυρίου, it is not necessary to read (with Richards) τίνος.

παρακαταθέσθαι καὶ σῶν εἶναι . The double expression is necessary to explain κοινῇ χρῆσθαι: the κοινωνία arises because one deposits the money and by the other it is kept safe.

ἄχρηστον ‐‐ χρήσιμος . ἄχρηστος fluctuates between ‘unused’ and ‘useless’: the latter sense is predominant here and gives an epigrammatic tone to the sentence (cf. ἐν μὲν χρήσει ἄχρηστος, ἐν δὲ ἀχρηστίᾳ χρήσιμος in D). It is noticeable that Plato does not take into account the possibility of money being deposited at interest: in this case the money could not be said to be useless.

καὶ κοινῇ καὶ ἰδίᾳ : not ‘to the individual and to the state,’ but ‘both in dealings with others, and in personal concerns.’ The words καὶ ἰδίᾳ are, strictly speaking, irrelevant, for it is with κοινωνήματα (in the widest sense) that we are concerned. They are to be regarded merely as a rhetorical amplification for the sake of emphasis: cf. infra 350 A, 351 A notes

οὐκ ἂν οὖν κτλ. See cr. n. Some may think that we should read οὐκοῦν (with the majority of MSS) and cancel εἴη after σπουδαῖον (so also Vind. D), understanding ἐστι. The accidental omission of ἄν is however not uncommon in Plato's MSS: see on IV 437 B.

οὗτος καὶ φυλάξασθαι . Because knowledge of anything implies knowledge also of its opposite, according to the usual Socratic view. See Phaed. 97 D οὐδὲν ἄλλο σκοπεῖν προσήκειν ἄνθρωπον—ἀλλ’ ἢ τὸ ἄριστον καὶ τὸ βέλτιστον· ἀναγκαῖον δὲ εἶναι τὸν αὐτὸν τοῦτον καὶ τὸ χεῖρον εἰδέναι, Charm. 166 E , Hipp. Min. 367 A ff. See also Stewart's Notes on the Nicomachean Ethics Vol. I p. 378.

φυλάξασθαι κτλ. See cr. n. With the emendation in the text, the argument is as follows: (1) he who can πατάξαι, can φυλάξασθαι: (2) he who can φυλάξασθαι ( νόσον), can λαθεῖν ἐμποιήσας ( νόσον): (3) he who can κλέψαι ( τὰ τῶν πολεμίων), is a good φύλαξ of an army. Thus the predicate of each step in the argument corresponds to the subject of the step next following: for λαθεῖν ἐμποιήσας ( νόσον) is to be taken as parallel to κλέψαι ( τὰ τῶν πολεμίων). The argument is unsound, and not intended to be serious: it is enough that it suffices to bewilder Polemarchus. For a further discussion on this passage see App. II.

στρατοπέδου γε κτλ. The στρατηγός must be both φυλακτικος τε καὶ κλέπτης according to Socrates in Xen. Mem. III 1. 6.

κλέπτειν and

κλέμμα were used (especially by Spartans) with reference to military operations involving surprise and stealth (Classen on Thuc. v 9. 5).

κλέπτης ‐‐ ἀναπέφανται . Cf. Hipp. Min. 365 C ff., where this view is worked out at length, ib. 369 B ἀναπέφανται ὁ αὐτὸς ὢν ψευδής τε καὶ ἀληθής and Xen. Mem. IV 2. 20 ff. ἀναπέφανται, as J. and C. remark, expresses an unexpected result—here a paradox. Like ὁ ἑκὼν ἁμαρτάνων ἀμείνων, the conclusion is a logical inference from the Socratic identification of virtue and knowledge, made without regard to experience.

ἀγαπᾷ , ‘esteems,’ is said with reference to ἐσθλόν in Hom. Od. XIX 395 f. μητρὸς ἑῆς πατέρ’ ἐσθλόν, ὃς ἀνθρώπους ἐκέκαστο | κλεπτοσύνῃ θ’ ὅρκῳ τε. The suggested ἄγαται for ἀγαπᾷ τε would be too strong: see Symp. 180 B μᾶλλον— θαυμάζουσιν καὶ ἄγανται—ὅταν ὁ ἑρώμενος τὸν ἐραστὴν ἀγαπᾷ, where the meaning of ἀγαπᾷ is shewn by οὕτω περὶ πολλοῦ ἐποιεῖτο in 180 A.

τοῦτο ‐‐ ἔτι . So Euthyphro (15 B) harks back to his first definition of piety (6 E) after he has been refuted by Socrates. Cf. also VII 515 E note

δοκεῖ does double duty, first with τοῦτο and then with δικαιοσύνη: cf. VI 493 A, VII 517 B, 525 B, 530 B and (with Stallbaum) Ap. 25 B . Hartman needlessly doubts the text.

φίλους δὲ λέγεις κτλ. The same mode of argument recurs in 339 B ff. Cf. also Hipp. Maior 284 D.

φίλοι κτλ. Schneider rightly observes that κατὰ δὴ τὸν σὸν λόγον below tends to shew that ἀλλ’ ὅμως —βλάπτειν is interrogative. The argument is in the form of a dilemma: either a it is just to injure those who do us no injustice (and benefit those who do), or b it is just to injure friends and benefit foes. The first alternative is immoral ( πονηρός), and the second directly opposed to Simonides' view. Socrates suppresses the words which I have put in brackets, because they lessen rather than increase the immorality of the conclusion: the second alternative is expressed in full as the αὐτὸ τοὐναντίον ἢ τὸν Σιμωνίδην ἔφαμεν λέγειν .

ὅσοι κτλ. : not ‘those of mankind who are in error’ (J. and C.) but ‘those who have mistaken their men’: cf. Phaedr. 257 D τοῦ ἑταίρου συχνὸν διαμαρτάνεις. So also Schneider, and Davies and Vaughan.

πονηροὶ γὰρ κτλ. Stallbaum (followed by D. and V.) wrongly takes αὐτοῖς as ‘in their eyes.’ The reasoning is difficult only from its brevity. If it is δίκαιον βλάπτειν ἀδίκους, and men sometimes suppose that a man is good when he is bad ( πονηροὶ γὰρ αὐτοῖς εἰσὶν ‘for they have bad friends’), then since friend has been defined as one whom we suppose to be good (334 C), it is sometimes δίκαιον βλάπτειν φίλους. Stallbaum's view is quite inconsistent with the definition of friends in 334 C as οὓς ἄν τις ἡγῆται χρηστούς.

τὸν φίλον καὶ ἐχθρόν . Hartman (with some inferior MSS) wishes to insert τόν before ἐχθρόν; but cf. infra III 400 D and many other examples cited by himself. To pronounce them all corrupt is to destroy the basis on which our knowledge of Platonic idiom rests.

τὸν δοκοῦντά τε ‐‐ καὶ τὸν ὄντα . The meaning required—‘he who both seems and is good’—would be more correctly expressed by τὸν δοκοῦντά τε—καὶ ὄντα (so Ast and others), but “aliquid tribuendum interpositis ἧ δ’ ὅς, quae negligentiam repetendi, si est negligentia, saltem excusant” (Schneider, who compares also infra 341 B ποτέρως λεγεις τὸν ἄρχοντά τε καὶ τὸν κρείττονα). In τὸν δὲ δοκοῦντα μέν, ὄντα δὲ μή Polemarchus expresses himself more accurately.

ὁ ἀγαθός ‐‐ ὁ πονηρός . Socrates unfairly neglects the δοκῶν, although according to Polemarchus' amended definition the ἀγαθός who seemed πονηρός would not be a friend, nor the πονηρός who seemed ἀγαθός an enemy. Polemarchus' theory indeed points to a division of men into three classes: friends, enemies, and those who are neither (viz. those who seem good and are bad, and those who seem bad and are good). The somewhat ideal view that the ἀγαθός is φίλος and the πονηρὸς ἐχθρός is genuinely Socratic (cf. Mem. II 6. 14 ff.): it is part of the wider view that all men desire the good ( Symp. 206 A , Gorg. 467 C ff.).

προσθεῖναι ‐‐ βλάπτειν . ἤ after τῷ δικαίῳ must mean ‘or in other words’: cf. infra 349 E πλεονεκτεῖν ἢ ἀξιοῦν πλέον ἔχειν and Phaed. 85 D ἐπὶ βεβαιοτέρου ὀχήματος, ἢ λόγου θείου τινός (so the Bodleian, but ἤ is cancelled by many editors). The late expression Φαίδων ἢ περὶ ψυχῆς involves essentially the same use of ἤ. .The clause ὡς—κακῶς is summed up in τούτῳ, and the whole sentence means: ‘do you wish us to make an addition to our account of justice, or in other words to say now—in addition to our original definition where we said it was just to do good to friends and harm to enemies—that it is just to do good to friends if they are good etc.’ This explanation is (I think) the least vulnerable one, if the text is to be retained. With προσθεῖναι used absolutely cf. 339 B. For other views see App. III.

ἔστιν ἄρα κτλ. Cf. Crito 49 A ff., Corg. 469 B , [ περὶ ἀρετῆς] 376 E. This chapter contains the only element of permanent ethical interest and value in the discussion with Polemarchus—the only element, moreover, which reappears in a later book of the Republic (11 379 B). The underlying principle—that κακῶς ποιεῖν = κακὸν ποιεῖν—is in accordance with the traditional Greek view of life. For illustrations we may cite Od. XVIII 136 f. τοῖος γὰρ νόος ἐστὶν ἐπιχθονίων ἀνθρώπων | οἷον ἐπ’ ἦμαρ ἄγῃσι πατὴρ ἀνδρῶν τε θεῶν τε, Arch. Fr. 70 (Bergk), and Simon. Fr. 5. 10—14 ἄνδρα δ’ οὐκ ἔστι μὴ οὐ κακὸν ἔμμεναι | ὃν ἀμάχανος συμφορὰ καθέλοι· | πράξας μὲν εὖ πᾶς ἀνὴρ ἀγαθός, | κακὸς δ’ εἰ κακῶς <τις>, | κἀπὶ πλεῖστον ἄριστοι, τούς κε θεοὶ φιλῶσιν. The same point of view is manifest in the transition of meaning in μοχθηρός and πονηρός from ‘laborious,’ ‘afflicted’ (e.g. Hesiod Fr. 95. 1 Göttling) to ‘depraved.’ Conversely, prosperity makes one morally better, as in Solon 13. 69 f. τῷ δὲ κακῶς ἔρδοντι θεὸς περὶ πάντα τίθησιν | συντυχίην ἀγαθήν, ἔκλυσιν ἀφροσύνης, and in the frequent identification of εὐπραγία or εὐδαιμονία with εὖ πράττειν e.g. Charm. 172 A , 173 D, Alc. I 116 B, Arist. Eth. Nic. I 8. 1098^{b} 20. It is by the analogy of the arts that Socrates in this chapter seeks to prove, first the identification κακῶς ποιεῖν = κακὸν ποιεῖν, and second that the good man cannot harm others: the Socratic conception of right conduct as an art is still predominant. It is important to observe that it was by means of this Socratic weapon that Plato achieved this noble anticipation of Christian ethical theory (St Matth. 5. 44 al.). Cf. also Gorg. 472 D ff.

ἀνθρώπους δὲ κτλ. Cf. 352 E— 353 E.

οὐκ ἦν σοφὸς ‐‐ εἰπών . Teichmüller ( Lit. Fehd. I p. 22 note) finds in this an allusion to Xenophon, who puts into the mouth of Socrates (addressing Critobulus in Mem. II 6. 35) the words ἔγνωκας ἀνδρὸς ἀρετὴν εἶναι, νικᾶν τοὺς μὲν φίλους εὖ ποιοῦντα, τοὺς δὲ ἐχθροὺς κακῶς: but the reference is only to 331 E σοφὸς γὰρ καὶ θεῖος ἀνήρ. The presents φησίν and νοεῖ are used in a general way, because such a theory and such an interpretation of it might be held by any one at any time: in οὐκ ἦν σοφὸς ὁ ταῦτα εἰπών the time is changed to the past to suggest οὐκ ἦν Σιμωνίδης ὁ ταῦτα εἰπών (Simonides being σοφός 331 E). But for ὁ ταῦτα εἰπών, ἦν would be ἐστι. It is a mistake to take ἦν as ‘is after all’: ἦν is hardly so used in Plato without ἄρα, nor is Phaedr. 230 A (cited by Goodwin MT. p. 13) an example of that idiom.

ἐάν τις αὐτὸ φῇ ‐‐ Σιμωνίδην : as Xenophon virtually does in Hier. II 2: see 331 E note

τῶν σοφῶν τε καὶ μακαρίων ἀνδρῶν . μακάριος is somewhat stronger than θεῖος, which it suggests, μάκαρες being a usual epithet of gods. The whole phrase is intended to carry us back to 331 E σοφὸς γὰρ καὶ θεῖος ἀνήρ. Ast's view that μακαρίων means “qui ante nostram aetatem floruerunt,” as if ‘sainted,’ misses the allusion to 331 E, and is a little far-fetched: it is enough that μακάριος conveys the same ironical commendation as θεῖος: cf. (with Stallbaum) Men. 71 A .

ἐγὼ γοῦν . See cr. n. With Hartman, I adopt Bekker's restoration: cf. VII 527 D. For γοῦν A everywhere writes γοὖν .

Περιάνδρου κτλ. Periander, Xerxes and Perdiccas are taken as types of tyrants, and no tyrant is σοφός ( Rep. IX 587 D). It is noticeable that Periander does not appear in the list of the seven wise men in Prot. 343 A . The expedition of Xerxes against Greece is cited by Callicles in Gorg. 483 D in connexion with the doctrine that might is right. In Περδίκκου the allusion is to Perdiccas II, father of Archelaus ( Gorg. 471 B ): he died late in 414 or early in 413, three years before the probable date of action of the Republic ( Introd. § 3), after proving himself a fickle friend and foe to the Athenians during the Peloponnesian war. Ismenias is mentioned again in Men. 90 A as having become rich δόντος τινός—<*> νῦν νεωστὶ εἰληφὼς τὰ Πολυκράτους χρήματα. There can be no doubt that he is to be identified with the Ismenias who (see Xen. Hell. III 5. 1 ) in 395 took money from Timocrates the Rhodian, envoy of the Persian King, in order to stir up war against Sparta, and who in 382, when the Spartans had seized the Cadmea, was condemned on this charge among others ( Xen. Hell. V 2. 35 ; Plut. Pelop. 5. 2). Plato implies that Ismenias kept enough Persian gold to enrich himself. he was no true Greek if he did not. But what is meant by saying that he had received the money of Polycrates? This question has been much discussed. Possibly ‘the money of Polycrates’ (with allusion, of course, to the riches of the Samian tyrant) was a sarcastic expression current in Athens for ‘the money of Timocrates’: this is perhaps the more likely as we are informed that the Athenians got no share of it themselves ( Hell. III 5. 2). Plato would naturally avail himself of such a political gibe to express his dislike of a man who took gold from the natural enemy of Greece ( Rep. V 470 C) to stir up not war, but sedition (ib. 470 B), and withdraw Agesilaus from fighting with the barbarian: for his political ideal in foreign policy was that of Cimon. See also on V 471 B. It is not however likely, I think, that the present passage was written after Ismenias' death, for Plato is not given to reviling his contemporaries after their death. That the other three persons cited by Plato were already dead would only make his reproof of the living more marked and scathing. The present passage—so far as it goes—is on the whole in favour of Teichmüller's view ( Lit. Fehd. I p. 25) that the first book of the Republic was written soon after 395, when the disgraceful affair was still fresh in men's minds. See Introd. § 4.

οἰομένου is to be pressed (as in III 395 D, 409 C: cf. IV 431 C): their power is fancied, not real: they cannot even do the thing they want: cf. Gorg. 467 A ff. πῶς ἂν οὖν οἱ ῥήτορες μέγα δύναιντο ἢ οἱ τύραννοι ἐν ταῖς πόλεσιν, ἐὰν μὴ Σωκράτης ἐξελεγχθῇ—ὅτι ποιοῦσιν ἃ βούλονται;— οὔ φημι ποιεῖν αὐτοὺς ἃ βούλονται. He alone (says Plato) is truly powerful who wills what is good and has the power to obtain it.

ὡς δὲ διεπαυσάμεθα . Cobet's suggestion ὡς δὲ δὴ ἐπαυσάμεθα misses the point. No doubt διαπαύομαι is (as he says) “intermitto orationem post aliquam moram denuo dicturus” (cf. Tim. 78 E , Symp. 191 C ), but this is precisely the sense required, for the question with which Socrates concludes ( τί ἂν ἄλλο κτλ.) shews that he desires to resume the discussion.

ταῦτ’ εἶπον refers to εἶεν—φαίη εἶναι.

συστρέψας ‐‐ διαρπασόμενος : ‘gathering himself up he sprang at us like a wild beast as though he would seize and carry us off.’ Thrasymachus comes down like a wolf on the fold. ἧκεν is not from <*>κω, but from ἵημι: this is also Ast's view (in his Lex. Plat.). The expression ἥκειν ἐφ’ ἡμᾶς would be too weak after συστρέψας ἑαυτὸν ὥσπερ θηρίον. The object to ἧκεν is ἑαυτόν, easily supplied from συστρέψας ἑαυτόν: lit. ‘he let himself go at us.’ Cf. Ar. Frogs 133 . It should be noted also that compounds of ἵημι occasionally drop ἑαυτόν altogether and become intransitive (e.g. VIII 563 A, Prot. 336 A ). Hartman's ᾖττεν for ἧκεν is not likely to find favour. For διαρπασόμενος Cobet would read διασπασόμενος. Plato however does not use διασπᾶν of harrying by wild beasts, but in the sense of disiungere, seiungere (VI 503 B, Laws 669 D): and even Cobet does not propose to change Pol. 274 B διηρπάζοντο ὑπ’ αὐτῶν (i.e. θηρίων). J. and C.'s citation of Il. XVI 355 αἶψα διαρπάζουσιν (i.e. οἱ λύκοι τὰς ἄρνας) seems to me (in spite of Hartman's wonder) strictly relevant, if only we take διαρπάζειν as ‘harry,’ and not (with J. and C.) as ‘tear in pieces.’

τίεὐηθίζεσθε κτλ. εὐηθίζεσθε refers to the readiness of the interlocutors to assent to one another's questions: cf. Charm. 175 C οὕτως ἡμῶν εὐηθικῶν τυχοῦσα ἡ σκέψις καὶ οὐ σκληρῶν.

ὑποκατακλινόμενοι : a metaphor, not from the wrestling schools, but from taking a lower or inferior seat at table or the like: cf. Symp. 222 E ἐὰν οὖν ὑπὸ σοὶ κατακλινῇ Ἀγαθών and Plut. quomodo adul. ab amico internoscatur 58 D τὰς τοιαύτας ὑποκατακλίσεις (alluding to men who take the front seats at theatres etc., in order to flatter the rich by giving up their seats to them). Thrasymachus' brutal frankness is not intended by Plato to be altogether wide of the mark: see App. II and 335 A note

μὴ ‐‐ φιλοτιμοῦ ἐλέγχων . A common reproach against Socrates: cf. Theae<*>. 150 C.

ὅπως μοι κτλ. This idiom is colloquial and abrupt, almost rude: cf. 337 B and the examples cited in Goodwin MT. p. 94. Thrasymachus will not tolerate the stale and barren platitudes— note ὕθλους below—of ordinary disputation: cf. [ Clitoph.] 409 C οὗτος μὲν—τὸ συμφέρον ἀπεκρίνατο, ἄλλος δὲ τὸ δέον, ἕτερος δὲ τὸ ὠφέλιμον, ὁ δὲ τὸ λυσιτελοῦν and Stewart's Nicomachean Ethics Vol. I p. 16, with the references there quoted.

εἰ μὴ πρότερος ‐‐ γενέσθαι . The θηρίον of 336 B has become a wolf. This is the earliest allusion in Greek literature to the belief that if a wolf sees you first you become dumb. Like Virgil Ecl. IX 53 the present passage favours Schaefer's emendation Αύκος εἶδέ ς); for Αύκον εἶδες in Theocr. XIV 22.

μὴ χαλεπὸς ἡμῖν ἴσθι : ἀεὶ θρασύμαχος εἶ, said Herodicus on one occasion to the sophist (Arist. Rhet. II 23. 1400^{b} 20).

ἐξαμαρτάνομεν ‐‐ ἁμαρτάνομεν : the preposition is often dropped in repeating a compound verb: cf. V 452 A, VII 533 A, X 608 A and my note on Prot. 311 A . I can see no sufficient reason for inserting τι before ἐξαμαρτάνομεν (with Π and some other MSS), although Stallbaum and others approve of the addition.

μὴ γὰρ δὴ οἴου κτλ. Cf. Laws 931 C, where there is a similar a fortiori sentence couched in the imperatival form.

ἑκόντας εἶναι . This phrase is used sixteen times by Plato, always in negative clauses, and generally in the nominative or accusative (Grünenwald in Schanz's Beiträge zur hist. Synt. d. gr. Spr. II 3. 1 ff.).

οἴου γε σύ, ὦ φίλε : i.e. ἡμᾶς σπουδάζειν ὅ τι μάλιστα φανῆναι αὐτό. For the justification of this view see App. IV.

εἰκός ἐστιν . There is no reason for omitting ἐστίν (with Hartman and apparently also Usener Unser Platotext p. 40).

χαλεπαίνεσθαι . This strained use of the passive of χαλεπαίνω in order to make the antithesis to ἐλεεῖσθαι formal as well as real is not found elsewhere in Plato. For parallels see Cope's Rhetoric of Aristotle Vol. 1 p. 299.

σαρδάνιον . Plato uses this expression as Homer does, of a sinister smile which bodes pain to others: Od. XX 301 f. μείδησε δὲ θυμῷ | σαρδάνιον μάλα τοῖον (of Odysseus among the suitors). Among later authors it more frequently denotes the forced smile which disguises the sufferer's own pain; and so apparently Simonides used the phrase ( Fr. 202 A Bergk). The explanations volunteered by the ancients apply only to the non-Homeric usage: the Scholiast, however, at the end of his note on this passage correctly remarks, μήποτε οὖν τὸ Ὁμηρικόν, ὅθεν καὶ ἡ παροιμία ἴσως ἐρρύη, “ μείδησε δὲ κτλ.,” τὸν ἀπ’ αὐτῶν τῶν χειλῶν γέλωτα καὶ μέχρι τοῦ σεσηρέναι γιγνόμενον σημαίνει. The spelling σαρδόνιον came into vogue through the popular etymology from the bitter Sardinian herb, ἧς οἱ γευσάμενοι δοκοῦσι μὲν γελῶντες, σπασμῷ δὲ ἀποθνῄσκουσιν (Schol.). The Scholiast's suggested derivation from σαίρειν ( ringi, as of an angry dog) suits the meaning which the phrase bears in Homer and Plato, and is probably right. Photius' σαρδάζων· μετὰ πικρίας γελῶν preserves the δ .

ποιήσοις is rejected by Cobet and Herwerden. “Post οὐδὲν ἄλλο ἤ, τί ἄλλο ἤ, πάντα μᾶλλον ἤ verbum omittunt” (says Cobet, quoting Theophr. Char. c. 25). ποιήσοις is not however otiose, but suggests the phrase πάντα ποιεῖν, ‘leave nothing undone,’ as in Euthyph. 8 C πάντα ποιοῦσι καὶ λέγουσι φεύγοντες τὴν δίκην: cf. Ap. 39 A .

ἐρωτᾷ . I formerly read ἐρωτῷ (with Goodwin MT. p. 277). A few inferior MSS have ἔροιτο. The optative is certainly the regular periodic construction in clauses of this kind: but the indicative may perhaps be allowed in loose conversational style.

τυγχάνει ὄν . Stallbaum explains ὄν as ‘being true,’ and τι as the subject to τυγχάνει. This view is perhaps less natural than to make ὄν the copula and τι the predicate: for the pronoun ‘it’ i.e. τὸ ἐρωτώμενον (Schneider) can be quite easily understood. For the use of τυγχάνει ὄν (‘really is’) cf. II 379 A, VII 518 E, Euthyph. 4 E with my note ad loc.

ὡς δή . The force of ὡς in this common ironical expression ( quasi vero, cf. Gorg. 468 E , 499 B) is referred by Jebb ( Soph. O. C. 809 ) to an ellipse: ‘(do you mean) forsooth that.’ An objection to this theory is that it will not explain ὡς δή τοι in cases like II 366 C, Phaedr. 242 C , Tim. 26 B . It seems better to explain these usages on the same principle. The view that ὡς is exclamatory will not account for II 366 C, and is not specially appropriate in the other places. Neither is it easy to make ὡς= ἐπεί (‘your illustration is excellent, seeing that the cases are so very similar!’ Tucker). Schneider (on II 366 C) regards ὡς as nearly equivalent to ὥστε (cf. note on II 365 D). Probably ὡς is in reality consequential (like the English ‘so’), the relative retaining its original demonstrative sense. This explanation will, I believe, suit all the passages in question.

περὶ δικαιοσύνης κτλ. περὶ δικαιοσύνης and τούτων are rejected by Herwerden, but the fulness of expression suits the arrogant tone of Thrasymachus.

τί ἀξιοῖς παθεῖν ; Here and in what follows there is a play on the judicial formula παθεῖν ἢ ἀποτεῖσαι, where παθεῖν refers to δεσμός φυγή θάνατος ἀτιμία, and ἀποτεῖσαι to fines. In a δίκη τιμητός, the defendant if found guilty would be asked in the words τί ἀξιοῖς παθεῖν καὶ ἀποτεῖσαι to propose an alternative penalty to that demanded by the accuser; after which it was the duty of the judges finally to assess ( τιμᾶν) the penalty: cf. Ap. 36 B and Laws 933 D. It is partly the paronomasia in the words παθεῖν μαθεῖν (cf. the ancient text πάθος μάθος Aesch. Ag. 176 ) which draws from Thrasymachus the mock compliment ἡδὺς γὰρ εἶ (‘you are vastly entertaining’) although (cf. ἆ <*>διστε 348 C) Thrasymachus is also jeering at the simplicity of Socrates.

πρὸς τῷ μαθεῖν καὶ ἀπότεισον . Hertz and Herwerden conjecture παθεῖν for μαθεῖν: but this would make Thrasymachus ignore Socrates' identification of παθεῖν with μαθεῖν. In ἀπότεισον ἀργύριον Plato no doubt satirizes (somewhat crudely, it must be allowed) the avarice of Thrasymachus and his class, in contrast with whom Socrates has no money, because his conversations are gratis.

εἰσοίσομεν . The metaphor is from a banquet to which each contributes his share: cf. Symp. 177 C ἐγὼ οὖν ἐπιθυμῶ ἅμα μὲν τούτῳ ἔρανὀν εἰσενεγκεῖν κτλ.

ἀπειρημένον αὐτῷ . See cr. n. The retention of εἴη after αὐτ ῷ can only be defended by regarding μὴ εἰδὼς μηδὲ φάσκων as equivalent to εἰ μὴ εἰδείη μηδὲ φάσκοι and carrying on the εἰ; but this is excessively harsh and no parallel has yet been adduced. Of the two alternatives, to insert an εἰ before ἀπειρημένον or εἴη, and to drop εἴη (with Bremius), I prefer the latter as simpler in itself and accounting more easily for the corruption. The accusative absolute may have been misunderstood and εἴη inserted by a negligent reader owing to εἰ in the previous line. Richter (in Fl. Jahrb. 1867 p. 137) inserts δ) before αὐτῷ and retains εἴη, regarding εἴ τι καὶ οἴεται and ἀπειρημένον δ’ αὐτῷ εἴη as coordinate clauses under the rule of the same εἰ; but to this there are many objections. Tucker's suggestion εἰ, ὅ τι καὶ οἴεται περὶ τούτων, ἀπειρημένον αὐτῷ εἴη κτλ. (‘if, in regard to whatever he thinks about them, it were forbidden’ etc.) strikes me as heavy and cumbrous.

μὴ οὖν ἄλλως ποίει : 328 B note

ἄκουε δή calls for attention, ostentatiously, like a herald: cf. X 595 C, Ap. 20 D , Prot. 353 C .

Πουλυδάμας ‐‐ ὁ παγκρατιαστής . οὗτος ὁ Πουλυδάμας ἀπὸ Σκοτούσσης ἦν, πόλεως θεσσαλίας, διασημότατος παγκρατιαστής, ὑπερμεγέθης, says the Scholiast. He was victor in the ninety-third Olympian games 408 B.C. Stallbaum refers to Pausanias (VI 5) and others for the wonderful stories of his prowess. His statue at Olympia by Lysippus was very famous. Cf. Boeckh Kl. Schr. IV p. 446.

τοῦτο τὸ σιτίον κτλ. Teichmüller ( Lit. Fehd. II p. 196) finds in this a confirmation of his belief that Plato was a vegetarian: but it is implied merely that a beef diet was not considered wholesome for persons out of training. Aristotle may have had this passage in view in Eth. Nic. II 5. 1106^{a} 36 ff., though his illustration is there taken from quantity, and not from quality, of food. Cf. also Gorg. 490 C .

ξυμφέρον ἅμα καὶ δίκαιον . The sophistry is undisguised. If βόεια κρέα is Polydamas' συμφέρον and δίκαιον, and δίκαιον is assumed to be everywhere identical with itself, it follows that βόεια κρέα is our δίκαιον, but not our ξυμφέρον, otherwise we are also κρείττονες. To avoid this, Wohlrab ingeniously takes ἐκείνου not with ἥττοσιν but with ξυμφέρον ἅμα καὶ δίκαιον, as if the meaning were ‘Polydamas' συμφέρον καὶ δίκαιον is also δίκαιον for us.’ This explanation is however linguistically harsh and comparatively pointless. On βδελυρὸς γὰρ εἶ Tucker aptly reminds us that the prevailing feature in Theophrastus' description of the βδελυρός ( Char. c. 11) is παιδιὰ ἐπιφανὴς καὶ ἐπονείδιστος (‘obtrusive and objectionable pleasantry’ Jebb).

κακουργήσαις . Cope observes that the word is used “of the knavish tricks and fallacies which may be employed in rhetorical and dialectical reasoning” ( Aristotle's Rhetoric Vol. 1 p. 17). Cf. Gorg. 483 A (cited by Tucker).

εἶτ’ οὐκ οἶσθα κτλ. ‘Do you mean to say you don't know’ etc. The division of constitutions into Monarchy, Oligarchy (for which Aristocracy is here substituted) and Democracy was familiar to everybody: see Aeschin. Ctes. 6, Tim. 4 ὁμολογοῦνται γὰρ τρεῖς εἶναι πολιτεῖαι παρὰ πᾶσιν ἀνθρώποις, τυραννὶς καὶ ὀλιγαρχία καὶ δημοκρατία. Cf. Whibley Greek Oligarchies pp. 17, 24. Thrasymachus proceeds to define κρείττων as ὁ κρατῶν (not ὁ ἰσχυρότερος, as Socrates had insinuated): ‐κρατοῦνται in δημοκρατοῦνται and ἀριστοκ ρατοῦνται well brings out his meaning. Cf. Laws 714 B νόμων εἴδη τινές φασιν εἶναι τοσαῦτα ὅσαπερ πολιτειῶν, and C οὔτε γὰρ πρὸς τὸν πόλεμον οὔτε πρὸς ἀρετὴν ὅλην βλέπειν δεῖν φασι τοὺς νόμους, ἀλλ) ἥτις ἂν καθεστηκυῖα ᾖ πολιτεία, ταύτῃ δεῖν τὸ ξυμφέρον ὅπως ἄρξει τε ἀεὶ καὶ μὴ καταλυθήσεται, καὶ τὸν φύσει ὅρον τοῦ δικαίου λέγεσθαι κάλλισθ) οὕτως. Πῶς; Ὅτι τὸ τοῦ κρείττονος ξυμφέρον ἐστί .

τίθεται δέ γε : Laws l. c. τίθεται δήπου, φασί, τοὺς νόμους ἐν τῇ πόλει ἑκάστοτε τὸ κρατοῦν. ἦ γάρ; Ἀληθῆ λέγεις. Ἆῤ οὖν οἴει, φασί, ποτὲ δῆμον νικήσαντα ἤ τινα πολιτείαν ἄλλην ἢ καὶ τύραννον θήσεσθαι ἑκόντα πρὸς ἄλλο τι πρῶτον νόμους <*> τὸ συμφέρον ἑαυτῷ τῆς ἀρχῆς τοῦ μένειν; Πῶς γὰρ ἄν; Aristotle makes it the distinguishing mark of his three perverted forms ( παρεκβάσεις) of constitution ( τυραννίς, ὀλιγαρχία, δημοκρατία) that they seek their own and not τὸ κοινῇ συμφέρον: Pol. Γ 7. 1279^{b} 4 ff.

τὸν τούτου ἐκβαίνοντα κτλ. Laws 714 D οὐκοῦν καὶ ὄς ἂν ταῦτα τὰ τεθέντα παραβαίνῃ, κολάσει ὁ θέμενος ὡς ἀδικοῦντα, δίκαια ταῦτ) εἶναι ἐπονομάζων; Ἔοικε γοῦν. Ταῦτ) ἄῤ ἀεὶ καὶ οὕτω καὶ ταύτῃ τὸ δίκαιον ἂν ἔχοι. Φησὶ γοῦν οὗτος ὁ λόγος. νόμος and δίκαιον are identified by this theory.

ταὐτὸν εἶναι δίκαιον . Herwerden would expunge ταὐτόν, but ταὐτὸν is not more otiose here than τὸ αὐτὸ below.

ξυμφέρον γέ τι . There is here a hint of the main purpose of the Republic, which is to prove that δίκαιον is ξυμφέρον in the truest sense for the individual and the state.

οὐ ‐‐ μέντοι . “In interrogationibus haec particula” ( μέντοι) “ita cum οὐ negatione coniungitur, ut gravissima sententiae vox intercedat, quo modo aliquis eis quae ex altero quaerit summam veritatis ingerit speciem” (Hoefer de part. Plat. p. 34). μέντοι is simply ‘of course,’ ‘surely’: ‘surely you regard it as just to obey the rulers, do you not?’ The idiom is frequent in Plato. The other examples of it (cited by Stallbaum) in the Republic are infra 346 A, VII 521 D, IX 581 A, 584 A, X 596 E.

πότερον δὲ ἀναμάρτητοι κτλ. The reasoning echoes that of 334 C above.

τιθέναι ‐‐ τίθεσθαι : we should expect τιθέναι in both cases, as the ἄρχοντες according to the theory we are discussing are κρείττονες and supreme as legislators: but the middle of personal interest is naturally used in combination with τὰ ξυμφέροντα ἑαυτοῖς: cf. infra 341 A.

τί λέγεις σύ ; a favourite eristic formula: see Ar. Clouds 1174 τοῦτο τοὐπιχώριον | ἀτεχνῶς ἐπανθεῖ, τὸ τί λέγεις σύ;

τοίνυν : not ‘therefore,’ but ‘also,’ a frequent use in Plato. In the Republic it occurs 29 times, according to Kugler de particulae τοι eiusque comp. ap. Pl. usu p. 34.

ὅταν οἱ μὲν ‐‐ τοῖς δέ (i.e. τοῖς ἀρχομένοις). These two clauses depend, not on ὡμολογῆσθαι, but on ποιεῖν: it is just to do τὰ ἀξύμφορα τοῖς ἄρχουσιν as often as the rulers unwillingly prescribe what is evil for themselves and so long as Thrasymachus says it is just for subjects to do what the rulers have prescribed. Desire for brevity and balance leads Plato to put both clauses under the government of ὅταν, although ‘since’ rather than ‘whenever’ is the more appropriate conjunction for introducing the second: for Thrasymachus does not sometimes but always assert that it is just to obey the rulers. The suggested reading φῄς for φῇς would require us to take τοῖς δέ κτλ. as an independent sentence, and leave μὲν in οἱ μέν without a corresponding δέ.

αὐτό is ‘the matter,’ ‘the case before us’: cf. IV 428 A ( αὐτῷ), VII 518 B ( αὐτῶν), 524 E ( αὐτῷ ), Theaet. 172 E al. The text has been needlessly suspected by Madvig and other critics.

οὑτωσί : not ‘in that case’ (Campbell), but (with Jowett) simply ‘thus,’ as explained in δίκαιον—λέγεις: cf. Ap. 26 E οὑτωσί σοι δοκῶ; οὐδένα νομίζω θεὸν εἶναι;

ναὶ μὰ Δία κτλ. The interlude is intended to mark that the first stage has been reached in the refutation of Thrasymachus.

ἐὰν σύ γε is of course ironical. The disciples of the rival disputants now enter the fray.

τὸ γὰρ τὰ κελευόμενα κτλ. If this, and no more, had been Thrasymachus' definition, it would remain unrefuted; commands would be commands, whether expedient for the rulers or not.

δ ἡγοῖτο ‐‐ ξυμφέρειν . This explanation is involved in Clitophon's earlier statement τὸ τὰ κελευόμενα ποιεῖν ὑπὸ τῶν ἀρχόντων: that which the rulers κελεύουσι is what they believe to be in their interests. Clitophon's defence finds no justification in the terms of Thrasymachus' definition; but it was the most obvious way of attempting to reconcile that definition with the admission that rulers are capable of erring.

τὸ τοῦ κρείττονος ξυμφέρον κτλ. Bonitz ( Zeitschr. f. öst. Gymn. 1865 p. 648), followed by Wohlrab, proposes to add the words τὸ ξυμφέρον after ξυμφέρον, “parum venuste,” as Hartman thinks. Neither is it well (with Hartman) to omit τοῦ κρείττονος. The apparent harshness of the construction (‘that which seems to be the stronger's interest to the stronger’) is justified by its brevity and precision, and by the desire to introduce the exact words of the original definition into its amended form.

συκοφάντης . Cf. (with Tucker) Arist. Soph. El. 15. 174^{b} 9 σοφιστικὸν συκοφάντημα τῶν ἐρωτώντων and Rhet. II 24. 1402^{a} 14 ἐπὶ τῶν ἐριστικῶν τὸ κατὰ τί καὶ πρὸς τί καὶ πῇ οὐ προστιθέμενα ποιεῖ τὴν συκοφαντίαν.

λέγομεν τῷ ῥήματι οὕτως . Bekker (with whom Shilleto on Dem. F. L. § 91 agrees) would insert μέν after λέγομεν: but (as Schneider remarks) the emphasis on τῷ ῥήματι does duty instead of the particle, and even otherwise, μέν is not essential: cf. III 398 A (where Shilleto would also add μέν), infra 343 C, II 363 E, X 605 C al.

τὸ δέ =‘whereas in point of fact’ is a favourite Platonic idiom: cf. IV 443 C, VII 527 A, 527 D al.

ἐπιλιπούσης . See cr. n. The present, which Stallbaum and others adopt, may be right, but the older reading is at least as good. The failure in knowledge must precede the actual error. For the mistake see Introd. § 5.

ἀποκρίνεσθαι . The imperfect infinitive, as Schneider remarks ( Addit. p. 6).

συκοφαντεῖν is explained in ἐξ ἐπιβουλῆς—κακουργοῦντά σε, where κακουργοῦντα (as Schneider observes) is not used as in 338 D of putting an evil or sophistical interpretation on a theory, but of damaging a man's personal reputation and credit: “scilicet existimationis et pecuniae detrimentum facturus sibi videbatur sophista ideoque Socratem se, quamquam frustra, impugnare in sequentibus quoque criminatur.”

μὴ λαθών : “si non latueris” (Schneider). Stephanus conjectured μήν and Ast με for μή: but either change would destroy the antithesis between λανθάνειν and βιάζεσθαι—secret guile and open fraud: cf. II 365 D ἀλλὰ δὴ θεοὺς οὔτε λανθάνειν, οὔτε βιάσασθαι δυνατόν. Hirschig's excision of μὴ λαθών greatly impairs the emphasis.

τὸν ὡς ἔπος εἰπεῖν . The only exact parallel to this use of ὡς ἔπος εἰπεῖν in Plato is Laws 656 E σκοπῶν δ’ εὑρήσεις αὐτόθι τὰ μυριοστὸν ἔτος γεγραμμένα ἢ τετυπωμένα οὐχ ὡς ἔπος εἰπεῖν μυριοστὸν ἀλλ’ ὄντως. This idiomatic phrase is rare before Plato, who uses it 77 times with the meaning ‘to put the matter in a word,’ implying that other and possibly more exact means of describing the thing in question might be found. In 52 of these cases the phrase is combined with πᾶς or οὐδείς and their family of words, in the sense of fere, propemodum: its use in other connexions is in part a return to old poetic usage; cf. Aesch. Pers. 714 , Eur. Hipp. 1162 , Heracl. 167. See Grünenwald in Schanz's Beiträge zur hist. Synt. d. gr. Spr. II 3, pp. 21 ff. The other examples in the Republic are V 464 D, VIII 551 B, IX 577 C.

ὃ νῦν δὴ ἔλεγες : viz. in 340 E κατὰ τὸν ἀκριβῆ λόγον. The antecedent is the phrase ἀκριβεῖ λόγῳ. The conjecture of Benedictus, ὅν for ὅ, though adopted by several editors, would (as Schneider remarks) leave it uncertain whether λόγῳ or τόν was referred to by the relative. There is no MS authority for ὅν.

ξυρεῖν ‐‐ λέοντα . παροιμία ἐπὶ τῶν καθ) ἑαυτῶν τι ἢ ἀδύνατα ποιεῖν ἐπιχειρούντων λεγομένη (Schol.). The proverb is very rare, and does not seem to occur elsewhere in classical Greek.

οὐδὲν ὢν καὶ ταῦτα : ‘though you were a nonentity at that too’: i.e. at bluffing me, as well as in other respects. So (I think) Schneider, rightly (“aber auch darin ist's nichts mit dir”). Others (e.g. Shorey in A. J. Ph. XVI p. 234) explain ‘and that too though you are a thing of naught.’ But in that case καὶ ταῦτα would surely precede οὐδὲν ὤν. Tucker can hardly be right in making καὶ ταῦτα simply ‘moreover’ ‘too,’ ‘and proved a failure, too.’ Nor (in spite of J. B. Mayor in Cl. Rev. X p. 110) is it quite enough to translate (with Campbell) ‘though here again you are nobody,’ i.e. ‘with as little effect as ever.’

καὶ λέγε ‐‐ ὄντα is expunged by Herwerden, but the emphatic reiteration is in keeping with the whole tone of the passage. For the sense we may recall the words of the so-called oath of Hippocrates εἰς οἰκίας δὲ ὁκόσας ἂν ἐσίω, ἐσελεύσομαι ἐπ’ ὠφελείῃ τῶν καμνόντων (Vol. 1. p. 2 ed. Kühn).

ἑκάστῳ τούτων : viz. τοῖς κάμνουσι, τοῖς ναύταις, and in general the sub jects upon whom the art is exercised. The expression is a little vague (cf. VIII 543 C note) but it is rash and unnecessary to insert εἴδει or write ἑκάστῳ <τῶν εἰδῶν> τούτων, as Tucker recommends.

ἆῤ οὖν ‐‐ τελέαν εἶναι . I have retained this reading, in deference to the MSS, but it is open to grave objection. As the sentence stands, the meaning is that every art (as well as every object of an art—this is implied by καί) has one συμφέρον, viz. to be as perfect as it can, but no other. In the sequel this is interpreted to mean that no art needs any additional ἀρετή; since it is ( qua art) perfect already: οὔτε γὰρ πονηρία οὔτε ἁμαρτία οὐδεμία οὐδεμιᾷ τέχνῃ πάρεστιν κτλ. (342 B). But the words of the sentence ἆρ’ οὖν—τελέαν εἶναι have to be taken very loosely in order to admit of this interpretation. We must suppose them equivalent to ‘No art has a συμφέρον of its own, unless you are to call the fact that it is perfect its συμφέρον.’ If Plato had written the passage as it stands in q and in the margin of Flor. U (both MSS probably of the fifteenth century), it would be open to no objection: ἆρ’ οὖν καὶ ἑκάστῃ τῶν τεχνῶν ἔστι τι ξυμφέρον ἄλλο <οὗ προσδεῖται>, ἢ <ἐξαρκεῖ ἑκάστη αὐτὴ αὑτῇ, ὥστε> ὅ τι μάλιστα τελέαν εἶναι; This reading was adopted by Bekker, and by Stallbaum in his first edition; and a careful study of the whole passage confirms the judgment of Schneider, “Platonem non solum potuisse, sed etiam debuisse vel haec ipsa vel consimilia scribere.” The same sense, expressed more briefly, may be obtained by the insertion of δεῖ before εἶναι: ‘has every art also a ξυμφέρον besides (i.e. besides the ξυμφέρον of its object), or must it be as perfect as possible?’ <*>κάστην does not require to be repeated any more than in 346 A below. The alteration is very slight; for δεῖ εἶναι, δεῖναι may have been written by mistake and δ afterwards ejected.

νῦν κτλ. : ‘has now been invented.’ The art of medicine is not coeval with body. I can see no reason for thinking (with Campbell) that νῦν is corrupt for ἡμῖν.

σῶμά ἐστι πονηρὸν κτλ. Lys. 217 B ἀναγκάζεται δέ γε σῶμα διὰ νόσον ἰατρικὴν ἀσπάζεσθαι καὶ φιλεῖν.

ὅπως ἐκπορίζῃ . This is said by Weber (Schanz's Beiträge II 2, p. 67) to be the only example in Plato of ὅπως with the subjunctive after a preterite tense.

ἐπ’ αὐτοῖς . Hartman proposes ἔτ’ αὐτοῖς. αὐτοῖς (sc. ὀφθαλμοῖς, ὠσίν) may be emphatic (ipsis), and ἐπί, ‘over and above,’ ‘besides’: ‘we require in addition to the organs themselves, an art’ etc. But it is perhaps simpler to make ἐπί=‘ to preside over’: cf. ἐφ’ οἷς ἔστιν VI 511 E note

εἰς ταῦτα means εἰς ὄψιν καὶ ἀκοήν. The art in question considers what is advantageous with respect to ( εἰς) seeing and hearing.

ἐκποριούσης . See cr. n. ἐκποριούσης appears in three Florentine MSS. The present is difficult, if not impossible, in so close a union with the future: cf. X 604 A and VI 494 D. See Introd. § 5.

ἢ αὐτὴ ‐‐ σκέψεται ; This question (which is of course to be answered in the negative) shews the awkwardness of the reading of A in ἆρ’ οὖν—τελέαν εἶναι (341 D), which might almost be construed to mean that each art does seek its own συμφέρον, viz. the perfection of itself.

ἕωσπερ κτλ. : ‘that is, so long as an art, taken in its strict sense’ (“streng genommen” Schneider) ‘preserves its essence entire and unimpaired.’ The predicate is ὅλη ἥπερ ἐστί, and ἀκριβής = ἀκριβὴς οὖσα in the sense which ἀκριβής bears throughout this passage (341 B al.). Hartman's insertion of ἡ before ἀκριβής is unsatisfactory; his alternative proposal to change ἀκριβής to ἀκριβῶς spoils the emphasis, and gives a wrong sense.

ἐπιστήμη is here a synonym for τέχνη. All arts rule: and ruling is itself an art or science, not a happy inspiration (cf. Mem. III 6). Like other arts, ruling seeks only the good of that which it rules.

ὡμολόγηται γὰρ ‐‐ χρηματιστής . Ast compares Arist. Pol. A 9. 1258^{a} 10 ff. ἀνδρίας γὰρ οὐ χρήματα ποιεῖν ἐστιν ἀλλὰ θάρσος, οὐδὲ στρατηγικἧς καὶ ἰατρικῆς, ἀλλὰ τῆς μὲν νίκην, τῆς δ’ ὑγίειαν.

ᾧ ἂν ‐‐ βλέπων . ᾧ is of course ( τῷ) ὃ, and τῷ ἀρχομένῳ is also neuter (not masculine), like ἀρχομένου in D. Bremius took τῷ ἀρχομένῳ as masculine, and consequently changed (with inferior MS authority) πρὸς ἐκεῖνο into πρὸς ἐκεῖνον: he has been followed by Stallbaum and others. But as ᾧ must be neuter, it would be intolerable to make ἀρχομένῳ masculine, since both words (as denoting the same object) are covered by the same article, viz. τῷ before ἀρχομένῳ. ἐκείνῳ is of course neuter also.

εἰς τοὐναντίον . Justice has now become τὸ τοῦ ἥττονος (rather than κρείττονος) συμφέρον.

κορυζῶντα : ‘snivelling,’ μωραίνοντα, μυξάζοντα· κόρυζα γὰρ ἡ μύξα, ἣν οἱ Ἀττικοὶ κατάρρουν φασίν (Schol.). Ruhnken on Timaeus Lex. s.v. quotes among other passages Lucian Alex. § 20 ἦν δὲ τὸ μηχάνημα τοῦτο ἀνδρὶ μὲν οἵῳ σοι, εἰ δὲ μὴ φορτικὸν εἰπεῖν, καὶ οἵῳ ἐμοί. πρόδηλον καὶ γνῶναι ῥᾴδιον, τοῖς δὲ ἰδιώταις καὶ κορύζης μεστοῖς τὴν ῥῖνα τεράστιον καὶ πάνυ ἀπίστῳ ὅμοιον, and Horace Sat. I 4. 8 (of Lucilius) emunctae naris.

ὅς γε αὐτῇ κτλ. “Apte αὐτῇ interpositum; nam ipsi nutrici Socratis insipientiam opprobrio esse, Thrasymachus vult significare” Ast. Richter ( Fl. Jahrb. for 1867 p. 140) ought not to have suggested ὅς γε αὐτός. The sense is ‘for she cannot teach you to recognise even sheep or shepherd,’ not ‘you do not know either sheep or shepherd’ (J. and C.), which would require οὔτε— οὔτε. The phrase is clearly a half-proverbial expression borrowed from the nursery.

ὅτι οἴει τοὺς ποιμένας κτλ. Thrasymachus gives a new turn to the nursery saying. The illustration from the shepherd and his sheep (which is now for the first time introduced) was used by the historical Socrates to justify the opposite conclusion ( Xen. Mem. III 2. 1 ) ἐντυχὼν δέ ποτε στρατηγεῖν ᾑρημένῳ τῳ, Τοῦ ἕνεκεν, ἔφη, Ὅμηρον οἴει τὸν Ἀγαμέμνονα προσαγορεῦσαι ποιμένα λαῶν; ἆρά γε ὅτι, ὥσπερ τὸν ποιμένα ἐπιμελεῖσθαι δεῖ, ὅπως σῶαί τε ἔσονται αἱ ὄϊες, καὶ τὰ ἐπιτήδεια ἕξουσιν, οὕτω καὶ τὸν στρατηγὸν ἐπιμελεῖσθαι δεῖ, ὅπως σῶοί τε οἱ στρατιῶται ἔσονται, καὶ τὰ ἐπιτήδεια ἕξουσι, καὶ οὗ ἕνεκα στρατεύονται τοῦτο ἔσται; So also Arist. Eth. Nic. VIII 13. 1161^{a} 12 ff. εὖ γὰρ ποιεῖ τοὺς βασιλευομένους, εἴπερ ἀγαθὸς ὢν ἐπιμελεῖται αὐτῶν, ἵν’ εὖ πράττωσιν, ὥσπερ νομεὺς προβάτων· ὅθεν καὶ Ὅμηρος τὸν Ἀγαμέμνονα ποιμένα λαῶν εἶπεν. In Plato Pol. 271 D ff. the deities of the golden age are compared to shepherds, and the comparison of a good ruler to a shepherd is very frequent in Plato: see Ast's Lex. Plat. s. v. νομεύς. In Socrates' view ‘the shepherd careth for his sheep.’ With Thrasymachus' attitude should be compared the picture of the tyrant in Theaet. 174 D as a συβώτην ἢ ποιμένα ἤ τινα βουκόλον—πολὺ βδάλλοντα (he squeezes as much milk as he can out of his flock): also Solon ap. Arist. Rep. Ath. ch. 12 εἰ γάρ τις ἄλλος ταύτης τῆς τιμῆς ἔτυχεν, οὐκ ἂν κατέσχε δῆμον οὐδ’ ἐπαύσατο, | πρὶν ἀνταράξας πῖαρ ἐξεῖλεν γάλα. In the word ἀμοργοί or ἀμολγοί used by Cratinus in the sense of πόλεως ὄλεθροι (Meineke Fr. Com. Graec. II 1, p. 140) the image is the same. Compare the eloquent words of Ruskin in Sesame and Lilies § 43 and Milton's Lycidas 113—129.

ἡγεῖ διανοεῖσθαι . The conjecture διακεῖσθαι for διανοεῖσθαι is tempting in view of διατεθείη which follows, but διανοεῖσθαι is better suited to σκοπεῖν and βλέποντας just above. For the somewhat rare construction Schneider compares Laws 626 D αὐτῷ δὲ πρὸς αὑτὸν πότερον ὡς πολεμίῳ πρὸς πολέμιον διανοητέον, ἢ πῶς ἔτι λέγομεν; and 628 D.

πόρρω εἶ περί . πόρρω can hardly (I think) mean ‘far from’ (sc. knowing): this would require πόρρω εἶ <τοῦ τι εἰδέναι> περί, as Herwerden suggests: cf. Lys. 212 A οὕτω πόρρω εἰμὶ τοῦ κτήματος ὥστε κτλ. The meaning is (I believe) ‘so far on’; ‘so profoundly versed are you in justice’ etc.: cf. πόρρω ἤδη ἐστὶ τοῦ βίου Ap. 38 C and phrases like πόρρω σοφίας ἐλαύνειν: see also Blaydes on Ar. Wasps 192 . Such biting sarcasm is appropriate in the mouth of Thrasymachus.

ἀλλότριον ἀγαθόν . Arist. Eth. Nic. V 3. 1130^{a} 3 f. διὰ δὲ τὸ αὐτὸ τοῦτο καὶ ἀλλότριον ἀγαθὸν δοκεῖ εἶναι ἡ δικαιοσύνη μόνη τῶν ἀρετῶν, ὅτι πρὸς ἕτερόν ἐστιν· ἄλλῳ γὰρ τὰ συμφέροντα πράττει, ἢ ἄρχοντι ἢ κοινωνῷ (with Stewart's note) and ib. 10. 1134^{b} 5.

τῷ ὄντι is not τῷ ὄντι δικαιῳ, but revera (as Stallbaum observes).

ὡς ἀληθῶς as well as ἀληθῶς, τῷ ὄντι, and the like, is used to indicate that a word is to be taken in its strict and full etymological sense ( εὐ‐ηθικῶν): cf. Phaed. 80 D εἰς Ἅιδου ὡς ἀληθῶς, and infra II 376 B, V 474 A, VI 511 B, VIII 551 E notes

τά γε οἰκεῖα ‐‐ μοχθηροτέρως . Wells aptly cites the refusal of Deioces in Herod. I 97 to continue as an arbiter: οὐ γάρ οἱ λυσιτελέειν τῶν ἑαυτοῦ ἐξημεληκότα τοῖσι πέλας δἰ ἡμέρης δικάζειν. Cf. also Ap. 23 B , 31 B. In like manner Aristotle mentions it as one of the safeguards of a democracy engaged in agriculture that the necessity of looking after their private interests will prevent the citizens from often attending the assembly ( Pol. Z 4. 1318^{b} 11). Plato is fond of the comparative ending in ‐ως (affected, says Cobet, by those “qui nitidissime scribunt”): see Kühner-Blass Gr. Gramm. I p. 577.

ἀπεχθέσθαι . ἀπέχθομαι as a present is not well attested in Plato's time; and the aorist ‘to incur the enmity of’ is at least as suitable in point of meaning here.

λέγω γὰρ ὅνπερ νῦν δὴ ἔλεγον . Ast points out that nothing in what has been already said corresponds to the words τὸν μεγάλα δυνάμενον πλεονεκτεῖν, and reads ὅπερ on slight MS authority. But no special reference is intended: the words mean simply ‘I mean the man I meant just now.’ Thrasymachus asserts that he has all along been referring to τὸν μεγάλα κτλ.

ἢ τὸ δίκαιον : i.e. ἢ τὸ δίκαιον εἶναι τῷ δικαίῳ. The reading αὑτῷ (found in A, but no dependence can be put on this MS in such matters) would require the omission of the article before δίκαιον (so Stallbaum and others). Tucker inclines to render ‘how much more he is personally benefited by being unjust than by justice,’ but the ordinary view is preferable.

ἣ οὐ κτλ. This laboured sentence is perhaps intended as a parody of some sophistic style: cf. Gorg. 448 C .

ὧν depends on

μέρει .

ἀνδραποδισταί : ‘kidnappers.’ The word is defined by Pollux III 78 as ὁ τὸν ἐλεύθερον καταδουλωσάμενος ἢ τὸν ἀλλότριον οἰκέτην ἀπαγόμενος. Thessaly had an evil name for this kind of crime (Blaydes on Ar. Plut. 521 ); but the frequent references to it in Attic literature shew that Greece itself was not exempt. See on IX 575 B and the article in Stephanus-Hase Thes. s.v.

τῶν τοιούτων κακουργημάτων is usually explained as depending on κατὰ μέρη, but as κατὰ μέρη is adverbial, this is somewhat awkward. It is perhaps better to regard the genitive as partitive, τι being omitted as in κινήσειεν ἂν τῶν ἀξίων λόγου νόμων IV 445 E, where see note.

πρὸς τοῖς ‐‐ χρήμασιν is virtually equivalent to πρὸς τῷ τὰ τῶν πολιτῶν ἀφελέσθαι, and combined by zeugma with δουλώσηται. Cf. I 330 E note

εὐδαίμονες ‐‐ κέκληνται . The generic singular τις has become a plural, as in Phaed. 109 D , infra VII 536 A. Envy of tyranny and tyrants was common in the Athens of Plato's younger days: compare Gorg. 484 A , 470 D (where it is maintained by Polus that Archelaus of Macedon is εὐδαίμων, and Socrates says ὀλίγου σοι πάντες συμφήσουσι ταῦτα Ἀθηναῖοι καὶ οἱ ξένοι 472 A) and Alc. II 141 A ff. The plays of Euripides in particular (see VIII 568 A) often eulogised the tyrant: e.g. Troad. 1169 ff., Fr. 252, Phoen. 524 ff. In earlier days Solon's friends had blamed him for not making himself tyrant of Athens: see the dramatic fragment (33 ed. Bergk), where the prevalent passion for tyranny is forcibly expressed in the lines ἤθελον γάρ κεν κρατήσας, πλοῦτον ἄφθονον λαβὼν | καὶ τυραννεύσας Ἀθηνῶν μοῦνον ἡμέραν μίαν, | ἀσκὸς ὕστερον δεδάρθαι κἀπιτετρῖφθαι γένος (4—6). See also Newman's Politics of Aristotle I pp. 388—392.

οὐ γὰρ ‐‐ τὴν ἀδικίαν . Cf. Gorg. 483 A φύσει μὲν γὰρ πᾶν αἴσχιόν ἐστιν, ὅπερ καὶ κάκιον, τὸ ἀδικεῖσθαι, νόμῳ δὲ τὸ ἀδικεῖν.

ἱκανῶς γιγνομένη : ‘realised on an adequate scale’ (D. and V.). For the construction of γίγνεσθαι with an adverb cf. (with Ast) Soph. 230 C and infra VI 504 C. After τὸ δ’ ἄδικον below, Herwerden would insert τό to go with ἑαυτῷ λυσιτελοῦν τε καὶ ξυμφέρον, but only τυγχάνει (and not τυγχάνει ὄν) is to be understood after ἄδικον; nor is the last clause intended as a strict and formal definition of injustice.

καταντλήσας . For the metaphor cf. infra VII 536 B, Lys. 204 D , Lucian Dem. Enc. 16 (imitated from this passage) and other examples in Blaydes on Ar. Wasps 483 .

ἐμβαλών : cf. Theaet. 165 D , Prot. 342 E . The whole expression recalls the Latin proverb scrupulum abeunti (Cic. de Fin. IV 80).

διαγόμενος . The use of this verb in Soph. El. 782 χρόνος διῆγέ με, Dem. 18. 89 πόλεμος—διῆγεν ὑμᾶς, Xen. Rep. Lac. I 3 and elsewhere is in favour of regarding διαγόμενος (‘living’) as grammatically passive and not middle both here and in Laws 758 A. Cf. Stephanus-Hase Thes. s.v. διάγω.

ἐγὼ γὰρ κτλ. I agree with Stallbaum and others in taking this sentence as interrogative: ‘do you mean that I think otherwise about this matter?’ i.e. think that it is not a question of βίου διαγωγή. J. and C. complain that this interpretation is “wanting in point.” It is surely much to the point to make Thrasymachus repudiate the imputation of trifling. His doctrine appears all the more dangerous when he confesses that it is no sophistic paradox, but a rule of life. I can see nothing to justify Apelt's conjecture ἔγωγ’ ἄῤ for ἐγὼ γὰρ ( Observ. Crit. p. 11).

ἤτοι ἡμῶν γε . ἤτοι or ἤτοι—γε= ‘or else’ (not ‘or rather’ as J. and C.). The regular construction is ἤτοι—ἤ, and ἤ—ἤτοι was condemned by the grammarians as a solecism, though it occurs in Pind. Nem. 6. 5. With the use of ἤτοι in this passage cf. III 400 C, IV 433 A τοῦτό ἐστιν—ἤτοι τούτου τι εἶδος ἡ δικαιοσύνη. Emendations have been suggested on all these passages of Plato: here ἤ τοι (van Prinsterer, Hartman) and in the other two passages ἤ: but we are not justified in altering the text. Cf. Kugler de partic. τοι eiusque comp. ap. Pl. usu p. 14.

ἔστω μὲν ἄδικος κτλ. The subject is ὁ ἄδικος, supplied from ἀδικίαν. To πείθει also ὁ ἄδικος is the subject; but ἡ ἀδικία or τὸ ἀδικεῖν is the subject of ἔστι. The effect is exactly as in the English ‘let him be unjust’ etc., ‘nevertheless he cannot convince me that it is really more profitable than justice.’ J. and C. understand τις before ἔστω, needlessly, as I think, and suppose that the “supposed impunity of injustice” is the subject to πείθει, but πείθει is much better with a personal subject. Although the sentence is a trifle loose, it is clear enough, and there is no occasion for reading πείθεις (with Vind. D and Ficinus).

ἐνθῶ . ἐντιθέναι (as Wohlrab points out) was used of nurses feeding children: cf. Ar. Knights 716 f., supra 343 A, and (for the general idea) Theognis 435 and Pl. Symp. 175 D . In μὰ Δία, μὴ σύ γε Socrates shudders at the prospect of having Thrasymachus for his intellectual nurse.

ποιμαίνειν . See cr. n. Cobet ( Mnem. IX p. 355) calls for πιαίνειν, but the “addita verba καθ’ ὅσον ποιμήν ἐστιν circa universum pastoris negotium errantem a Socrate Thrasymachum notari docent” (Schneider). How Thrasymachus errs is explained in οὐ πρὸς τὸ κτλ. πιαίνειν might perhaps be read, if the ἀλλά clause is taken closely with what precedes: you did not think it necessary (says Socrates) to adhere rigidly to the genuine shepherd, but think he fattens his sheep qua shepherd. In that case, however, we should expect ἀλλ’ οὐ—βλέπειν in place of οὐ—βλέποντα, to form the antithesis to πιαίνειν .

οὕτω δὲ ᾤμην . Some inferior MSS (with Eusebius Praep. Ev. XII 44. 2) read δή for δέ, and so Ast and Stallbaum. The connecting particle is better than the illative here, where Socrates is merely recalling his former train of reasoning: ‘and it was thus that I came to think’ etc.

οὐδεὶς ‐‐ μισθόν . Cf. Arist. Eth. Nic. V 10. 1134^{b} 5 ff. καὶ διὰ τοῦτο ἀλλότριον εἶναί φασιν α<*>γαθὸν τὴν δικαιοσύνην—μισθὸς ἄρα τις δοτέος.

αὐτοῖσιν : see 330 B note

οὐχὶ ‐‐ μέντοι : 339 B note

παρὰ δόξαν is simply ‘contrary to your opinion’ (“gegen deine Ueberzeugung” Schneider) as in Prot. 337 B , cf. 349 A ἕως ἄν σε ὑπολαμβάνω λέγειν ἅπερ διανοεῖ and 350 E. The words could hardly mean an ‘unexpected or paradoxical’ reply (as Tucker construes). Socrates is appealing—note ὦ μακάριε— to Thrasymachus not to obstruct the discovery of the truth by want of candour and sincerity.

ἀλλὰ ‐‐ ἑτέρα sc. ἐστίν. The reading ἑτέραν is in itself equally good, but has inferior MS authority. Herwerden needlessly recommends the omission of ἑτέρα, or (as alternatives) ἀλλὰ τούτῳ, ἔφη, τῷ ἑτέραν, or ἀλλὰ τούτῳ, ἔφη, ἑτέραν, τῷ ἑτέραν .

οὐκοῦν κτλ. Aristotle agrees with this analysis: see Pol. A 3. 1258^{a} 10 ff. It should be noted that the antecedent to αὕτη is not μισθόν, but τὸ παρέχεσθαι μισθόν.

διὰ τὸ ξυμφέρειν . See cr. n. To ξυμφέρον there are two objections: first that διά with the participle used like διά with the infinitive is rare and dubious; second that ξυμφέρον is more naturally to be taken as a virtual adjective than as a participle. The last objection might be surmounted by reading ξυμφέρον<ὄν>, but the more serious flaw would still remain, and ξυμφέρειν is in itself so much superior, that (like most editors) I feel bound to adopt it.

οὐδέ γ’ οἶμαι τὴν μισθωτικήν sc. καλεῖς ἰατρικήν. The reasoning is somewhat subtle. ἰατρική, κυβερνητική, μισθωτική ( μισθαρνητική), says Socrates, are three distinct arts. κυβερνητική is not to be called ἰατρική, even if ἰατρική should accompany its operation, nor is μισθωτική to be called ἰατρική in a similar case. Nor is ἰατρική to be called μισθωτική, even if ἰατρική should be accompanied by μισθωτική.

κοινῇ ‐‐ προσχρώμενοι : ‘from the common use of some additional element which is the same in all.’

τὸ μισθὸν ἀρνυμένους . δέ γε as usual introduces the minor premise. The semi-poetic word ἀρνυμένους is used to suggest μισθαρνεῖν and μισθαρνητική, the word μισθόν at the same time bringing the product of the art well into view. As τὸ—δημιουργούς is the subject to γίγνεσθαι, the masculine τόν—so most MSS—for τό is impossible.

ἔλεγον μηδένα ἐθέλειν . μή with the infinitive after verbs of saying, thinking and the like “carries with it the emphasis of the witness on oath, so to speak the emphasis of desire” (Gildersleeve in A. J. Ph. 1 50). Cf. Theaet. 155 A , Euthyph. 6 B , Phaed. 94 C al., and infr. III 407 E, IV 419 A.

ὡς ἔοικε belongs to

ὧν ἕνεκα , and δεῖν is in indirect narration after ἔλεγον above. There would be no object in qualifying the force of δεῖν; it is not disputed that rulers must have their reward. Hence Stallbaum is wrong in regarding δεῖν as under the influence of ἔοικε, an illogical idiom which is common in Herodotus (Stein on 1 65), and found occasionally in Tragedy (Jebb on Trach. 1238) and in Plato ( Phil. 20 D , Soph. 263 D , Euthyd. 280 D ). That ὡς ἔοικε has no influence on δεῖν in this passage may also be seen from the fact that δεῖν (not δεῖ) would still be used if ὡς ἔοικε were removed. δεῖν is not for δέον; the late participial form δεῖν is not found in Plato: see my note on Euthyph. 4 D .

ἄρχῃ . The transition from plural to singular and conversely is common: see for examples III 408 B, 411 C, 413 D, E, IV 426 A, C, V 463 D, VI 496 C, 500 C, VIII 554 A, C, 558 A, IX 591 A, X 601 D, E, 604 D, and cf. Heindorf on Gorg. 478 C , Prot. 319 D .

ὡς ἐν μισθοῦ μέρει . ὡς is not (with Wohlrab) to be taken with <*>ν μισθοῦ μέρει, but stands for the indirect interrogative ὅπως.

αὐτοί = ‘by themselves,’ ‘ultro,’ should be construed with λαμβάνοντες. The conjecture αὐτόν for αὐτοί is very tame.

ὅθεν κινδυνεύει ‐‐ νενομίσθαι . These words are intended to indicate parenthetically that Socrates' thesis finds support in the common judgment of men. Good men, he says, require to be compelled to rule. This may be why ( ὅθεν) it is accounted a disgrace to enter on office willingly: that is to say, if you do so, you may be inferred to be, not ἀγαθός, but φιλότιμος or φιλάργυρος, which ὄνειδος λέγεταί τε καὶ ἔστιν 347 B. There is no good reason for rejecting the clause, as some have proposed to do.

πόλις ἀνδρῶν ἀγαθῶν is the first express allusion to an Ideal City in the Republic. The principle here laid down—the reluctance of the best men to undertake the task of government—is fully recognised in Plato's commonwealth, where the ἄρχοντες are represented as unwilling to desert the life of contemplation for the cares of office. ‘Nolo episcopari’ is in fact one of the leading guarantees which Plato gives against the abuse of political power (Nohle Die Staatslehre Plato's in ihr. gesch. Entwick. p. 119). See VI 520 E, 521 A, where this topic is resumed. Cf. also Sesame and Lilies § 43 “The true kings—rule quietly, if at all, and hate ruling; too many of them make ‘il gran rifiuto.’”

τῷ ὄντι κτλ . τῷ ὄντι belongs to οὐ πέφυκε, not to ἀληθινός (as Ast supposes). Richter suggests ἁληθινός for ἀληθινός, but what is said of a single ruler applies to all: cf. (with Schneider) Laws 733 E λέγωμεν δὴ σώφρονα βίον ἕνα εἶναι καὶ φρόνιμον ἕνα καὶ ἕνα τὸν ἀνδρεῖον.

πᾶς ἂν κτλ. The articular infinitive with αἱρεῖσθαι is hard to parallel, and on this ground Richards would cancel τό. I once thought that τὸ ὠφελεῖσθαι might be taken as the object after γιγνώσκων (‘he who knows what being benefited is,’ i.e. virtually ‘who knows his own interests’); but this is harsh, and I now acquiesce in the usual interpretation. With γιγνώσκων ( intellegens) used absolutely cf. (with Schneider) Laws 733 E σώφρονα μὲν οὖν βίον ὁ γιγνώσκων θήσει πρᾶον ἐπὶ πάντα. For the sentiment cf. Soph. O. T. 584 —598, Eur. Ion 621 —632, Hipp. 1016—1020.

εἰσαῦθις σκεψόμεθα . The reference has been much discussed. Pfleiderer's idea ( Zur Lösung d. Pl. Fr. p. 72) that the words were introduced by Plato “bei der Gesammtredaktion des Werkes” to prepare us for the second half of Book X is most unlikely, because (among other reasons) Book X does not expressly revert to this topic at all. Siebeck ( Zur Chron. d. Pl. Dialoge pp. 121 ff.) holds that phrases of this sort always refer either to some future dialogue contemplated by Plato, or to a later part of the same dialogue. It is difficult to establish either alternative in the present case; nor has Siebeck, I think, succeeded in proving his point even elsewhere. It is simplest to suppose that such formulae (like εἰσαῦθις ἐπισκεπτέον in Arist. Eth. Nic. I 5. 1097^{b} 14) are in general only a convenient way of dropping the subject, although there may occasionally be a specific reference. Here there is none. So also Hirmer Entst. u. Komp. d. Pl. Polit. in Fl. Jahrb. Supplementband XXIII p. 607 note 2.

τὸν τοῦ ἀδίκου βίον ‐‐ δικαίου . In these words Socrates sums up the remarks of Thrasymachus from 343 B ( καὶ οὕτω πόρρω κτλ.) to 344 C ( λυσιτελοῦν τε καὶ ξυμφέρον).

ποτέρως ‐‐ λέγεσθαι . Ast's suggestion πότερον, ἦν δ) ἐγώ, ὦ Γλαύκων, αἱρεῖ; καὶ ποτέρως ἀληθεστέρως δοκεῖ σοι λέγεσθαι; is now generally adopted, but (apart from its considerable divergence from the MS reading) the juxtaposition of ποτέρως and ἀληθεστέρως is unpleasing. The ποτέρως αἱρεῖ of A is quite unobjectionable: cf. VII 528 A οὕτως—αἱροῦμαι; and it is (I think) an objection to πότερον αἱρεῖ that it would represent Socrates as asking Glauco not which view he elected to take, but which life—the just or the unjust—he chose for himself. Schneider (after Bekker) retains the reading of the best MSS in πότερον ὡς ἀληθεστέρως, and explains the last two words as equivalent to ὥσπερ ὃ ἀληθεστέρως λέγεται: but ὡς ἀληθεστέρως could not (if written by Plato) be anything but the comparative of ὡς ἀληθῶς, and that is quite different in sense from ἀληθεστέρως. I have omitted ὡς (with Bremius and a few MSS of inferior authority), “ut ortum ex varia lectione πότερον et ποτέρως in ποτερόνως conflata” (Schneider). I am glad to find that Tucker adopts the same solution.

διῆλθε : i.q. διῆλθεν ὄντα or διελθὼν ἔλεξεν εἶναι (Schneider). Cf. II 363 A ἄφθονα ἔχουσι λέγειν ἀγαθὰ τοῖς ὁσίοις with note ad loc. In view of ἐν ἑκατέρῳ λέγομεν in B below, it is easy to suggest διῆλθεν <ἐν>; but the text is probably sound.

ἂν μὲν τοίνυν κτλ. The alternatives are between continuous speech and dialectic. By λόγον in παρὰ λόγον Thrasymachus' speech in 343 A ff. is meant: to this Socrates would reply, after which Thrasymachus would speak again, and finally Socrates. Thus each party would have delivered two speeches. In Athenian lawsuits there were often two speeches delivered by the accuser and two by the defendant (Meier und Schömann Attische Process p. 924), so that Plato's imagery is borrowed from the law-court, whence δικαστῶν τινῶν τῶν διακρινούντων just below. This point escaped Ast, who reads καὶ αὖθις οὗτος ἄλλον ἡμῖν (after Ficinus and Stephanus).

ἀντικατατείναντες is intransitive: cf. II 358 D κατατείνας ἐρῶ τὸν ἄδικον βίον ἐπαινῶν and 367 B: the notion (as in ξυντείνω, ξυντεταμένως and the like) is of nervous tension. The word cannot mean ‘replying to one another in set speeches’ (J. and C.). “Setting out alternative lists of advantages” (remarks Bosanquet) “was the well-known method of fable or poetry. See Book II” 361 D—362 C and 362 E— 365 A: “and compare Prodicus' Choice of Heracles (Xen. Mem. II 1) and the discussion between the Just and Unjust arguments in the Clouds of Aristophanes.”

ὁποτέρως is virtually indirect: translate ‘whichever you please, then.’ Hermann reads ποτέρως, but the text ought not to be changed either here or in Euthyd. 271 A τίς ἦν, ὦ Σώκρατες, ᾧ χθὲς ἐν Λυκείῳ διελέγου;—τίς ἦν; Ὁπότερον καὶ <*>ρωτᾷς, ὦ Κρίτων· οὐ γὰρ εἶς, ἀλλὰ δὔ ἤστην, i.e. (it depends on) which of these you are asking about etc. Cf. also <*>τις—αὐτῶν ἡ ἀρετή 353 C. In Rep. IX 578 E ἐν ποίῳ ἄν τινι καὶ ὁπόσῳ φόβῳ οἴει γενέσθαι αὐτόν and Gorg. 522 A , the ὁπόσῳ is perhaps due to the proximity of οἴει, which gives the question a certain semblance of indirectness; ὁποίῳ in Alc. I 110 C and ὁποίου infra 400 A may be similarly explained; while in Meno 74 D ἀλλὰ μή μοι οὕτως—ἀλλ’ ὅ τι ἐστὶν τοῦτο, it is casy to supply a verb of saying. Possibly (as Heindorf thinks) ὅτι (B ὃ τί) in Euthyd. 287 B is corrupt for τί, as <*>πως for πῶς in Charm. 170 C . In Lys. 212 C ὁπότερος οὖν αὐτῶν ποτέρου φίλου ἐστιν; ὁ φιλῶν τοῦ φιλουμένου—ἢ ὁ φιλούμενος τοῦ φιλοῦντος; we ought no doubt to read ὁ πότερος (with Hermann).

εἰκός γε ‐‐ τοὐναντίον . Thrasymachus' view of δικαιοσύνη is like Callicles' theory of ἀρετή in Gorg. 491 E ff. esp. 492 B τρυφὴ καὶ ἀκολασία καὶ ἐλευθερία, ἐὰν ἐπικουρίαν ἔχῃ, τοῦτ’ ἐστὶν ἀρετή τε καὶ εὐδαιμονία. The irony is clearly marked by ὦ ἥδιστε, and Hartman should not have revived Hirschig's proposal to read <οὔκουν> εἰκός γε .

ἀλλὰ τί μήν ; ‘Well, what else?’ Cf. (with J. and C.) Symp. 206 E .

πάνυ γενναίαν εὐήθειαν : ‘sublime simplicity.’ Such contempt for εὐήθεια recalls Thucydides' description of contemporary morals: cf. especially III 83. 1 καὶ τὸ εὔηθες, οὖ τὸ γενναῖον πλεῖστον μετέχει, καταγελασθὲν ἠφανίσθη.

εὐβουλία was preeminently a political virtue: cf. Alc. I 125 E πολιτείας κοινωνούντων τίνα καλεῖς ἐπιστήμην; Εὐβουλίαν ἔγωγε, Prot. 318 E , and infra IV 428 B. It is therefore fitly used by Thrasymachus to describe his theory, which is a theory of political rather than of private morality: cf. πόλεις τε—ποιεῖσθαι below.

ὑφ) ἑαυτοὺς ποιεῖσθαι . ἑαυτοῖς is found in some inferior MSS, but the accusative is also admissible. Cf. Thuc. IV 60 (cited by Schneider) εἰκὸς—αὐτοὺς τάδε πάντα πειράσασθαι ὑπὸ σφᾶς ποιεῖσθαι. In τελέως Thrasymachus recalls the τελέαν ἀδικίαν of 348 B.

σὺ δὲ οἴει ‐‐ λέγειν . Baiter (with Paris A) assigns these words to Socrates; but they come much more naturally from Thrasymachus: cf. 344 B. βαλλάντια and not βαλάντια is the spelling of A here and in VIII 552 D ( βαλλαντιοτόμοι): in IX 575 B ( βαλλαντιοτομοῦσι) the second λ is due to an early corrector. The double ‐λλ‐ has also the best MS authority in Gorg. 508 E , Symp. 190 E . See also Blaydes on Ar. Frogs 772 . For

ἦ δ’ ὅς below after ἔφη cf. Phaed. 78 A and VII 522 A.

ἤδη στερεώτερον : ‘still more stubborn.’ στερέος is like σκληρός in Theaet. 155 E σκληροὺς—καὶ ἀντιτύπους ἀνθρώπους, but stronger, suggesting castiron hardness and inflexibility.

ῥᾴδιον . See cr. n. Schneider refers to Laws 757 B τὴν δὲ ἀληθεστάτην καὶ ἀρίστην ἰσότητα οὐκέτι ῥᾴδιον παντὶ ἰδεῖν. ῥᾷον is not (I think) possible here: and a scribe might easily omit ΙΔ in ΡΑΙΔΙΟΝ. Cf. Introd. § 5.

εἰ γὰρ κτλ. Gorg. 483 C νόμῳ μὲν τοῦτο ἄδικον καὶ αἰσχρὸν λέγεται, τὸ πλέον ζητεῖν ἔχειν τῶν πολλῶν, καὶ ἀδικεῖν αὐτὸ καλοῦσιν. Dümmler ( Zur Comp. d. Pl. St. p. 13) goes so far as to assert that ὥσπερ ἄλλοι τινές is an express reference to Polus in the Gorgias; but nothing is gained by so hazardous a conjecture.

προσετίθεμεν : ‘used to attribute to,’ sc. before you announced your view—with ironical deference, like ἐλέγομεν in Prot. 353 C τί οὖν φατε τοῦτο εἶναι, ὃ ἡμεῖς <*>ττω ε<*>ναι τῶν ἡδονῶν ἐλέγομεν; Stallbaum takes the imperfect as referring to 345 C, but neither there nor in 348 C (cited by Schneider) is there anything to justify a particular reference.

ἐμοὶ γὰρ ‐‐ λέγειν . A similar remark is made after Callicles has expounded kindred views in Gorg. 492 D σαφῶς γὰρ σὺ νῦν λέγεις ἃ οἱ ἄλλοι διανοοῦνται μέν, λέγειν δὲ οὐκ ἐθέλουσι.

τὰ δοκοῦντα κτλ. can only mean ‘what you think about the truth,’ not ‘what you think to be the truth’ (D. and V.) or ‘your real mind’ (Schneider and Jowett). We should expect ἀδικίας for ἀληθείας, as H. Wolf proposed to read, for it is Injustice, not Truth, which is the subject of dispute. But as ἀδικίας has not a vestige of support from the MSS, I have not ventured to make the change. The truth in question must be understood as the truth about justice and injustice. Herwerden's ἐπὶ τῆς ἀληθείας (for which he compares Dem. de Cor. 17, 226, and 294) will hardly command assent.

τί δέ ‐‐ ἐλέγχεις ; Cf. Charm. 161 C πάντως γὰρ οὐ τοῦτο σκεπτέον ὅστις αὐτὸ εἶπεν, ἀλλὰ πότερον ἀληθὲς λέγεται ἢ οὔ.

πλέον ἔχειν . The literal and derived significations of this phrase are treated as identical throughout the curious reasoning which follows. Primarily, πλέον ἔχειν refers to quantitative superiority; in its derived sense, it is used (together with πλεονεκτεῖν) more generally of ‘overreaching.’

τῆς δικαίας πράξεως . ‘To have more than the just action’ means ‘to do more than is just’ (cf. πλείω—αἱρεῖσθαι— πράττειν 350 A), outdo, overreach what is just in action. The notion of virtue as a μεσότης is implied.

οὐδὲ τῆς κτλ. See cr. n. I do not think that οὐδὲ τῆς δικαίας can be right. The whole emphasis (as οὐδὲ shews) must be on πράξεως, and the emphatic word should be expressed. οὐδὲ τῆς πράξεως τῆς δικαίας (sc. any more than the ἀνδρὸς δικαίου) gives exactly the emphasis required. In the cases quoted by Schneider ( Laws 754 B, 916 B, infra VII 516 B) the omitted word is unemphatic and easily supplied. For the error cf. Crito 50 B where the first hand of the Bodleian MS reads τὰς δικασθείσας by mistake for τὰς <δίκας τὰς> δικασθείσας. See also Introd. § 5.

οὐκοῦν καὶ ‐‐ λάβῃ . The ἄδικος πρᾶξις which the unjust man overreaches is to be regarded as ἄδικος because it has itself overreached (not fallen short of) the mean. ὡς with the subjunctive after verbs of striving does not seem to occur elsewhere in Plato: like its use in a pure final clause (of which there is only one example in Plato, viz. Tim. 92 A ) it is almost exclusively confined (among Attic writers) to Xenophon and the tragedians. See Weber's tables in Goodwin MT. p. 398, and cf. Gildersleeve in A. J. Ph. IV p. 419.

τοῦ μὲν ὁμοίου ‐‐ τοῦ δὲ ἀνομοίου . This generalisation of ‘like’ and ‘unlike’ into abstract notions, without regard to their relativity, is suggestive of (but does not of course presuppose) the Ideas of τὰ πρός τι which we meet with in Phaed. 74 A .

οὐκοῦν κτλ. A proviso which is made use of in 350 C ( ἀλλὰ μὴν —ἑκάτερον εἷναι ).

ὁ δὲ μὴ ἐοικέναι . ὁ δέ is simply ‘the other’ (as is marked in A by a pause after δέ), i.e. ὁ μὴ τοιοῦτος: cf. 339 E ( τοῖς δέ for τοῖς δὲ ἀρχομένοις), 343 D ὁ μὲν δίκαιος ἀπὸ τῶν ἴσων πλέον εἰσφέρει, ὁ δ’ ἔλαττον and IX 587 B. J. and C., with most of the editors, adopt the reading of Stephanus ( ὁ δὲ μὴ μὴ ἐοικέναι), which has the support of some inferior MSS; but the idiom is sufficiently well authenticated, and the collocation of the two negatives would be unpleasing. I am glad to see that Tucker takes the same view.

οἷσπερ ἔοικεν . Madvig's οἵοισπερ ἔοικεν is refuted by 350 C ὡμολογοῦμεν ᾧ γε (i.e. οἷος ᾧ γε) ὅμοιος ἑκάτερος εἴη, τοιοῦτον καὶ ἑκάτερον εἶναι. Cf. also Arist. Pol. H 13. 1332^{a} 22. The construction was supported by Schneider from Phaed. 92 B , but ὅ and not ᾧ is now read there on the authority of the best MS.

ἀλλὰ τί μέλλει (sc. εἶναι); A rare formula, occurring also in Hipp. Min. 377 D: cf. τί δ’ οὐ μέλλει; VIII 566 D, X 605 C. With the force of τί (‘what else’) cf. ἀλλὰ τί οἴει supra 332 C.

μουσικὸν δέ τινα κτλ. Here begin the usual Socratic illustrations from the arts, with the concomitant identification of virtue and knowledge ( ὁ δὲ σοφὸς ἀγαθός; Φημί 350 B).

δοκεῖ ἂν οὖν ‐‐ ἀξιοῦν πλέον ἔχειν . Socrates ignores the proverb καὶ κεραμεὺς κεραμεῖ κοτέει καὶ ἀοιδὸς ἀοιδῷ. Strictly speaking, however, it is not qua κεραμεύς, but qua moneymaker (or the like) that the κεραμεὺς κοτέει. J. and C. cite an admirable parallel from Shakespeare ( King John IV 2) “When workmen strive to do better than well, They do confound their skill in covetousness.” The words ἢ ἀξιοῦν πλέον ἔχειν have a suspicious look, and are rejected by Heller ( Fl. Jahrb. 1875 p. 171) and others, but such duplicate expressions are common in Plato, and as the illustration from the harp introduces a new and important stage in the argument, Plato may have wished to remind us that after all πλεονεκτεῖν is only the πλέον ἔχειν with which we started (349 B). It should be noted, too, that ἀξιοῦν is a little more than ἐθέλειν.

ἐν τῇ ἐδωδῇ ἢ πόσει refers of course to the patient's diet. Plato carefully writes πλεονεκτεῖν here in preference to πλέον ἔχειν. The ‘overreaching’ in such a case might well consist in giving the patient less.

ἢ πράττειν ἢ λέγειν . The idea of πλεονεκτεῖν in speaking has not been introduced before, nor is it made use of in the sequel. We must regard the addition of ἢ λέγειν as merely a rhetorical device to increase the emphasis: see on 333 D and 351 A.

τί δὲ ὁ ἀνεπιστήμων; κτλ. Proclus' commentary on these words is interesting, though he probably reads more into them than Plato intended here: καὶ ὅλως τῷ μὲν ἀγαθῷ τὸ κακὸν ἠναντίωται μόνον, τῷ δὲ κακῷ καὶ τὸ καλὸν (leg. κακὸν) καὶ τὸ ἀγαθόν· ἀναιρετικὸν οὖν ἐστι τοῦ ἀγαθοῦ καὶ τοῦ πρὸς αὐτὸ ἐναντίου κακοῦ ( in Alc. I p. 323 ed. Creuzer). The identifications in ὁ δὲ ἐπιστήμων σοφός and ὁ δὲ σοφὸς ἀγαθός below have been allowed before in the special cases of the μουσικός and the ἰατρικός (349 E).

ἀναπέφανται . Stallbaum naïvely reminds us that ἀναπέφανται is often used of a conclusion which “praeter exspectationem emergit et elucet.” The pervading fallacy in the discussion is akin to the a dicto secundum quid ad dictum simpliciter. Thus ‘like’ and ‘unlike’ are used absolutely, and each of them is equated with itself. The wise man is held to be good, because one is good in that in which one is wise (this might however be justified on the “stricter mode of reasoning”). Finally, the just man is inferred to be wise and good, on the principle that one is what one resembles: but whether the resemblance be in essence or in accident, we are not told. The argument should be regarded as a dialectical tour de force,— φιλόνικον μᾶλλον ἢ φιλάληθες. The reasoning in the next section of the argument strikes a deeper note.

ὁ δὲ Θρασύμαχος κτλ. ‘Now Thrasymachus’ etc. δέ is not “flat” (Tucker), but at least as good as δή, and much better supported by the MSS.

οὐχ ὡς ἐγὼ νῦν ῥᾳδίως λέγω . “Expectabam certe: οὐχ ὡς ἐγὼ νῦν λέγω ῥᾳδίως,” says Herwerden; but the antecedent in Greek is idiomatically attracted into the relative clause (Kühner Gr. Gramm. II p. 922). Translate ‘not in the easy way in which I now repeat them.’

ἅτε καὶ θέρους ὄντος . The action is probably laid in Hecatombaeon (roughly our July): see Introd. § 3. Bekker (following the punctuation of A) takes τότε with ὄντος, but πρότερον δὲ οὔπω shews that it belongs to καὶ εἶδον.

τότε καὶ is simply ‘then too’; I cannot see anything “mock-heroic” in the expression, as J. and C. do.

εἰ οὖν λέγοιμι κτλ. εἰ δ) οὖν is read by Ast: “sed sufficit externum, ut ita dicam. vinculum οὖν (Schneider).” δημηγορεῖν and εἰπεῖν ὅσα βούλομαι are the opposites of διαλέγεσθαι and βραχυλογία ( Prot. 336 B , 335 A).

ὥσπερ ταῖς γραυσί . Cf. Gorg. 527 A τάχα δ’ οὖν ταῦτα μῦθός σοι δοκεῖ λέγεσθαι, ὥσπερ γραός, καὶ καταφρονεῖς αὐτῶν: Pol. 268 E ἀλλὰ δὴ τῷ μύθῳ μου πάνυ πρόσεχε τὸν νοῦν, καθάπερ οἱ παῖδες. παῖς for ταῖς was read before Ast on the authority of one MS; but ταῖς is quite satisfactory.

ὅπερ ἄρτι . The words ἔφαμεν δὲ δὴ καὶ ἰσχυρὸν εἶναι τὴν ἀδικίαν· ἢ οὐ μέμνησαι; (350 D), which are referred to in ἄρτι, involve the general question of the relation between justice and injustice; whence we have ὁποῖόν τι τυγχάνει ὂν δικαιοσύνη πρὸς ἀδικίαν. ὁποῖον depends on ἐρωτῶ, not on λόγον.

ἐλέχθη γάρ που : 344 C, 348 E. It has nowhere been expressly said that Injustice is δυνατώτερον than Justice, but καὶ δυνατώτερον is added for emphasis (see on ἢ λέγειν in 350 A); and indeed according to the theory of Thrasymachus δύναμις (power in a general sense) rests solely on ἰσχύς (physical strength). δύναμις and ἰσχύς are clearly distinguished in Prot. 351 A .

ἁπλῶς . The Platonic use of ἁπλοῦν has been investigated by Bonitz in Hermes II (1867) pp. 307 ff. Its antitheses are διπλοῦν, διάφορον, σύνθετον, πεπλεγμένον, ποικίλον, and the like, and it denotes that which is uniform, or single and simple, or true without any difference or qualifications. ἁπλῶς οὕτως means merely ‘in this simple or general way’ (“im Allgemeinen” Schneider): a more elaborate and profounder proof (thinks Socrates) is necessary.

καὶ καταδεδουλῶσθαι is rejected by Cobet, but successfully defended by Heller ( Fl. Jahrb. 1875 p. 172). There is in reality no pleonasm: we have first an attempt ( ἐπιχειρεῖν), then a successful attempt ( καταδεδουλῶσθαι), then the results of success ( πολλὰς δὲ καὶ ὑφ’ ἑαυτῇ ἔχειν δουλωσαμένην). A powerful city like Athens might, and often did, display her energy in all three directions simultaneously. For the collocation of δουλοῦσθαι and καταδουλοῦσθαι (middle) Heller compares infra IX 589 D, E and Menex. 240 A .

ἡ ἀρίστη . Thrasymachus refuses to withdraw from the position that ἀδικία is ἀρετή, in spite of Socrates' refutation. This is why Socrates says μανθάνω κτλ. ‘I understand: (you say so) because this was your theory.’ ὅτι is not ‘that’: see above on 332 A. Richter suggested κρατίστη for ἀρίστη on account of κρείττων just below; but κρείττων is said not by Thrasymachus, but by Socrates.

εἰ ‐‐ ἔχει . After ἔχει, εἰ is inserted by Stallbaum, following a suggestion of Baiter's. Cf. also J. B. Mayor in Cl. Rev. X p. 111. It so happens that ἡ is written in A over an erasure large enough to have contained εἰ ἡ, but there is no trace of εἰ, and mere erasures in A are seldom useful in determining the text. For ἡ Richter suggests ᾖ, which would however give a wrong meaning. Tucker also offers a variety of conjectures, but the text is perfectly sound: cf. II 359 B μάλιστ’ ἂν αἰσθοίμεθα, εἰ τοιόνδε ποιήσαιμεν τῇ διανοίᾳ· δόντες (i.e. εἰ δόντες ) ἐξουσίαν —εἶτ’ ἐπακολουθήσαιμεν κτλ. and IX 589 D εἴπερ τοιόνδε τι γίγνεται, λαμβάνων (i.e. εἰ λαμβάνων )— καταδουλοῦται.

ἢ λῃστὰς κτλ. Cf. (with Ast) Isocrates Panath. 226 οὐδεὶς ἂν αὐτοὺς ( τοὺς Σπαρτιάτας) διά γε τὴν ὁμόνοιαν δικαίως ἐπαινέσειεν, οὐδὲν μᾶλλον ἢ τοὺς καταποντιστὰς καὶ λῃστὰς καὶ τοὺς περὶ τὰς ἄλλας ἀδικίας ὄντας· καὶ γὰρ ἐκεῖνοι σφίσιν αὐτοῖς ὁμονοοῦντες τοὺς ἄλλους ἀπολλύουσιν. There must be some honour even among thieves.

ὁμόνοιαν καὶ φιλίαν . The conception of δικαιοσύνη which meets us in Book IV 433 A—434 E is dimly outlined here.

ἐν ἐλευθέροις κτλ. : ‘whether it makes its appearance among freemen or among slaves.’ Plato wishes to emphasize the universality of the rule, and that is why he specifies the two classes into which society is divided. Cf. Gorg. 514 D , 515 A. It is less natural and easy to construe (with Tucker) ‘in a society where there are both freemen and slaves.’

ἀλλήλοις τε καὶ τοῖς δικαίοις . So in 349 C above it is said that the unjust try to overreach both one another and the just.

ἐν ἑνὶ κτλ. The results of Book IV are foreshadowed more clearly in what follows. The notion that justice present in the individual keeps the individual at peace with himself is more fully developed in 441 D, and implicitly assumes a psychological theory like that in Book IV, where soul is shewn to have ‘parts’ (435 C ff.). Further, in Book IV, Plato first describes justice in the State, and afterwards justice in the individual, using the larger aggregate to assist him to find it in the smaller. The same method is observed here in the description of injustice, and afterwards in Books VIII and IX, where the varieties of ἀδικία in states and individuals are described. The present passage (351 A —352 A), in fact, contains the undeveloped germ of the whole method and doctrine of the Republic (with the exception of Books V—VII). Cf. Hirmer Entst. u. Kompos. d. Pl. Pol. p. 608.

μῶν μὴ (a strengthened num) occurs only twice in the Republic, here and in VI 505 C. In the later dialogues μῶν is especially frequent (Frederking in Fl. Jahrb. 1882 p. 539). A classified list of examples is given by Kugler de part. τοι eiusque comp. ap. Pl. usu p. 40.

οἵαν ‐‐ ποιεῖν . See cr. n. ποιεῖ would involve (as even Schneider admits) “durissimum et haud scio an vitiosum anacoluthon.” Cf. οἷοι μὴ ἀδικεῖν in 334 D. Tucker proposes to eject οἵαν and retain ποιεῖ, but the reading of II is preferable in every way. For the error see Introd. § 5.

παντὶ : i.e. whether just or unjust: cf. 351 E ἐχθροὶ ἔσονται (viz. οἱ ἄδικοι) ἀλλήλοις τε καὶ τοῖς δικαίοις.

ἔστων . On the form see Introd. § 5.

τὰ λοιπά κτλ. : viz. the discussion which begins in D below.

ὅτι μὲν γὰρ κτλ. The whole sentence is summed up in ταῦτα μὲν οὖν ὅτι οὕτως ἔχει (352 D) and placed in this recapitulated form under the government of μανθᾴνω. The introduction of the antithesis ( ἀλλὰ δὴ κτλ.) to οἱ δὲ ἄδικοι οὐδὲν πράττειν μετ’ ἀλλήλων οἶοί τε, and of the explanations required by that antithesis, complicates the sentence, without, however, rendering it obscure. For similar anacolutha with ὅτι see V 465 A, VI 493 D notes and cf. Engelhardt Anac. Plat. Spec. III pp. 38, 40. The whole sentence forms a kind of transition to “the rest of the feast” by summing up what has been so far proved; viz. that Justice is wisdom and virtue ( καὶ σοφώτεροι καὶ ἀμείνους), and more capable of action than Injustice ( δυνατώτεροι πράττειν); even the difficulty raised in ἀλλὰ δὴ—ἀδύνατοι is not new, having been briefly explained in 351 C. Liebhold's ἔτι for ὅτι is an unhappy suggestion; nor should ὅτι be rendered ‘quoniam,’ as Hartman proposes.

μήτοι ‐‐ γε : a strong negative somewhat rarely used by Plato: cf. Phil. 67 A and infra III 388 B, C. See Kugler de part. τοι eiusque comp. ap. Pl. usu p. 11.

ὅντινα τρόπον χρὴ ζῆν . A reminiscence of the πῶς βιωτέον of Socrates: cf. 344 E.

ὃ ἂν ‐‐ ἄριστα . The political applications of this principle are developed from II 369 E onwards: cf. IV 433 A ff.

ἀκούσαις κτλ. The rapid succession of questions makes it possible to dispense with ἄν in the second: cf. II 382 E.

φαῖμεν . See cr. n. If φαμέν is retained, ἄν will belong to εἶναι (cf. VI 493 C), but it is inappropriate here to make εἶναι future or hypothetical. Schneider, while retaining φαμέν, refers ἄν to δικαίως, “ut sensus sit: οὐκοῦν, εἰ ταῦτα τούτων φαμὲν ἔργα εἶναι, δικαίως ἂν φαῖμεν”—a harsh and unnatural view. We may either drop ἄν and keep φαμέν, as (with one of Stobaeus' MSS Flor. 9. 63) I formerly did: or change φαμέν to φαῖμεν. The latter solution is easier and better. Similarly in φαῖμεν below (353 D) the ι is due to A^{2}. See also Introd. § 5.

ἀποτέμοις —see cr. n.—can hardly, I think, dispense with the particle ἄν. It should be noted that the illustrations are of two kinds—the first to illustrate ἢ μόνῳ ἐκείνῳ, the second to illustrate ἄριστα; after each division the conclusion is stated, in the second case more diffidently ( ἆρ’ οὖν οὐ—θήσομεν), perhaps because it is less obvious.

μόνον τι . Cornarius unhappily suggested τις for τι and Stephanus μόνῳ τις for μόνον τι (cf. 352 E). μόνον τι is of course the subject to ἀπεργάζηται .

οὐκοῦν ‐‐ προστέτακται . Cf. Men. 72 A οὐκ ἀπορία εἰπεῖν ἀρετῆς πέρι ὅ τι ἔστιν. καθ’ ἑκάστην γὰρ τῶν πράξεων καὶ τῶν ἡλικιῶν πρὸς ἕκαστον ἔργον ἑκάστῳ ἡμῶν ἡ ἀρετή ἐστιν. ὡσαύτως δὲ—καὶ ἡ κακία: also infra X 608 E with Arist. Eth. Nic. II 5. 1106^{a} 15 ff.

ἦν : ‘is, as we saw,’ viz. at 352 E: cf. infra IV 441 D, VI 490 A, VII 522 A.

ἀπεργάσαιντο . Heindorf (on Crat. 424 E ) would read ἀπεργάσαιτο, and Baiter adopts his suggestion; but (as Stallbaum observes) the use of ὀφθαλμοί just above may affect the construction. In the same way, perhaps, the occurrence of γυναῖκες καὶ τἄλλα θηρία immediately before causes Plato to write δεήσοιντο (the reading of A) rather than δεήσοιτο in Tim. 76 E . Of the other alleged cases of a plural verb after a neuter plural in Plato, some (e.g. Laws 634 E, 683 B) are not supported by the best MSS; one— ἐξ ὧν τά τε ὀνόματα καὶ τὰ ῥήματα συντίθενται (so AT) Crat. 424 E —is distributive; some refer to living objects, e.g. Laws 658 C (with which contrast κρίνοι just before) and Lach. 180 E ; at least one ( Phil. 24 E ) is perhaps corrupt. See also on Rep. II 365 B.

τυφλότητα κτλ. τυφλότης is also said to be the disease or vice of the eyes in Alc. I 126 B, a passage probably imitated from this. In the stricter discussion of X 608 E it is not τυφλότης but ὀφθαλμία which is the vice to which the eyes are subject.

οὐ γάρ πω ‐‐ ἐρωτῶ is ‘I do not, at this stage, enquire’; but the words do not, I think, contain an express promise that the subject will be afterwards resumed. Although the peculiar vice of the eyes is specified in Book X (l.c.), their virtue is not; and τοῦτο refers to ἥτις αὐτῶν ἡ ἀρετή. Cf. 347 E note

ψυχῆς ἔστιν τι ἔργον : cf. III 407 A and Arist. Eth. Nic. I 6. 1097^{b} 22—1098^{a} 17, where this discussion is closely imitated. That it is the ἔργον of soul (and in particular of νοῦς) to rule ( ἄρχειν, ἐπιμελεῖσθαι, and the like), is continually asserted in Plato: see for example Phaedr. 246 B πᾶσα ἡ ψυχὴ παντὸς ἐπιμελεῖται τοῦ ἀψύχου, Crat. 400 A , Phil. 30C, Laws 896 A. The same doctrine is made the ground of the subjection of body to soul which is inculcated in the Phaedo (80 A, 94 B), and in Alc. I 130 A. Cf. also Isocrates περὶ ἀντιδόσεως 180 ὁμολογεῖται μὲν γὰρ τὴν φύσιν ἡμῶν ἔκ τε τοῦ σώματος συγκεῖσθαι καὶ τῆς ψυχῆς· αὐτοῖν δὲ τούτοιν οὐδεὶς ἔστιν ὅστις οὐκ ἂν φήσειεν ἡγεμονικωτέραν πεφυκέναι τὴν ψυχὴν καὶ πλείονος ἀξίαν· τῆς μὲν γὰρ ἔργον εἶναι βουλεύσασθαι καὶ περὶ τῶν ἰδίων καὶ περὶ τῶν κοινῶν, τοῦ δὲ σώματος ὑπηρετῆσαι τοῖς ὑπὸ τῆς ψυχῆς γνωσθεἶσιν.

ἐκείνου . The reading ἐκείνης—see cr. n.—can only be defended by supposing that Plato was guilty of a strange confusion, unless we make a pause at ἄλλῳ, and take ἤ as ‘or,’ not ‘than’; but ἤ after ἄλλῳ would certainly here be understood as ‘than,’ and an alternative question should be less ambiguously expressed. After ψυχῇ the corruption to ἐκείνης was natural enough. Madvig would eject the word.

τὸζῆν is κατ’ ἐξοχήν the ἔργον of ψυχή in Plato: cf. Crat. 399 D , E τοῦτο ἄρα (sc. ψυχή), ὅταν παρῇ τῷ σώματι, αἴτιόν ἐστι τοῦ ζῆν αὐτῷ, τὴν τοῦ ἀναπνεῖν δύναμιν παρέχον καὶ ἀναψῦχον, ἅμα δὲ ἐκλείποντος τοῦ ἀναψύχοντος τὸ σῶμα ἀπόλλυταί τε καὶ τελευτᾷ· ὅθεν δή μοι δοκοῦσιν αὐτὸ ψυχὴν καλέσαι, and Phaed. 105 D . The influence of this idea makes itself felt in all the proofs of immortality in Plato, and not least in X 608 E ff. See notes ad loc.

συνεχωρήσαμεν κτλ. The reference is to 350 C, D: cf. also 348 C. In these passages Justice has been identified with Virtue, but not expressly with virtue of soul. For this reason Hartman would eject ψυχῆς. But as Plato has just been using ἀρετή ‘excellence’ in connexion with things other than soul (ears and eyes), it is important that he should now make it clear that in identifying δικαιοσύνη and ἀρετή, he meant soul's ἀρετή. Otherwise a soul may possess its ἀρετή without being just; in which case the conclusion which he is aiming at will not follow.

ὅ γε εὖ ζῶν κτλ. The ambiguity (as it appears to us) of εὖ ζῆν and εὖ πράττειν is frequently used by Plato to suggest that the virtuous life is the happy one, e.g. Charm. 172 A , 173 D: see note on 335 B. Aristotle says that Plato was the first to establish this identification: see the third fragment of his elegies vv. 4—6 ed. Bergk ὃς μόνος ἢ πρῶτος θνητῶν κατέδειξεν ἐναργῶς | οἰκείῳ τε βίῳ καὶ μεθόδοισι λόγων | ὡς ἀγαθός τε καὶ εὐδαίμων ἅμα γίνεται ἀνήρ.

εἱστιάσθω . The metaphor occurs again in 352 B, V 458 A, IX 571 D. It is one of the formal links connecting the Timaeus with the Republic: see Tim. 17 A . Cf. Shakespeare Macbeth Act I Scene 4 “In his commendations I am fed: It is a banquet to me.”

Βενδιδείοις . See Introd. § 3.

ὑπὸ σοῦ γε κτλ. Plato seems to be making the amende honorable to Thrasymachus: cf. VI 498 C, D μὴ διάβαλλε— ἐμὲ καὶ Θρασύμαχον ἄρτι φίλους γεγονότας, οὐδὲ πρὸ τοῦ ἐχθροὺς γεγονότας.

παραφερομένου . Casaubon's conjecture περιφερομένου is neat, but inappropriate, the reference being to the successive courses at a feast, which were not usually carried round among the Greeks. In Athen. IV 33 the carrying round of viands is mentioned as an Egyptian custom: τρίτη δ’ ἐστὶν ἰδέα δείπνων αἰγυπτιακή, τραπεζῶν μὲν οὐ παρατιθεμένων, πινάκων δὲ περιφερομένων .

ἐγώ μοι δοκῶ κτλ. Lys. 222 E δέομαι οὖν ὥσπερ οἱ σοφοὶ ἐν τοῖς δικαστηρίοις, τὰ εἰρημένα ἅπαντα ἀναπεμπάσασθαι.

ἐγὼ κτλ. λόγου is abstract = τοῦ λέγειν, not ‘the discussion’ (Jowett), which would be τοῦ λόγου. For τὸ δέ see on I 340 D.

ἦν ἄρα : ‘was after all,’ as in IV 443 C τὸ δέ γε ἦν ἄρα—εἴδωλόν τι τῆς δικαιοσύνης and Soph. Tr. 1172 τὸ δ’ ἦν ἄρ’ οὐδὲν ἄλλο πλὴν θανεῖν ἐμέ. With προοίμιον cf. infra VII 531 D, Aesch. P. V. 740 f. ο<*>ς γὰρ νῦν ἀκήκοας λόγους | εἶναι δοκεῖ σοι μηδέπω )ν προοιμίοις, and Shakespeare Macbeth I 3 “As happy prologues to the swelling act Of the imperial theme.” For the sense see the last note on Book I. There is no good ground for supposing (with von Sybel De Platonis Proemiis Academicis) that either Book I of the Republic or the rest of Plato's dialogues were intended merely as προοίμια or ‘Programs’ to attract pupils to his lectures.

βούλει κτλ. The antithesis is between δοκεῖν πεπεικέναι and πεῖσαι, and βούλει is used in its natural sense, not (as Ast thinks) with the force of μᾶλλον βούλει.

λέγε γάρ μοι . Other classifications of ‘goods’ in Plato will be found in Laws 631 B ff. and 697 B ff. (with which compare Arist. Eth. Nic. I 8. 1098^{b} 12 ff.). See also Euthyd. 279 A ff., Gorg. 467 E , Phil. 66 A ff. The nearest parallels to the present classification are furnished by Stoicism, in which goods were classified as a τελικά, b ποιητικά, c both τελικά and ποιητικά, and the προηγμένα as a δἰ αὑτά, b δἰ ἕτερα, c καὶ δἰ αὑτὰ καὶ δἰ ἕτερα see D. L. VII 96, 107.

χαίρειν ‐‐ ἀβλαβεῖς . These ‘innocent pleasures’ are defined in Laws 667 E as those which bring no consequences in their train, good, bad, or otherwise (cf. καὶ μηδὲν εἰς τὸν ἔπειτα χρόνον διὰ ταύτας γίγνεται ἄλλο ἢ χαίρειν ἔχοντα). They are not quite identical with the ‘pure pleasures’ of Phil. 51 B , which are not necessarily devoid of all results, but only of pain. The same conception recurs in Aristotle, who regards the ἀβλαβεῖς ἡδοναί both as conducive to the ethical end and as useful for purposes of recreation ( Pol. Θ 5. 1339^{b} 25).

καὶ μηδὲν κτλ. The relative passes into a demonstrative ( ταύτας) in the second half of the sentence, as in III 412 D, VI 505 D, E, VII 521 B, and elsewhere. The idiom is regular in Greek, but the second pronoun is more usually some case of αὐτός than of οὗτος, e.g. III 395 D, VI 511 C, Gorg. 452 D , Theaet. 192 A . Cobet however ( Mnem. XI p. 167) goes too far in maintaining that αὐτός is alone permissible in this idiom. Cf. Engelhardt Anac. Plat. Spec. III pp. 41—43. μηδέν is used in preference to οὐδέν: for “cogitatione circumscriptum genus significatur” (Schneider). With the sentiment Muretus compared Arist. Eth. Nic. X 2. 1172^{b} 22 οὐδένα γὰρ ἐπερωτᾶν τίνος ἕνεκα ἥδεται, ὡς καθ’ αὑτὴν οὖσαν αἱρετὴν τὴν ἡδονήν.

ἔχοντα : sc. αὐτάς (so also Schneider), not the idiomatic ‘to continue rejoicing’ (as Campbell suggests). The essential mark of these pleasures, viz. that they give pleasure only while they last, is brought out by ἔχοντα, which recalls δεξαίμεθ’ ἂν ἔχειν just above, and is used without an expressed object as in 366 E.

τὸ φρονεῖν ‐‐ ὑγιαίνειν . ἀκούειν is added in 367 C. Cf. Arist. Eth. Nic. I 4. 1096^{b} 16 καθ) αὑτὰ δὲ ποῖα θείη τις ἄν; ἢ ὅσα καὶ μονούμενα διώκεται, οἷον τὸ φρονεῖν καὶ ὁρᾶν καὶ ἡδοναί τινες καὶ τιμαί; ταῦτα γὰρ εἰ καὶ δἰ ἄλλο τι διώκομεν, ὅμως τῶν καθ) αὑτὰ ἀγαθῶν θείη τις ἄν: also Met. A 1. 980^{a} 2 ff. Aristotle himself does not suggest that a special class should be made of things desirable both in themselves and for their results; but integri sensus and bona valetudo are included in the Stoic category of προηγμένα καὶ δἰ αὑτὰ καὶ δἰ ἕτερα (Cic. De Fin. III 56: cf. D. L. VII 107).

γυμνάζεσθαι κτλ. Cf. Prot. 354 A and Gorg. 467 C , D (where χρηματισμός is again said to belong to this class). ἰάτρευσις as an example of χρηματισμός (in spite of the ἀκριβὴς λόγος of I 342 B ff.) is suggested by ἰατρεύεσθαι. ὁ ἄλλος is ‘the rest of,’ and should not be taken (with Stallbaum) as praeterea: cf. Gorg. l.c. οἱ πλέοντές τε καὶ τὸν ἄλλον χρηματισμὸν χρηματιζόμενοι and Crito 53 E.

μισθῶν θ’ ἕνεκα κτλ. Herwerden would read μισθῶν τε μὲν ἕνεκα, but for δέ without μέν preceding see I 340 D note The words διὰ δόξαν, which are condemned by the same critic, may no doubt be a gloss on εὐδοκιμήσεων ἕνεκα. I incline however to think them genuine. Plato is not averse to duplicate expressions of this kind (see Schanz Nov. Comm. Plat. pp. 12—15), and the emphatic addition of διὰ δόξαν helps in the absence of μέν to prepare us for the antithesis αὐτὸ δὲ δἰ αὑτὸ κτλ. Cf. 363 A below.

ψέγεται . See cr. n. The words ἀδικία δ’ ἐπαινεῖται are probably genuine: for the mention of ἀδικία seems to be necessary to justify the pronoun ἑκατέρου just below: cf. also in D βούλομαι καὶ σοῦ ἀκούειν ἀδικίαν μὲν ψέγοντος, δικαιοσύνην δὲ ἐπαινοῦντος. For the omission see Introd. § 5.

ὡς ἀναγκαῖον ἀλλ’ οὐχ ὡς ἀγαθόν . Cf. infra 360 C and VI 493 C τὴν δὲ τοῦ ἀναγκαίου καὶ ἀγαθοῦ φύσιν ὅσον διαφέρει τῷ ὄντι κτλ.

ἀμείνων ἄρα . ἄρα disclaims responsibility for the theory: cf. 362 A, 364 B, E al.

Θρασυμάχου ‐‐ ἄλλων . See on I 337 A ff.

κατατείνας κτλ. : ‘I will speak vehemently in praise of the unjust life.’ The explanation of Photius and Suidas ( κατατείνας ἐρῶ· ἀντὶ τοῦ μακρὸν λόγον διεξελεύσομαι) does not suit II 367 B ὡς δύναμαι μάλιστα κατατείνας λέγω. For this intransitive use of κατατείνω cf. I 348 A and Boeckh's emendation of Eur. Iph. Aul. 336 οὔτε κατατενῶ ( καταινῶ MSS) λίαν ἐγώ.

εἴ σοι βουλομένῳ . In Crat. 384 A ἐστί is again omitted in this phrase. A still bolder example is cited by Stallbaum from Antipho 6. 8 ἐὰν ὑμῖν ἡδομένοις. See Schanz Novae Comm. Plat. pp. 31— 35.

οἷόν τέ τι . The reading of A τί ὄν τε καὶ ὅθεν γέγονε involves the separation of ὅθεν from γέγονε, and is otherwise much too harsh to be right. There is something to be said in favour of Schneider's περὶ τούτου ἅκουε τί οἴονται, καὶ ὅθεν γέγονε δικαιοσύνη (see cr. n.), especially as the confusion between οἷόν τε and οἴονται occurs rather frequently in Platonic MSS (see Schneider on 1 329 E), but the specific reference in ὃ ἔφην πρῶτον ἐρεῖν to 358 C πρῶτον μὲν ἐρῶ δικαιοσύνην οἷον εἶναί φασι καὶ ὅθεν γεγονέναι points to the presence of οἷον here. The reading οἷόν τε, adopted by Stallbaum, as well as by Jowett and Campbell, on the authority of three MSS (Vind. F, Flor. RT), is unexceptionable in point of sense, but fails to account for the presence of τί in the best MSS. I have ventured to read οἷόν τέ τι (SC. ἐστί), supposing that the confusion arose from the accidental omission of τι, which was afterwards (as τί) wrongly inserted before οἶόν (where it remained in Π), οἷον itself being afterwards changed to ὄν in order to provide a kind of construction (‘being what, and whence, it arises,’ J. and C.). This ὄν was itself fortified by τυγχάνει in Flor. B and the Aldine edition. Campbell's suggestion that “ τί ὄν τε may be a corruption of τί ἐστί” is improbable: still less can Herwerden and Hartman induce us to reject the whole clause. Few will approve of Tucker's conjecture τί τῷ ὄντι καὶ ὅθεν κτλ. Dr Jackson suggests ἄκουέ τι, οἷόν τε καὶ κτλ., and a reviewer of my Text of the Republic in Lit. Centralblatt 1898 p. 296 οἷόν τ’ ἐστί κτλ.

πεφυκέναι γὰρ ‐‐ κακόν . Cf. Gorg. 483 A φύσει μὲν γὰρ πᾶν αἴσχιόν ἐστιν ὅπερ καὶ κάκιον, τὸ ἀδικεῖσθαι, νόμῳ δὲ τὸ ἀδικεῖν. That the natural relation between man and man is one of war is a view expressed in Laws 626 A ἢν γὰρ καλοῦσιν οἱ πλεῖστοι τῶν ἀνθρώπων εἰρήνην, τοῦτ’ εἶναι μόνον ὄνομα, τῷ δ’ ἔργῳ πάσαις πρὸς πάσας τὰς πόλεις ἀεὶ πόλεμον ἀκήρυκτον κατὰ φύσιν εἶναι. A similar theory is contained in the myth of Protagoras ( Prot. 322 B ff.).

τοῖς μὴ δυναμένοις κτλ. : i.e. (according to the theory of Callicles) τοῖς ἀσθενέσι ἀνθρώποις καὶ τοῖς πολλοῖς ( Gorg. 483 B ). In place of δοκεῖ in 359 A I have adopted Ast's conjecture δοκεῖν. Throughout this paragraph Glauco consistently presents his view at second hand. For the collocation of infinitives cf. ἀδικεῖν, ἀδικεῖν 360 D, and for the error itself Introd. § 5.

ξυνθήκας αὑτῶν : ‘covenants between one another,’ ‘mutual covenants.’ Reading αὐτῶν, Tucker suggests that the meaning is, ‘they established laws and covenants concerning them,’ i.e. concerning matters connected with ἀδικεῖν and ἀδικεῖσθαι—a very improbable view.

νόμιμόν τε καὶ δίκαιον · φημὶ γὰρ ἐγὼ τὸ νόμιμον δίκαιον εἶναι, said Socrates ( Mem. IV 4. 12).

τοῦ μὲν ἀρίστου κτλ. Cf. the reasoning of Philus (whose position in Cicero's work corresponds to that of Glauco here) in Cic. de Rep. III 23 “nam cum de tribus unum esset optandum, aut facere iniuriam nec accipere, aut et facere et accipere, aut neutrum, optimum est facere, impune si possis, secundum nec facere nec pati, miserrimum digladiari semper tum faciendis tum accipiendis iniuriis.” Cicero is following Carneades ( ibid. 8), who may have been thinking of the present passage. ἀγαπᾶσθαι below (as J. and C. observe) “implies acquiescence rather than decided preference.”

ἐπεὶ τὸν δυνάμενον κτλ. is further elaborated with much vigour in Gorg. 484 A . With ὡς ἀληθῶς ἄνδρα should be compared the emphatic ἀνήρ in that passage ( ἐὰν δέ γε, οἶμαι, φύσιν ἱκανὴν γένηται ἔχων ἀνήρ), and Eur. Phoen. 509 ἀνανδρία γάρ, τὸ πλέον ὅστις ἀπολέσας | τοὔλασσον ἔλαβε.

εἰ τοιόνδε ‐‐ δόντες . δόντες κτλ. explains τοιόνδε. εἰ need not be twice expressed: cf. I 351 C note

νόμῳ ‐‐ παράγεται . The language is perhaps suggested by the lines of Pindar cited in Gorg. 484 B νόμος ὁ πάντων βασιλεὺς θνατῶν τε καὶ ἀθανάτων— ἄγει δικαιῶν τὸ βιαιότατον ὑπερτάτᾳ χειρί κτλ. (cf. Prot. 337 D ). but the preposition in παράγεται adds the further notion that equality is not Nature's highway. For βίᾳ i.q. βιαἴως in conjunction with another dative Schneider cites VIII 552 E οὓς ἐπιμελείᾳ βίᾳ κατέχουσιν αἱ ἀρχαί. In the next line it is better to regard τοιάδε as explained by εἰ—γενέσθαι, than as balancing οἴαν, in which case εἰ αὐτοῖς γένοιτο would be superfluous. The opportunity ( ἐξουσία) of working their will comes from the possession ( εἰ αὐτοῖς γένοιτο) of a certain active faculty ( δύναμις) like that of Gyges.

τῷ Γύγου κτλ. Cf. X 612 B τὸν Γύγου δακτύλιον. In Appendix I I have given reasons for believing that the Gyges of the proverbial ‘Gyges' ring’ was not “Gyges the Lydian”—the hero of Herodotus' story (I 7), but a homonymous ancestor of his. If so, we must (on the hypothesis that the text is sound) suppose that Plato here omits the name of the original Gyges either because he wishes tacitly to contradict a prevalent misconception, or (more probably) because his readers might be presumed to know or to be capable of inferring that the ancestor of Gyges the Lydian was also called Gyges. The MS reading is supported by Proclus ( τῷ κατὰ τὸν Γύγου πρόγονον διηγήματι in Schöll Procli Comm. in Remp. Pl. part. ined. p. 60. 30). For other views of this passage see App. I.

ὡς φαίνεσθαι : with νεκρόν, as Schneider saw: “utrum vere mortuus fuerit, an specie, fabula incertum reliquit.” Stallbaum wrongly interprets ‘nimirum videbatur Gyges cernere’ etc.: this would be expressed by δοκεῖν. Ast connects the phrase with μείζω ἢ κατ’ ἄνθρωπον: but this is very weak in point of sense. The words are omitted by Cicero ( De Off. III 38).

ἔχειν . See cr. n. and (for the omission in A) Introd. § 5. ἔχειν in the sense of ‘have on’ ‘wear,’ i.q. φορεῖν, is tolerably frequent in Homer, though rarer in Attic: see Stephanus-Hase Thes. s.v. For the change of subject in ἔχειν—ἐκβῆναι cf. III 414 D note Other views on the text and interpretation of this passage are discussed in App. II.

χειρί . Herwerden's δακτύλῳ is unnecessary, and even unpleasant with δακτύλιον so near. Cf. χρυσόχειρες in Luc. Tim. 20. “Etiamnunc homines ita loquuntur” (Hartman).

ἵν’ ἐξαγγέλλοιεν κτλ. : ‘to report, as was done every month.’ The present expresses the habit (J. and C.).

σφενδόνην : the ‘collet’ or ‘bezel’ (Lat. funda or pala annuli)— which is as it were the sling in which the stone is set.

ὡς δόξειεν . “Optativus eandem vim habet, quam solet in oratione obliqua habere, efficitque, ut verba οὐδεὶς ἂν γένοιτο οὕτως etc. ex aliorum ore missa videantur" (Schneider). This explanation appears to me better than any other, although I can discover no exact parallel in Greek. Glauco is most careful throughout the whole of this section to disclaim responsibility for the views he advocates: cf. ὡς ὁ λόγος 359 B, ἐπεὶ— ἀδικεῖν in C, ὡς φήσει κτλ. in D below: also 361 E al. Tucker would translate ‘as it might seem,’ defending the optative by Ar. Birds 180 ὥσπερ εἴποι τις and Eur. Andr. 929 ὡς εἴποι τις. Others erroneously hold that ἄν may be supplied from ἂν γένοιτο, while Ast is desirous of inserting the particle on conjecture. I do not think that the optative can be explained as an instance of irregular assimilation or attraction.

ἂν μείνειεν . For ἄν cf. Symp. 179 A and other examples in Kühner Gr. Gr. II p. 934.

ἰσόθεον ὄντα . The halfconscious irony of ἰσόθεος foreshadows Plato's attack on the popular theology.

οὐδεὶς ἑκὼν δίκαιος here and in 366 D sums up the Thrasymachean theory in a phrase which suggests the Socratic and Platonic antithesis οὐδεὶς ἑκὼν πονηρός .

περὶ ‐‐ λέγων . περί can hardly be for ὑπέρ, nor dare we write ὑπέρ for περί (as Badham suggests). The words mean simply ‘qui de hoc argumento verba facit,’ ‘the exponent of such a theory.’ Cf. 362 D ἱκανῶς εἰρῆσθαι περὶτοῦλόγου. Muretus seems to have desiderated πατήρ for περί: cf. expressions like Phaedr. 275 E ( λόγος) τοῦ πατρὸς ἀεὶ δεῖται βοηθοῦ. On the strength of this Herwerden would read ὁ πατὴρ τοῦ τοιούτου λόγου, rejecting λέγων (“posteaquam ex πήρ factum est περί, corrector addidit λέγων”). The ‘father of the theory’ would mean Thrasymachus: see on ὦ παῖδες in 368 A. It is just possible that πατήρ was read by Ficinus (“ut sermonis huius perhibent auctores”), and if so, the variant may have some ancient authority now lost; but Herwerden's proposal is too drastic, and the text is probably sound.

ἀθλιώτατος . Apelt conjectures ἠλιθιώτατος, but cf. (with Hartman) I 344 A τοὺς δὲ—ἀδικῆσαι οὐκ ἂν ἐθέλοντας ἀθλιωτάτους.

τὴν δὲ κρίσιν κτλ. αὐτήν opposes the third division of Glauco's speech to the other two (see 358 C), and marks it as the most important. A kindred use of αὐτός recurs at 370 E ἀλλὰ μήν,—κατοικίσαι γε αὐτὴν τὴν πόλιν —ἀδύνατον. I formerly read αὖ τὴν for αὐτήν, but the MS reading is quite de fensible. It should be noticed that κρίσιν is at first a kind of pendent accusative, afterwards “resumed as a cognate accusative with κρῖναι” (J. and C.). Tucker strangely makes κρίσιν=‘choice.’ The word means of course (our) ‘judgment’ concerning etc. Cf. 361 D ἵν’ ἀμφότεροι —κρίνωνται and εἰς τὴν κρίσιν ἐκκαθαίρεις.

εἰς goes with τέλεον: cf. δόξαν εἰς 361 A.

οὕτω ‐‐ λανθανέτω . ἐπιχειρῶν ὀρθῶς means of course attempting possible, and abstaining from impossible, ἀδικήματα. But as an ἀδίκημα is possible only if the ἀδικῶν is able to conceal it (the alternative of open violence is recognised later 361 B), it is necessary that the unjust man should escape detection. Hence λανθανέτω, although λανθάνειν was not attributed (because not essential) to the pilot and doctor (360 E).

φαῦλον means a ‘bungler’ (D. and V.). With the sentiment cf. Prot. 317 A τὸ οὖν ἀποδιδράσκοντα μὴ δύνασθαι ἀπξδρᾶναι, ἀλλὰ καταφανῆ εἶναι, πολλὴ μωρία καὶ τοῦ ἐπιχειρήματος: also Laws 845 B, and the Spartan practice of punishing boys not for stealing, but for being caught (Xen. Rep. Lac. 2. 8). With ἐσχάτη γὰρ ἀδικία κτλ. the editors compare Cicero de Off. I 41 “totius autem iniustitiae nulla capitalior est, quam eorum, qui, cum maxime fallunt, id agunt, ut viri boni esse videantur.”

κατ) Αἰσχύλον ‐‐ ἀγαθόν . Sept. 592—594 (of Amphiaraus) οὐ γὰρ δοκεῖν ἄριστος, ἀλλ’ εἶναι θέλει | βαθεῖαν ἄλοκα διὰ φρενὸς καρπούμενος, | ἐξ ἧς τὰ κεδνὰ βλαστάνει βουλεύματα. Herwerden would expunge ἀγαθόν (“mente repetatur ἁπλοῦν καὶ γενναῖον”), on the ground that if Plato had added any adjective, it would have been δίκαιον. (The Scholiast substitutes δίκαιος for ἄριστος in Aeschylus.) ἀγαθόν gives excellent sense, and is nearer to the poet's words.

εἴη is explained by Stallbaum as an optative of wish (though in a subordinate clause): ‘it is not clear therefore whether he is fain to be just,’ etc. This gives a fair sense, but the idiom is obscure, and unsupported by other examples. J. and C. remark that “the optative accords with the conditional nature of the case in an imagined future,” taking ἄδηλον as for ἄδηλον ἂν εἴη. But an omitted ἂν εἴη cannot be responsible for the mood of τοιοῦτος εἴη, nor could ἂν εἴη easily be omitted (see Schanz Nov. Comm. Pl. p. 33). Still less should we accept Hartman's ἄδηλον <ἂν> οὖν, sc. εἴη. Madvig ejects εἴη altogether, understanding ἐστι after τοιοῦτος. This may be right, but its intrusion is not easy to explain. I think the word is genuine, and means ‘was’: ‘it is not clear then, say they, whether he was just,’ etc. Glauco again disclaims responsibility: cf. 360 B note εἴη would in direct speech be ἦν: and the idiom is like that in III 406 E, where see note. For the sequence of moods and tenses cf. VI 490 A note Failing this interpretation, the word must (I think) be spurious. Herwerden's proposal— τοιούτῳ εἶναι, ἄδηλον ὄν (retaining εἴη)—does not surmount the difficulty and is also wrong in point of sense.

ἀπ’ αὐτῆς . See cr. n. The sense required is not ‘what is produced by’ ( ὑπό) ‘it,’ but ‘what results from it’: cf. γίγνεσθαι ἀπό (in a similar connexion) 357 C and 358 B. The scribe no doubt assimilated the preposition to the preceding ὑπό.

ἔστω . See cr. n. I formerly read ἴτω with A^{1} and the majority of editors, but I now agree with Schneider that ἔστω is right. ἴτω cannot be used by itself as a synonym for ‘live,’ or as a copula: we should require ἴτω διὰ βίου, instead of ἴτω μέχρι θανάτου (to transpose the two phrases would of course be too violent a change). The sole authority for ἴτω is the first hand in A: and this is certainly insufficient to outweigh the inherent superiority of ἔστω. Most MSS have ἤτω, a late form for ἔστω.

ἐκκαθαίρεις : not ‘polish up’ (J. and C.) but rather ‘scour clean’ (D. and V.), ‘purge’ from all extraneous matter: see 361 C γυμνωτέος δὴ πάντων πλὴν δικαιοσύνης.

ἀγροικοτέρως is said with reference to the exaggeration and coarseness of the description: cf. Ap. 32 D , Gorg. 509 A .

δεδήσεται : ‘will be kept in chains.’ δεθήσεται (so v and some other MSS) is required by Herwerden, and may be right. But in Xen. Cyr. IV 3. 18 δεδήσομαι is similarly combined with several first futures.

ἐκκαυθήσεται κτλ. Schneider refers to Hdt. VII 18 θερμοῖσι σιδηρίοισι ἐκκαίειν—τοὺς ὀφθαλμούς, and Gorg. 473 C ἐὰν—στρεβλῶται καὶ ἐκτέμνηται καὶ τοὺς ὀφθαλμοὺς ἐκκάηται. That ἐκκαυθήσεται (and not ἐκκοπήσεται, the reading of some inferior MSS, and of the ancient authorities who cite this passage) is right here, is probable also from X 613 E ἃ ἄγροικα ἔφησθα σὺ εἶναι ἀληθῆ λέγων, εἶτα στρεβλώσονται καὶ ἐκκαυθήσονται, whether the last clause is genuine or not. It is not clear that Cicero ( de Rep. III 27) did not find ἐκκαυθήσεται in his text; for though he has effodiantur oculi, he adds afterwards vinciatur, uratur. Herwerden recasts the words of Plato to suit Cicero's translation, but Cicero is a much less trustworthy witness than Paris A.

ἄρα : see on 358 C. τῷ ὄντι in the same line belongs not to φήσουσι, but to τὸν ἄδικον—ἐθέλειν.

βαθεῖαν κτλ. : “reaping in his thoughts the fruit of the deep furrow, from which good counsel grows” (Verrall). Plato takes τὰ κεδνὰ βουλεύματα more concretely, and places in apposition thereto ἄρχειν and the other infinitives down to ὠφελεῖσθαι, δοκοῦντι being the dative of interest after βλαστάνει. For the change from the dative δοκοῦντι to the accusative κερδαίνοντα cf. Euthyph. 5 A and infra IV 422 B, C.

κοινωνεῖν . Cobet deletes this word, as well as καὶ κοινωνήματα in Laws 738 A πρὸς ἅπαντα τὰ ξυμβόλαια καὶ κοινωνήματα. In view of the same passage Platt ( Cl. Rev. III p. 72) would read καὶ κοινωνεῖν. No change is necessary, for κοινωνεῖν is a term of wider connotation than ξυμβάλλειν (see I 333 A note), and the asyndeton has a rhetorical effect: cf. III 407 B, V 465 C, VI 488 C, IX 590 A notes

πλεονεκτεῖν recalls I 343 D, E, 349 B ff., as τούς τε φίλους εὐ ποιεῖν κτλ. recalls the theory attributed to Simonides in I 334 B. Here however it is not Justice, but Injustice masquerading as Justice, which is said to benefit friends and injure enemies.

μᾶλλον προσήκειν . The comparative is attached to the verb as well as to the adjective, so as to combine the force of two expressions, viz. (1) ὥστε καὶ θεοφιλῆ αὐτὸν εἶναι μᾶλλον προσήκειν and (2) ὥστε καὶ θεοφιλέστερον αὐτὸν εἶναι προσήκειν. In cases like λαθραιότερον μᾶλλον Laws 781 A, μᾶλλον is quite redundant: in Hipp. Mai. 285 A ἔστι δέ γε—ὠφελιμώτερον—παιδεύεσθαι μᾶλλον ἢ κτλ. it is resumptive. See on the whole subject Kühner Gr. Gr. II p. 25.

παρεσκευάσθαι ‐‐ ἄμεινον . For ἄμεινον Richards would read ἀμείνον) or ἀμείνονα: cf. 358 C πολὺ γὰρ ἀμείνων ἄρα ὁ τοῦ ἀδίκου ἢ ὁ τοῦ δικαίου βίος. The change is tempting at first sight; but Plato generally uses ἀμείνω and not ἀμείνονα, and the adverb expresses what is virtually the same meaning, since a βίος ἄμεινον παρεσκευασμένος (cf. πόλιν εὖ παρεσκευασμένην Laws 751 B) is (according to the views here described) a βίος ἀμείνων. Hermann's χείρον) for χεῖρον in Phaed. 85 B , though adopted by Schanz, is also unnecessary, for ἔχειν may be intransitive.

ἔφη . See cr. n. ἔφη is present in the majority of MSS, and cannot be dispensed with, where the interlocutor is specified, as here. See Introd. § 5.

ἀδελφὸς ἀνδρὶ παρείη : frater adsit fratri. Ast proposed to insert ἂν before ἀνδρί, making the sentence interrogative. The rhythm would thus approximate to the usual paroemiac rhythm of proverbs: but the brevity and force of the proverb would suffer. If change were needed it would be better to adopt Shilleto's elegant suggestion ἀδελφεὸς ἀνδρὶ παρείη (note on Dem. F. L. § 262), but even if this was the original expression, it would be quite in Plato's manner to substitute the modern for the archaic word, in defiance of rhythm. The source of the proverb (with which compare συγγνώμη ἀδελφῷ βοηθεῖν F. L. § 264) is found by the Scholiast in Od. XVI 97 f. ἤ τι κασιγνήτοις ἐπιμέμφεαι, οἷσί περ ἀνὴρ | μαρναμένοισι πέποιθε, καὶ εἰ μέγα νεῖκος ὄρηται. Cf. also Il. XXI 308 f. and Xen. Mem. II 3. 19.

ἐναντίους . Adimantus' λόγοι are ἐναντίοι, because they praise Justice, and censure Injustice: whereas Glauco had done the reverse: κατατείνας ε<*>ρῶ τὸν ἄδικον βίον ἐπαινῶν (358 D).

αὐτὸ δικαιοσύνην . Not αὐτοδικαιοσύνην (with the second hand in A), which would be the (chiefly postPlatonic) expression for the Idea of Justice (cf. αὐτοάνθ ρωπος and the like). αὐτό is ipsum, ‘by itself,’ as in αὐτοὶ γάρ ἐσμεν: cf. Theaet. 146 E γνῶναι ἐπιστήμην αὐτὸ ὄ τι ποτ’ ἐστιν, and infra V 472 C, X 612 B (cited by J. and C.). αὐτό may be thus used even when the feminine of the article is present, e.g. Prot. 361 A αὐτὸ ἡ ἀρετή: cf. also Crat. 411 D .

γίγνηται . The nominatives are treated as equivalent to a neuter plural, whence the singular verb. Cf. Symp. 188 B , Laws 925 E, Andocides I 145. γίγνεσθαι is the verb in each of these examples. See also infra V 462 E.

τῷ δικαίῳ . Schneider is right in refusing to change the δικαίῳ of A, II and most MSS to ἀδίκῳ, which has the authority of a few inferior MSS. The reference in διῆλθεν ἄρτι is no doubt to 362 B, where the benefits accrue to the man who seems to be just, although in reality he is unjust. But ὄντα etc. should be taken, not with διῆλθεν, but as part of the parents' exhortation. This yields a better rhythm, and much better sense. The parents exhort their children to be just, in order that ( ἵνα depends on χρὴ δίκαιον εἶναι) they may obtain the rewards ἀπὸ τοῦ εὐδοκιμεῖν ὄντα τῷ δικαίῳ. They very properly assume that the surest way to seem to be just (and so to obtain the rewards of justice) is to be just: cf. Xen. Mem. II 6. 39 συντομωτάτη τε καὶ ἀσφαλεστάτη καὶ καλλίστη ὁδὸς—ὅ τι ἂν βούλῃ δοκεῖν ἀγαθὸς εἶναι, τοῦτο καὶ γενέσθαι ἀγαθὸν πειρᾶσθαι and ib. I 7. 1 with Heracl. Fr. 137 ed. Bywater συντομωτάτην ὁδὸν—εἰς εὐδοξίαν τὸ γενέσθαι ἀγαθόν. Glauco's picture of the just man as one who seems to be unjust is untrue to the facts of experience, as Socrates points out in X 612 D: nor did even Glauco go so far as to say that the unjust man, qua unjust, ηὐδοκίμει, but only ὁ δοκῶν δίκαιος εἶναι (who may, of course, be unjust). The divorce between appearance and reality is purely argumentative, and out of place in parental exhortations. Further, in order to make ἀπὸ τοῦ εὐδοκιμεῖν ὄντα etc. represent what Glauco said, we should have to read τῷ ἀδίκῳ μὲν δοκοῦντι δὲ δικαίῳ: otherwise the words δοξαζομένων δὲ ἀδίκων in the corresponding phrase (363 E) might just as well be omitted. If ὄντα is construed with διῆλθεν, the words τῷ δικαίῳ must (with Ast) be expunged: but that the clause represents what the parents say is further proved by the exact correspondence of ἀπὸ τοῦ εὐδοκιμεῖν ὄντα τῷ δικαίῳ with τὰς ἀπ’ αὐτῆς (sc. δικαιοσύνης ) εὐδοκιμήσεις, which is what the parents praise. I have dwelt on this point at some length because recent English editors (except Tucker) have wrongly deserted Paris A.

τοῖς ὁσίοις depends on ἀγαθά (‘good things for the pious’): cf. ἀγαθὰ διῆλθε τῷ τοῦ ἀδίκου I 348 A note This is much simpler than to punctuate ἀγαθά, τοῖς ο<*>σίοις ἅ as the other editors do. Such a postponement of the relative is rare, and here, I think, unduly harsh, in spite of the analogy of III 390 B and IV 425 C. Cobet felt the difficulty when in an unhappy moment he suggested ἀγαθά, ἃ τοῖς ὁσίοις κτλ.

Ἡσίοδός τε κτλ. Hesiod and Homer are appealed to as recognised theological authorities: see Hdt. II 53.

ἄκρας ‐‐ καταβεβρίθασι . OD. 232 f. τοῖσι (i.e. ἰθυδίκῃσιν ἀνδράσι) φέρει μὲν γαῖα πολὺν β<*>ον, οὔρεσι δὲ δρῦς | ἄκρη μέν τε φέρει βαλάνους, μέσση δὲ μελίσσας· | εἰροπόκοι δ’ ὄϊες μαλλοῖς καταβεβρίθασι. Further rewards of justice ( ἄλλα δὴ πολλὰ ἀγαθά) are enumerated in vv. 227—231, and 235—237. Many other illustrations in support of Plato's attack on Greek religion throughout this passage will be found in Nägelsbach's Hom. Theol. and Nachhom. Theol. passim.

ὥστε τευ ‐‐ ἰχθῦς . Od. XIX 109 ff. The ἤ before βασιλῆος is difficult: apparently the author intended to give two comparisons, but dropped the second. We are hardly justified, I think, in abolishing the anacoluthon by reading (with Platt) ὥστέ τεο βασιλῆος or (with Ameis) ὥστε τευ ἦ.

Μουσαῖος κτλ. By Musaeus' son Plato probably means Eumolpus (cf. Suidas s. vv. Εὔμολπος and Μουσαῖος). In this section of the argument Plato directs his attack against certain forms of the Orphic conception of a future life: see Lobeck Aglaophamus p. 807 with Rohde Psyche^{2} II pp. 127, 129 notes, and Dieterich Nekyia pp. 72 ff. 77 ff. notes Lobeck refers to Plut. Comp. Cim. et Lucull. 2 Πλάτων ἐπισκώπτει τοὺς περὶ τὸν Ὀρφέα τοῖς εὖ βεβιωκόσι φάσκοντας ἀποκεῖσθαι γέρας ἐν ᾅδου μέθην αἰώνιον and id. Ne suav. quidem vivi posse sec. Epic. 1105 B, where the allusion to Plato is less clear: also D. L. VI 4.

συμπόσιον τῶν ὁσίων . ὅσιοι was the regular appellation of the μύσται ( ὁσίους μύστας hymn. Orph. 84. 3 ed. Abel). For the συμπόσιον cf. [ Axioch.] 371 D συμπόσιά τε εὐμελῆ καὶ εἰλαπίναι αὐτοχορήγητοι καὶ ἀκήρατος ἀλυπία καὶ ἡδεῖα δίαιτα. The stock example in antiquity of earthly virtue rewarded by the delights of a sensuous paradise is Heracles: see e.g. Pind. Nem. I 71, Theocr. XVII 28 f. and Horace Od. III 3. 9 f., IV 8. 29 f. A somewhat higher note is struck in Pind. Ol. II 61 ff. and Fr. 129 f. Several of these passages shew traces of Orphic influence, but the special instance of Heracles is traceable to Homer ( Od. XI 602 f.).

μέθην αἰώνιον may be illustrated from the fragment of Pherecrates ap. Athen. VI 268 E ff.

ἀποτίνουσιν . See cr. n. The reading of A is defended by Stallbaum as an abbreviation for μακροτέρους λόγους ἀποτείνουσι περὶ μισθῶν παρὰ θεῶν; but no other example of this harsh condensation has been adduced, and the sense is far from satisfactory. A better meaning is conveyed by Schneider's translation, “Andere aber lassen die Belohnungen der Götter noch weiter reichen als diese”: for it is clear from the next clause that μακροτέρους (‘more extensive,’ not, of course, ‘greater,’ which would be μείζους) refers to the extension of the rewards of virtue beyond the personality of the individual concerned. But μακροτέρους ἀποτείνουσιν μισθούς is (to say the least) an obscure and difficult expression; and ἀποτίνουσιν (i.q. λέγουσιν ἀποτίνεσθαι) receives strong support from the parallel use of διδόασιν in C above, and κατορύττουσιν, ἀναγκάζουσι, and ἄγοντες below. The collocation of μακροτέρους with ἀποτίνουσι may easily have led to the corruption ἀποτείνουσι, owing to the frequency of such expressions as μακροὺς λόγους ἀποτείνειν. For the error see Introd. § 5.

παῖδας ‐‐ κατόπισθεν . The Scholiast remarks ἐξ Ἡροδότου (VI 86) ἀπὸ τοῦ δοθέντος χρησμοῦ Γλαύκῳ τῷ Λάκωνι ὡς Ἀνδρὸς δ’ εὐόρκου γενεὴ μετόπισθεν ἀμείνων. The story of Glaucus admirably illustrates the view herein expressed; but Plato is more probably thinking of Hesiod OD. 285 (a line which is identical with that quoted from the oracle), and also perhaps of some such lines as those of Tyrtaeus 12. 29 f. καὶ τύμβος καὶ παῖδες ἐν ἀνθρώποις ἀρίσημοι | καὶ παίδων παῖδες καὶ γένος ἐξοπίσω.

εἰς πηλόν τινα κατορύττουσιν . τινα is contemptuous: ‘something which they call mud’: cf. 372 B infra and Symp. 210 D ( ἀνθρώπου τινός). The ‘mud’ is Orphic: see Abel Orphic. p. 247 and cf. Phaed. 69 C , Rep. VII 533 D, and the σκῶρ ἀείνων of Ar. Frogs 146 , with Blaydes' note. See also Rohde Psyche^{2} I p. 313 note and Dieterich Nekyia pp. 82 f. The employment of the Danaid legend in Orphic teaching is illustrated by Gorg. 493 B : cf. also Dieterich Nekyia pp. 69 f., 75.

δοξαζομένων δέ . For δέ without μέν see I 340 D note

ἄλλα δὲ οὐκ ἔχουσιν : sc. λέγειν τιμωρήματα. Adimantus means that they dissuade men from injustice merely on account of its results, ignoring τίνα ἔχει δύναμιν αὐτὸ καθ’ αὑτὸ ἐνὸν ἐν τῇ ψυχῇ (358 B). J. and C. aptly cite Theaet. 176 D , E ἀγνοοῦσι γὰρ ζημίαν ἀδικίας, ὃ δεῖ ἥκιστα ἀγνοεῖν· οὐ γάρ ἐστιν ἣν δοκοῦσιν, πληγαί τε καὶ θάνατοι, ὧν ἐνίοτε πάσχουσιν οὐδὲν ἀδικοῦντες, ἀλλ’ ἣν ἀδύνατον ἐκφυγεῖν, viz. “that by their wicked acts they become like the pattern of evil.”

ἰδίᾳ has been understood of writing in prose, but the reference is only to the representations of private persons, e.g. parents, etc.)(to poets, who were in a sense the professional teachers of Hellas: cf. X 606 C, Laws 890 A ἰδιωτῶν τε καὶ ποιητῶν, and 366 E below.

καλὸν μὲν ‐‐ ἐπίπονον . See cr. n. For the omission of τε καὶ δικαιοσύνη see Introd. § 5. The sentiment may be illustrated by Hesiod OD. 289—292 and Simon. ap. Pl. Prot. 339 B ff. ἄνδρ’ ἀγαθὸν μὲν ἀλαθέως γενέσθαι χαλεπόν κτλ.; cf. also Simonides' imitation of Hesiod ( Fr. 58 ed. Bergk).

ὡς ἐπὶ τὸ πλῆθος : i.q. ὡς ἐπὶ τὸ πολύ. So also Phaedr. 275 B . The sentiment recurs in Isocr. de Pace § 31.

πονηρούς is the substantive, and ἄλλας δυνάμεις ἔχοντας balances πλουσίους. πλούτους, parallel to ἄλλας δυνάμεις, and also dependent on ἔχοντας, might appear neater. But there is no reason for deserting the MSS, although Plato is fond of the plural of πλοῦτος (cf. e.g. VI 495 A, X 618 B, 619 A). The sentiment is best illustrated from Polus's description of the happiness of Archelaus in Gorg. 471 A ff.

ὡς ἄρα ‐‐ μοῖραν . ἄρα hints dissent: cf. 358 C note The gnomic poets often express themselves in this vein: e.g. Solon 15. 1 πολλοὶ γὰρ πλουτεῦσι κακοί, ἀγαθοὶ δὲ πένονται, Theogn. 373— 380. A kindred sentiment occurs in Sophocles Phil. 447—452. For the most part however it is held that Justice asserts herself in the end: see for example Solon 4. 15 f., 13. 7—32. Euripides expresses the general teaching of Greek tragedy on this subject when he writes ( Ion 1621 f.) ἐς τέλος γὰρ οἱ μὲν ἐσθλοὶ τυγχάνουσιν ἀξίων, | οἱ κακοὶ δ’ , ὥσπερ πεφύκασ’ , οὔποτ’ εὖ πράξειαν ἄν. There is no occasion to write (with Richards) πολλάκις τοῖς for πολλοῖς.

ἐπὶ πλουσίων θύρας ἰόντες . This semi-proverbial expression (cf. VI 489 B, C) stigmatises the avarice of seers and mendicant priests ( ἀγύρται from ἀγείρω, cf. infra 381 D). Plato's contempt for μαντική in general is expressed in the Euthyphro and sporadically in various dialogues (see e.g. Tim. 71 E , with Archer-Hind's note); but his attack is here particularly directed (cf. infra 364 E) against such Ὀρφεοτελεσταί or Orphic friars as Theophrastus speaks of in his description of the δεισι‐ δαίμων ( Charact. 16) καὶ τελεσθησόμενος πρὸς τοὺς Ὀρφεοτελεστὰς κατὰ μῆνα πορεύεσθαι μετὰ τῆς γυναικός, ἐὰν δὲ μὴ σχολάζῃ ἡ γυνή, μετὰ τῆς τίτθης καὶ τῶν παιδίων. The kind of ceremonies which they practised may be seen from Dem. de Cor. §§ 258 ff. Plato agreed with the more enlightened section of his countrymen in condemning such degrading cults and superstitions on the ground of their immoral tendency: see especially Foucart des Assoc. religieuses chez les Grecs pp. 153 —157, where the opinions of ancient writers on this subject are collected. On ἀγύρται in general reference may be made to J. H. Wright in Harvard Studies in Cl. Philol. VI p. 66 note

ἐάν τε ‐‐ βλάψειν is in oratio obliqua: ‘et si quis inimicum laedere velit, nocituros se parvo sumptu iusto pariter et iniusto’ (Schneider Addit. p. 11). This explanation (which Tucker also proposes without knowing that Schneider had forestalled him) is by far the best and simplest. For other views see App. III.

ἐπαγωγαῖς ‐‐ καταδέσμοις . ἐπαγωγαί are ἀγωγαὶ δαίμονος φαύλου ἐπί τινα γενόμεναι (Timaeus Lex. s.v.). The datives are usually construed with πείθοντες, and καταδέσμοις understood as the binding formulae “by which the seer compels the invisible powers to work his will” (Rohde Psyche^{2} II p. 88 note). But in the κατάδεσμοι which have been discovered it is the victim and not the god who is bound down; see e.g. CIG 538 (an Athenian inscription of about 380 B.C.)— καταδῶ Κτησίαν—καὶ Κλεοφράδην καταδῶ—καὶ τοὺς μετὰ Κτησίου ἅπαντας καταδῶ. This and other instances from leaden tablets found in graves are given by Wachsmuth Rhein. Mus. XVIII (1863) pp. 560 ff.: cf. also Marquardt Röm. Staatsverwaltung III p. 109 note 6. On this account I think it better to connect ἐπαγωγαῖς τισὶν καὶ καταδέσμοις with βλάψειν, exactly as in Laws 933 D ἐὰν δὲ καταδέσεσιν ἢ ἐπαγωγαῖς ἤ τισιν ἐπῳδαῖς ἢ τῶν τοιούτων φαρυακειῶν ὡντινωνοῦν δόξῃ ὅμοιος εἶναι βλάπτοντι—τεθνάτω. Plato is still alluding to the debasing forms of oriental superstition which had gained a footing in Greece in his day: see Foucart l. c. p. 172.

θεοὺς ‐‐ σφισιν ὑπηρετεῖν : whereas true religion consists in man's ὑπηρεσία τοῖς θεοῖς Euthyph. 13 D ff.

οἱ μὲν κτλ. : ‘some declaiming about the easiness of vice, how that’ etc. οἱ μὲν—ᾄδοντες recalls 364 A, while οἱ δέ refers to the ἀγύρται καὶ μάντεις of 364 B. The reference in the first case is as precise as possible: πάντες γὰρ ἐξ ἑνὸς στόματος ὑμνοῦσιν ὡς καλὸν μὲν ἡ σωφροσύνη τε καὶ δικαιοσύνη, χαλεπὸν μέντοι καὶ ἐπίπονον· ἀκολασία δὲ καὶ ἀδικία ἡδὺ μὲν καὶ εὐπετὲς κτήσασθαι, δόξῃ δὲ μόνον καὶ νόμῳ αἰσχρόν (364 A). Those who ὑμνοῦσιν ὡς—ἀκολασία—καὶ ἀδικία—εὐπετὲς κτήσασθαι can be accurately described as κακίας περὶ εὐπετείας ᾄδοντες, but scarcely by οἱ κακίας πέρι εὐπετείας διδόντες, because ‘to offer facilities for vice’ is not the same thing as to say that vice is easy. Stallbaum attempts to evade this difficulty by taking διδόντες as equivalent to διδόσθαι λέγοντες, but neither is ‘saying that facilities are offered for vice’ quite the same as ‘saying that vice is easy.’ It is also difficult to find another instance of the plural of εὐπέτεια. The verbal echoes seem to me very strongly in favour of περὶ—ᾄδοντες. For ᾄδοντες=‘harping on’ (like the ὑμνοῦσιν to which it refers) cf. Lys. 205 C ἃ δὲ ἡ πόλις ὅλη ᾄδει and 205 D ἅπερ αἱ γραῖαι ᾄδουσι (with reference to the proverbial γραῶν ὕθλος): the use of ᾄδειν in Laws 854 C is different, but akin. For the corruption of ᾄδοντες to διδόντες see Introd. § 5. The conjectures of Liebhold ( Fl. Jahrb. 1888 p. 107) and Zeller ( Arch. f. Gesch. d. Phil. II p. 694) κακίας πέρι εὐπετείας διελθόντες and κακίας πέρι εὐπέτειαν διδόντας have little in their favour.

ὡς τὴν ‐‐ ἔθηκαν . Hesiod OD. 287—289. ὡς is due to Plato: Hesiod has τὴν μέν τοι κτλ. For λείη the MSS of Hesiod read ὀλίγη: λείη (also in Laws 718 E, Xen. Mem. II 1. 20 and elsewhere) proves the existence of a different recension. Cf. G. E. Howes Harvard Studies in Cl. Philol. VI p. 165. The verses are partially quoted or referred to again in Laws 718 E, Prot. 340 D ; their influence is also seen in Phaedr. 272 C .

καί τινα ὁδὸν κτλ. : Hesiod OD. 290 μακρὸς δὲ καὶ ὄρθιος οἶμος ἐς αὐτὴν | καὶ τρηχὺς κτλ. The last two words account for the marginal addition καὶ τραχεῖαν in A.

λιστοὶ ‐‐ ἁμάρτῃ . See cr. n. The words are spoken by Phoenix to Achilles in Il. IX 497—501. Plato edits the lines to suit his own purposes. For λιστοί our text of Homer has στρεπτοί. The word λιστοί (though implied in ἄλλιστος, τρίλλιστος) does not occur elsewhere, a fact which is strongly in favour of its genuineness here. We must suppose that the recension which Plato used had λιστοί. The theology contained in these lines meets us continually in ancient literature: cf. also the words of the king in Hamlet III 3 “And what's in prayer but this twofold force To be forestalled ere we come to fall Or pardoned being down?” Plato expresses his dissent in Laws 716 E ff., 905 D: in Alc. II 149 E we read οὐ γὰρ οἶμαι τοιοῦτόν ἐστι, τὸ τῶν θεῶν ὥστε ὑπὸ δώρων παράγεσθαι οἶον κακὸν τοκιστήν.

βίβλων ‐‐ ἐγγόνων . The allusion is to Orphic liturgies. Musaeus was the son of Selene, according to Philochorus quoted by the Scholiast on Ar. Frogs 1033 : cf. φαεσφόρου ἔκγονε Μήνης | Μουσαῖε in Abel Orphic. Fr. 4. Orpheus' mother was the Muse Calliope (Suidas s.v. Ὀρφεύς). There is no solid basis for the old view that ἔκγονος means ‘son,’ and ἔγγονος ‘grandson.’ The etymological form is ἔκγονος, but ἐκ‐ was often assimilated to ἐγ‐ before γ during the 4th century B.C., particularly in this word: cf. also ἐγγειτόνων etc. on Inscriptions. See Meisterhans^{3} p. 107. Elsewhere in the Republic ἔκγονος is the regular spelling.

καθ’ ἃς θυηπολοῦσιν : sacrificial liturgies. A θυηπολικόν is mentioned by Suidas (s.v. Ὀρφεύς) as one of the ‘works’ of Orpheus: see also Lobeck Aglaoph. p. 371 and Rohde Psyche^{2} II pp. 112, 113 notes

πόλεις : as for instance when Epimenides the Cretan purified Athens (see Grote III 85—89). Plato may be thinking of this event, which in defiance of chronology he placed ten years before the Persian wars ( Laws 642 D, E). Cf. also infra 366 A and Laws 909 B.

λύσεις ‐‐ καθαρμοί . λύσεις means ‘modes of absolution’ (Lobeck Aglaoph. p. 810): cf. 366 A οἱ λύσιοι θεοί and Arist. Pol. B 4 1262^{a} 32 τὰς νομιζομένας λύσεις. The Scholium on Ar. Frogs 1033 contains the remark: οὗτος (i.e. Musaeus) δὲ παραλύσεις καὶ τελετὰς καὶ καθαρμοὺς συντέθεικεν. For παραλύσεις Blaydes proposes λύσεις, while Rutherford reads περὶ λύσεις (apparently with the Ravenna Codex), inserting also on his own conjecture ποιήματα after συντέθεικεν. I have no doubt that the Scholiast wrote παρὰ λύσεις: ‘besides Absolutions, he has composed also τελεταί and καθαρμοί.’ καθαρμοί formed a distinct class of religious literature, and were written by Epimenides, Empedocles, and others: see Grote I p. 27 note 3.

παιδιᾶς ἡδονῶν : ‘pleasures of play.’ παιδιᾶς depends on ἡδονῶν, and is here used abstractly: cf. Thuc. III 38. 7 ἀκοῆς ἡδονῇ and (with Schneider) Paus. I 21. 7 θέας ἡδονήν. Madvig would eject ἡδονῶν, but without ἡδονῶν Plato would probably have written παιδιῶν (cf. Laws 829 B): other suggestions, such as καὶ παιδιᾶς καὶ ἡδονῶν, or καὶ παιδιᾶς διὰ ἡδονῶν, or καὶ παιδιῶν καὶ ἡδονῶν are open to graver objection. For παίζειν and the like in connexion with religious celebrations Stallbaum cites Hdt. IX 11 Ὑακίνθιά τε ἄγετε καὶ παίζετε and VIII 99 ἐν θυσίῃσί τε καὶ εὐπαθείῃσι: add Phaedr. 276 B , Laws 666 B. Plato's point is that atonement if it is made a pleasure and not a penance sets a premium on sin.

τελευτήσασιν ‐‐ τελετάς . The Orpheotelestae connected τελεταί with τελευτᾶν, sometimes on the ground assigned by Plato here, sometimes because they alleged that the sensations of dying resembled those of initiation into the great mysteries (Plut. Frag. de An. 725). This and other ancient derivations are given by Lobeck Aglaoph. pp. 124, 126, 172. For περιμένει Cobet needlessly conjectures περιμένειν.

τιμῆς i. q. τοῦ τιμᾶν. Cf. (with J. and C.) 359 C above.

τί ‐‐ ποιεῖν . The subject to ποιεῖν is ταῦτα πάντα—λεγόμενα: ψυχάς is its secondary object. Cf. infra 367 B τί ποιοῦσα ἑκατέρα τὸν ἔχοντα κτλ. and 367 E. This view, which Schneider also holds, is better than to make ψυχάς subject to ποιεῖν and ταῦτα πάντα κτλ. dependent on ἀκουούσας.

ἐπιπτόμενοι . The image, as Jowett remarks, suggests a bee gathering honey: cf. Ion 534 B λέγουσι γὰρ—οἱ ποιηταὶ ὅτι ἀπὸ κρηνῶν μελιρρύτων ἐκ Μουσῶν κήπων τινῶν καὶ ναπῶν δρεπόμενοι τὰ μέλη ἡμῖν φέρουσιν ὥσπερ αἱ μέλιτται, Simon. Fr. 47 ὁμιλεῖ δ’ ἄνθεσιν (viz. the poet) ὧτε μέλισσα ξανθὸν μέλι κηδομένα and Pind. Pyth. X 53 f.

πότερον δίκᾳ ‐‐ ἀναβάς . The fragment (which appears tolerably often in ancient citations) is restored as follows by Bergk ( Fr. 213) Πότερον δίκᾳ τεῖχος ὕψιον | ἢ σκολίαις ἀπάταις ἀναβαίνῃ | ἐπιχθονίων γένος ἀνδρῶν | δίχα μοι νόος ἀτρέκειαν εἰπεῖν. It is, I think, unlikely that θεσπέσιος βίος and κύριον εὐδαιμονίας below “si non a Pindaro, certe ex poetis petita sunt” (Bergk).

ἐὰν καὶ μὴ δοκῶ has been commonly altered to ἐὰν μὴ καὶ δοκῶ on the suggestion of Dobree and Boeckh (with a few inferior MSS): but the text is sound. We are dealing with ταῦτα πάντα— λεγόμενα κτλ.; and it has not been said that it is useless to be just, unless one is also believed to be just ( ἐὰν μὴ καὶ δοκῶ). This would imply that it is useful to be just, if one is also considered just; but what has been urged is that Justice is in itself never advantageous, although its εὐδοκιμήσεις (363 A) are: see 358 C, E, 360 C ( οὐδεὶς ἑκὼν δίκαιος, ἀλλ’ ἀναγκαζόμενος, ὡς οὐκ ἀγαθοῦ ἰδιᾳ ὄντος), 362 A ( οὐκ εἶναι δίκαιον, ἀλλὰ δοκεῖν δεῖ ἐθέλειν). The words ἐὰν καὶ μὴ δοκῶ mean ‘if I also seem unjust,’ for οὐ δοκῶ δίκαιος εἶναι, not δοκῶ οὐ δίκαιος εἶναι, is the Greek idiom. This meaning suits exactly. What has to be established is that δοκεῖν prevails over εἶναι in human life ( οὐκοῦν—βιᾶται). The proof is as follows. To be just and seem unjust is misery (see 361 E): to be unjust, and seem just is bliss (see 362 A, C): therefore δοκεῖν is everything, and ἐπὶ τοῦτο τρεπτέον ὅλως.

φασιν . Is τὰ λεγόμενα the subject? or is the sentence an anacoluthon? (“nam quo modo res ipsa comparata sit, nescio: quae quidem vulgo dicuntur, talia sunt, ut iusto mihi commodi quicquam fore negetur” Schneider). The latter view is the more likely. Similar anacolutha are cited by Engelhardt Anac. Pl. Spec. III p. 40.

οἱ σοφοί . Simonides ( σοφὸς γὰρ καὶ θεῖος ἀνήρ I 331 E) Fr. 76 B ergk. Plato himself sets no small store by a good name (coupled with virtue) in Laws 950 C.

πρόθυρα ‐‐ σχῆμα : ‘as my porch and trappings.’ The mixture of metaphors is thoroughly Platonic: cf. VII 527 D note With σχῆμα (any kind of external or adventitious means of impressing others or hiding one's own deficiencies) cf. Gorg. 511 E περιπατεῖ ἐν μετρίῳ σχήματι.

σκιαγραφίαν (‘perspective drawing’ VII 523 B, X 602 D) with its cognate words is continually used by Plato of things unreal, counterfeit, illusory: cf. infra IX 583 B note, 586 B al., and Wohlrab on Theaet. 208 E .

τοῦ σοφωτάτου κτλ. Archilochus seems to have canonized the fox as the embodiment of cunning in Greek literature: fragments are preserved of at least two fables of his in which the fox appears (86—88 and 89 ed. Bergk). In the second (89. 5, 6) occur the lines τῷ δ) (sc. πιθήκῳ) ἆρ’ ἀλώπηξ κερδαλέη συνήντετο | πυκνὸν ἔχουσα νόον. The κερδαλέαν καὶ ποικίλην of Plato corresponds in meaning to κερδαλέη—πυκνὸν ἔχουσα νόον, and may have ended one of the iambics in this or another Archilochean fable: it is at all events clear that they are from Archilochus. ‘The crafty and subtle fox of Archilochus’ means simply ‘the crafty and subtle fox of which Archilochus speaks’: the rest of the imagery is due to Plato. With the general sentiment cf. St Matth. vii 15 ἔρχονται πρὸς ὑμᾶς ἐν ἐνδύμασι προβάτων (this is the σκιαγραφία ἀρετῆς), ἔσωθεν δέ εἰσιν λύκοι ἅρπαγες: with ἑλκτέον ἐξόπισθεν (opposed to πρόθυρα μὲν καὶσχῆμα) Milton Samson Agonistes 358— 360 “Why are his gifts desirable, to tempt Our earnest prayers, then, given with solemn hand As graces, draw a scorpion's tail behind?” Unnecessary difficulty has been caused by an erroneous gloss of Timaeus ( τὴν ἀλωπεκῆν· τὴν πανουργίαν), which seems to imply that he read ἀλωπεκῆν ‘fox's skin’ for ἀλώπεκα in this passage. Ruhnken (followed by Ast and Stallbaum) while retaining ἀλώπεκα explained it of the fox's skin; but it would be pointless to ‘drag behind a fox's skin.’ With ἀλώπεκα—‘fox’ for ‘foxiness’—cf. infra 382 D ποιητὴς—ψευδὴς ἐν θεῷ οὐκ ἔνι , Phaed. 77 E , and the well-known “astutam vapido servas sub pectore vulpem” Persius V 117.

ἀλλὰ γάρ ‘at enim,’ like ἀλλὰ δή (infra D, X 600 A al.), introduces an objection: cf. infra 366 A al.

οὐδὲ γὰρ ‐‐ μεγάλων : an audacious application of the proverb χαλεπὰ τὰ καλά .

ὡς ‐‐ φέρει . For ὡς we might expect ᾗ (Ficinus has quâ). ταύτῃ must be taken as referring to what precedes, though further explained by ὡς—φέρει. ἴχνη and φέρει shew that the metaphor is still the ὁδὸς βίου. The words ἴχνη φέρει may be from Archilochus. For the sentiment cf. III 394 D.

ξυνωμοσίας ‐‐ ἑταιρίας . An allusion to the political life of Athens: cf. Ap. 36 B , Theaet. 173 D , Thuc. VIII 54 ξυνωμοσίας, αἵπερ ἐτύγχανον πρότερον ἐν τῇ πόλει οὖσαι ἐπὶ δίκαις καὶ ἀρχαῖς. In the Laws, Plato would suppress all such secret clubs and cabals with a strong hand: see 856 B ff. The πειθοῦς διδάσκαλοι mentioned presently are the Sophists.

ὡς for ὥστε (except in idiomatic phrases like ὡς ἔπος εἰπεῖν, ὥς γε ἐντεῦθεν ἰδεῖν) is a curious archaism, tolerably frequent in Xenophon (e.g. Cyrop. I 2. 8, V 2. 5, VI 4. 16, VIII 5. 1 and 7. 27), but almost unexampled in Plato. The Protagoras (330 E) furnishes an instance with οὕτως preceding (cf. Xen. Cyr. IV 2. 13 ). ὡς in Phaed. 108 E is perhaps to be explained in the same way: cf. also Alc. II 141 B and Symp. 213 B παραχωρῆσαι γὰρ τὸν Σωκράτη ὡς ἐκεῖνον καθίζειν. See also on ὡς δή in I 337 C. As βιάζομαι can be followed by the simple infinitive, it might seem preferable to connect ὡς πλεονεκτοῦντες as a participial explanatory clause either with βιασόμεθα or with δίκην μὴ διδόναι (‘not to be punished for aggrandisement’); but the first alternative gives a wrong sense to πλεονεκτοῦντες, and the second involves too harsh an inversion.

οὐκοῦν κτλ. Cf. Laws 885 B θεοὺς ἡγούμενος εἰναι κατὰ νόμους οὐδεὶς πώποτε οὔτε ἔργον ἀσεβὲς εἰργάσατο ἑκὼν οὔτε λόγον ἀφῆκεν ἄνομον, ἀλλὰ ἓν δή τι τῶν τριῶν πάσχων, ἢ τοῦτο ὅπερ εἶπον οὐχ ἡγούμενος, ἢ τὸ δεύτερον ὄντας οὐ φροντίζειν ἀνθρώπων, ἢ τρίτον εὐπαραμυθήτους εἶναι θυσίαις τε καὶ εὐχαῖς παραγομένους. These three classes of heretics are severally refuted in 886 A—899 D, 899 D— 905 D, 905 D—907 B. It is clear both from this passage and from the Laws that the air was full of such heresies in Plato's day. The first was doubtless fostered by the sceptical attitude of Protagoras— περὶ μὲν θεῶν οὐκ ἔχω εἰδέναι οὔθ’ ὡς εἰσὶν οὔθ’ ὡς οὐκ εἰσίν (ap. D. L. IX 51): for the second cf. Aesch. Ag. 369 —372 οὐκ ἔφα τις | θεοὺς βροτῶν ἀξιοῦσθαι μέλειν | ὅσοις ἀθίκτων χάρις | πατοῖθ’ · ὁ δ’ οὐκ εὐσεβής: the third—the most pernicious of all, according to Plato Laws 948 C—furnished the raison d'être of a degenerate priesthood.

τί καὶ ἡμῖν κτλ. ‘If the gods do not care for us, why should we in our turn ( καί) care’ etc. For the text see cr. n. and App. III.

ἀκηκόαμεν ‐‐ ποιητῶν . The first ἤ is ‘or’ and the second ‘than.’ In λόγων Plato may be thinking inter alia of the works of early λογογράφοι like Pherecydes, who wrote genealogies of gods and heroes in prose; but there is no occasion to change λόγων into λογίων with Muretus. γενεαλογησάντων ποιητῶν refers to Homer and the Hesiodic and Orphic theogonies.

θυσίαις ‐‐ ἀγανῇσιν : see 364 D.

ἀπό : ‘from the proceeds of.’ Cf. Laws 906 C, D τοῦτον δὴ τὸν λόγον ἀναγκαῖον λέγειν τὸν λέγοντα ὡς εἰσὶ συγγνώμονες ἀεὶ θεοὶ τοῖς τῶν ἀνθρώπων ἀδίκοις καὶ ἀδικοῦσιν, ἂν αὐτοῖς τῶν ἀδικημάτων τις ἀπονέμῃ, καθάπερ κυσὶ λύκοι τῶν ἁρπασμάτων σμικρὰ ἀπονέμοιεν, οἱ δὲ ἡμερούμενοι τοῖς δώροις συγχωροῖεν τὰ ποίμνια διαρπάζειν.

ὑπερβαίνοντες καὶ ἁμαρτάνοντες are subordinate to λισσόμενοι: “by praying when we transgress and sin, we shall persuade them,” etc. There is again a reference to λισσόμενοι ὅτε κέν τις ὑπερβήῃ καὶ ἁμάρτῃ quoted in 364 E. The position of the participles is justified by the allusion to this line.

ἢ ‐‐ ἤ . It was a common Greek belief that the sins of the fathers are visited upon the children: see the passages cited by Nägelsbach Nachhom. Theol. pp. 34 ff. If we take Plato at his word, Adimantus represents this vicarious punishment as extending even to the other world.

ἢ παῖδες παίδων . Baiter conjectures <ἢ παῖδες> ἢ παῖδες παίδων, and so I formerly printed. But παῖδες παίδων means little more than ‘descendants’ (cf. Laws 927 B), and the text may stand. Similarly in Ruskin Modern Painters Ch. 1 “all those labours which men have given their lives and their sons' sons' lives to complete.”

ὦ φίλε ‐‐ λογιζόμενος . ὦ φίλε is the objector who urges ἀλλὰ γὰρ—παίδων. In φήσει Plato recurs to the singular of 365 B λέγοι γὰρ ἂν κτλ. λογιζόμενος is not ‘reasoning,’ but ‘making his calculation,’ ‘calculos subducens’: such a man's morality is nothing but a balancing of profit and loss. Hermann's devotion to Paris A led him to conjecture ἀλλ’ ὠφελήσουσιν ἁγνιζομένους αἰ τελεταί rather than admit a simple case of omission arising from homoioteleuton: see cr. n. Vermehren proposes ἀλλ’ ὠφελήσουσιν αἱ νομιζόμεναι τελεταί ( Plat. Stud. p. 90), but we should certainly follow Π here. See also Introd. § 5.

λύσιοι : ‘givers of absolution’: cf. 364 E. Certain Chthonian deities of the Orphic theology are meant, such as Hecate, Demeter, Dionysus λύσιος or λυσεύς, and above all Ζεὺς μειλίχιος. See Lobeck Aglaoph. p. 303.

θεῶν παῖδες : e.g. Musaeus and Orpheus ( Σελήνης τε καὶ Μουσῶν ἔγγονοι 364 E). Madvig's rejection of οἵ (so also Ficinus) before ταῦτα in the last clause seriously impairs the rhythm of the sentence.

ἄκρων . ἄκρος was a fashionable expression to apply to the élite of any profession or art: cf. Theaet. 152 E τῶν ποιητῶν οἱ ἄκροι τῆς ποιήσεως ἑκατέρας, Pol. 292 E , supra 360 E, infra III 405 A, V 459 B.

ὡς δή τοι : see on I 337 C.

θείᾳ φύσει ‐‐ ἐπιστήμην . θείᾳ φύσει means a disposition which is good by divine grace or nature, not as the result of knowledge or compulsion. The virtue of such men is θείᾳ μοίρᾳ παραγιγνομένη ἄνευ νοῦ ( Men. 99 E ): they are ἄνευ ἀνάγκης, αὐτοφυῶς, θείᾳ μοίρᾳ ἀγαθοί ( Laws 642 C), resembling Wordsworth's “Glad Hearts! without reproach or blot, Who do thy work and know it not.” Cf. VI 493 A note ἐπιστήμην is scientific knowledge of the good in the Socratic, not yet in the Platonic, sense.

οὐδεὶς ἑκὼν δίκαιος gives the lie to the Socratic οὐδεὶς ἑκὼν ἄδικος: cf. 360 C. For ἀνανδρίας below see on 359 B.

ἡρώων . J. and C. think “Plato is referring to well-known tales and maxims, which the poets and logographers had put into the mouths of ancient heroes.” It is simpler to understand the expression of Orpheus, Musaeus, and other θεῶν παῖδες, ποιηταὶ καὶ προφῆται τῶν θεῶν γενόμενοι: see 366 B note So also Dreinhöfer Plato's Schrift üb. d. Staat nach Disposition u. Inhalt p. 2 note 16.

ἄλλως ἤ . Praise of the δόξαι of Justice is somewhat inaccurately spoken of as praise of justice itself: but it is unnecessary to insert διά (with Richards) before δόξας. Cf. 367 D τῶν μὲν ἄλλων ἀποδεχοίμην ἂν οὕτως ἐπαινούντων δικαιοσύνην καὶ ψεγόντων ἀδικίαν, δόξας τε περὶ αὐτῶν καὶ μισθοὺς ἐγκωμιαζόντων καὶ λοιδορούντων.

ἰδίοις : see on 363 E.

ἀλλ) αὐτὸς ‐‐ ξύνοικος ᾖ . This thesis is developed and elaborated in Gorg. 472 D —481 B.

ὑπέρ is here little if anything more than περί, cf. Laws 777 A ὑπὲρ τοῦ Διὸς ἀγορεύων. This usage, which appears on Inscriptions after 300 B.C. (Meisterhans^{3} p. 222), is very rare in Plato. It occurs occasionally in the Attic orators, especially with λέγειν, and is tolerably common in Polybius and later Greek: see Stephanus-Hase Thes. s. v. ὑπέρ and Jannaris Hist. Gr. Gr. § 1685. I do not think we are justified in translating (with Tucker) ‘on behalf of their view of the relations of justice and injustice.’

κατατείνας : 358 D note

ἀλλότριον ἀγαθόν : I 343 C note

ὡμολόγησας : 358 A.

πολὺ δὲ μᾶλλον . The sequence of δέ after τε is frequent in Plato with δὲ καί, εἰ δὲ βούλει, τί δέ, ἔτι δέ, μέγιστον δέ, τὸ δὲ κεφάλαιον and the like. For a classified list of examples see Hoefer de part. Plat. pp. 15—17.

ἀκούειν is added to Glauco's list (357 C) by Adimantus, who is also responsible for the exaggeration πολὺ μᾶλλον.

καὶ ‐‐ δή with ὑγιαίνειν marks it as different in kind from the other examples: cf. (with J. and C.) Men. 87 E καὶ πλοῦτος δή and infra 373 A.

γόνιμα : i.q. γνήσια, but more forcible: cf. Theaet. 151 E , Ar. Frogs 96 .

καὶ ἀδικία βλάπτει . The sense is: καὶ ψέγε τοῦτ’ αὐτὸ ἀδικίας ὃ αὐτὴ δἰ αὑτὴν τὸν ἔχοντα βλάπτει. Hartman would cancel the words, needlessly, although the zeugma is bolder than usual. For the stylistic effect cf. ἀδικία δ’ ἐπαινεῖται 358 A above.

ἀποδεχοίμην and ἀνασχοίμην are equally good Greek (cf. Prot. 339 D , Phaed. 92 A , E al.), but as ἀπο‐ is supported by both Α and Π, it is more probable that the error lies in ‐σχοίμην than in ἀπο‐, especially as ἀποδεχοίμην is found also in the margin of A. The ἀπο‐ is at least as old as the Scholium, which mentions the two readings ἀποσχοίμην and ἀνασχοίμην. The latter is an obvious correction of ἀποσχοίμην, and has survived in Ξ and a few inferior MSS besides.

ὦ παῖδες ἐκείνου τοῦ ἀνδρός . This curious phrase occurs once again in Plato viz. Phil. 36 D , where Protarchus is addressed in the words ὦ παῖ ἐκείνου τἀνδρός. Philebus has withdrawn from the discussion, his part in which he has bequeathed to Protarchus, who is therefore playfully called his son. That this is the meaning appears from Phil. 11 A , B, 11 C δέχει δὴ τοῦτον τὸν νῦν διδόμενον, ὦ Πρώταρχε, λόγον; Ἀνάγκη δέχεσθαι· Φίληβος γὰρ ἡμῖν ὁ καλὸς ἀπείρηκεν, 12 A, 16 B, 19 A: cf. also 15 C and 28 B. In precisely the same way Glauco and Adimantus are the ‘children of Thrasymachus.’ They are διάδοχοι τοῦ λόγου as appears from 357 A, 358 B ( ἐπανανεώσομαι τὸν Θρασυμάχου λόγον), 367 A and 367 C, as well as from the substance of their arguments. This image is in fact one of the links by means of which Plato binds the dialogue together: as Polemarchus is heir to Cephalus (331 E), so Glauco and Adimantus are heirs to Thrasymachus. In explaining ἐκείνου τοῦ ἀνδρός of Thrasymachus, Stallbaum is therefore not “ridiculous” (as J. and C. assert) but right. See my article in Cl. Rev. X p. 237.

ὁ Γλαύκωνος ἐραστής may be Critias, as Schleiermacher supposed; but there is no evidence in support of the conjecture: see Bergk Poet. Lyr. Gr.^{4} II p. 283.

τὴν Μεγαροῖ μάχην : perhaps in 409 B.C.: see Diod. Sic. XIII 65. If so, Plato is guilty of a slight anachronism, supposing that the scene of the dialogue is laid in 410. See Introd. § 3.

παῖδες ‐‐ ἀνδρός . By Ἀρίστωνος, the author of the line of course meant Aristo, father of Glauco and Adimantus; but Ἀρίστων suggests ἄριστος (cf. IX 580 B) and the pun conveys a friendly, if halfironical, compliment to ‘his excellency’ Thrasymachus, whose παῖδες (so far as the argument is concerned) Glauco and his brother are: see on ὦ παῖδες above. In Symp. 174 B , when inviting Aristodemus to come as an uninvited guest to sup with Agathon, Socrates indulges in a similarly playful pun: ἕπου τοίνυν, ἔφη, ἵνα καὶ τὴν παροιμίαν διαφθείρωμεν μεταβάλλοντες, ὡς ἄρα καὶ ἀγαθῶν ἐπὶ δαῖτας ἴασιν αὐτόματοι ἀγαθοί. (The διαφθορά consists in the substitution of ἀγαθῶν for δειλῶν, the form of the proverb which Plato had in view being αὐτόματοι δ’ ἀγαθοὶ δειλῶν ἐπὶ δαῖτας ἴασιν, as the Scholiast remarks. Arnold Hug is ill-advised in adopting Lachmann's suggestion to read Ἀγάθων) i.e. Ἀγάθωνι for ἀγαθῶν: see Cl. Rev. X p. 238.) Other plays on proper names in Plato are collected by Riddell Digest pp. 250 f. In κλεινοῦ Stallbaum finds a ‘lusus facetus’ on ἐκείνου; but this particular lusus (if it exists) is accidental and unmeaning.

θεῖον . The addition of τι (proposed by Herwerden) is unnecessary: cf. III 388 D note θεῖος is here used, like ἔνθεος, of inspiration: if the speaker does not understand or believe what he says, he is, like a rhapsodist or poet, nothing but the mouthpiece of the inspiring deity: cf. Phaedr. 245 A , Ion 533 E, 535 E— 536 D.

τί τέ ἐστιν ‐‐ ἔχει recalls the conclusion of Book I (354 B, C).

οἵανπερ ἂν sc. ἐποιησάμεθα, the verb being omitted as it frequently is with ὥσπερ ἂν εἰ.

ἕρμαιον ‐‐ τυγχάνει . I have followed Schneider in printing a colon before ἕρμαιον: for the sentence ἕρμαιον— τυγχάνει is not the grammatical apodosis to the εἰ clause, but a further result. The asyndeton with ἕρμαιον is the usual asyndeton of ampliative clauses. For the principle underlying the method of inquiry here enunciated, see Soph. 218 C ὅσα δ’ αὖ τῶν μεγάλων δεῖ διαπονεῖσθαι καλῶς, περὶ τῶν τοιούτων δέδοκται πᾶσι καὶ πάλαι τὸ πρότερον ἐν σμικροῖς καὶ ῥᾴοσιν αὐτὰ δεῖν μελετᾶν, πρὶν ἐν αὐτοῖς τοῖς μεγίστοις and Pol. 286 A . (Contrast Phil. 48 B , where the opposite course is recommended.) In the special case of the State versus the Individual, the words ἐν σμικροῖς, ἐν ἐλάττοσιν are not applicable, but ἐν ῥᾴοσιν πρότερον δεῖ μελετᾶν is the essential part of the principle, and Justice in the State is ῥᾴων καταμαθεῖν (368 E) than in the Individual. Cf. also infra 377 C ἐν τοῖς μείζοσιν—μύθοις ὀψόμεθα καὶ τοὺς ἐλάττους. Illustrations from letters are tolerably frequent in Plato: cf. e.g. IV 402 A f., Theaet. 205 D —206 A, Pol. 277 E ff.

πρῶτον ‐‐ ἐπισκοποῦντες lays down the method to be pursued in the rest of the treatise, except in books V—VII, which are professedly a ‘digression,’ and X, which is of the nature of an epilogue. At each suc cessive stage in the exposition of his subject, Plato reminds us more or less explicitly of the method which he here proposes to follow:—at the end of the first sketch of a State 371 E; in connexion with the φλεγμαίνουσα πόλις 372 E; before entering on the theory of education 376 C, D and again in III 392 C, when he has finished the treatment of λόγοι; at Adimantus' objection IV 420 B, C; at the end of the picture of the just state IV 427 D ff.; in passing to Justice in the Individual IV 434 D ff.; at V 472 B ff., where the question is raised ‘Is this State possible?’; on beginning the account of the degenerate commonwealths and men in VIII 545 B; and finally when the whole argument draws to a head at IX 577 C.

τὴν τοῦ μείζονος ὁμοιότητα . Justice in the State is in fact to be used as a means of explaining Justice in the Individual, which is after all the real Justice: cf. IV 443 B ff. notes The relation between the two is that of a παράδειγμα and that which the παράδειγμα is intended to explain: see Pol. 278 C οὐκοῦν τοῦτο μὲν ἱκανῶς συνειλήφαμεν, ὅτι παραδείγματός γ) ἐστὶ τότε γένεσις, ὁπόταν ὂν ταὐτὸν ἐν ἑτέρῳ διεσπασμένῳ, δοξαζόμενον ὀρθῶς καὶ συναχθὲν περὶ ἑκάτερον ὡς συνάμφω μίαν ἀληθῆ δόξαν ἀποτελῇ; Φαίνεται. Plato has been severely blamed (as e.g. by Grote Plato III pp. 123 ff.) for representing the Commonwealth as the Individual “writ large.” Plato, however, laid stress upon this view, as tending to cement the union between the citizen and the State, which was rapidly dissolving in his day. This is well brought out by Krohn Plat. Frag. p. 5. Cf. also Pöhlmann Gesch. d. antik. Kommunismus etc. pp. 146 ff.

εἰ γιγνομένην ‐‐ ἀδικίαν . This would lead us to expect that we are to discover Justice and Injustice in the same State. In the sequel we find Justice only in the Ideal City: it is the degenerate Cities of VIII and IX that furnish the picture of Injustice. Plato does not expressly announce his change of plan till IV 420 B, C: ᾠήθημεν γὰρ ἐν τῇ τοιαύτῃ μάλιστα ἂν εὑρεῖν δικαιοσύνην καὶ αὖ ἐν τῇ κάκιστα οἰκουμένῃ ἀδικίαν—νῦν μὲν οὖν—τὴν εὐδαίμονα πλάττομεν—αὐτίκα δὲ τὴν ἐναντίαν σκεψόμεθα. The discrepancy must, I think, be admitted (see Krohn Pl. St. p. 32, and Kunert die doppelte Recens. d. Pl. St. pp. 10 ff.), but such corrections and developments of plan are characteristic of the dialogue as a form of literature, and do not establish the theory of a double recension of the Republic. Cf. Grimmelt de reip. Pl. comp. et unit. p. 19, and Westerwick de Rep. Pl. pp. 43—45.

γίγνεται ‐‐ πόλις κτλ. The present episode is ostensibly an historical account of the genesis of society, and from this point of view should be compared with Laws III 676 A ff. Some of the features are derived from an analysis of the industrial basis of society as it exists in civilised times: others (see 372 B—D), are semi-mythical and idyllic, recalling pictures of the golden age such as we find in Pol. 269 C ff., and in the caricatures of the comedians (e.g. ap. Athen. VI 267 E ff.). But the prevailing atmosphere is not historical or legendary, but idealistic (note δεῖ in 369 E and elsewhere), and Plato's πρώτη πόλις (Arist. Pol. Δ 4. 1291^{a} 17) should primarily be regarded as—in its essential features—a preliminary and provisional description of the industrial foundation on which the higher parts of his own ideal city are to rest. Cf. also on 372 B, D, Rettig Proleg. in Plat. remp. p. 42 and Steinhart Einleitung p. 156.

τυγχάνει as a mere copula is very rare in Attic prose, and it would be easy here to insert ὤν after πολλῶν: see Porson on Eur. Hec. 782 . In the Platonic dialogues this usage recurs in Phaedr. 263 C , Gorg. 502 B , Alc. I 129 A, 133 A, Hipp. Mai. 300 A, Laws 918 C, Tim. 61 C , nor is it possible in the last three examples to account for its omission by lipography. The idiom occurs in Sophocles and Euripides, once in Aristophanes ( Eccl. 1141), and (though condemned by Phrynichus) must also be admitted (though rarely) in prose: see the instances cited by Blaydes on Ar. (l.c.) and cf. Rutherford's New Phrynichus p. 342.

πολλῶν ἐνδεής . In the account of the genesis of society given in the Laws (676 A—680 E), more stress is laid on the social instinct of man: in Prot. 322 B ff the operating cause is man's defencelessness against wild beasts. Grote ( Plato III p. 139 note) censures Plato for not mentioning the “reciprocal liability of injury” among the generative causes of civic life; but this (as well as assistance against external aggression) is hinted at in βοηθούς.

ἄλλος ‐‐ χρείᾳ . The words are short for ἄλλος ἄλλον, τὸν μὲν ἐπ’ ἄλλου, τὸν δ’ ἐπ’ ἄλλου χρείᾳ (for the omission of τὸν μέν cf. Prot. 330 A , Theaet. 181 D al.): ‘one taking to himself one man, another another—the one man for one, the other for another purpose.’ Essentially the same meaning would no doubt be conveyed without τὸν δ’ ἐπ’ ἄλλου, which Herwerden following two inferior MSS would omit; but the fuller form of expression is chosen in order, I think, to prepare us for the principle of ‘One man, one work’ to be presently enunciated.

ταύτῃ τῇ ξυνοικίᾳ . Stallbaum rightly regards the sentence as an anaco luthon, the antecedent to ταύτῃ being the words from παραλαμβάνων to βοηθούς. If the subject to ἐθέμεθα (a gnomic aorist) were ἄλλος—δεόμενοι—ἀγείραντες, we should probably have had παραλαμβάνοντες for παραλαμβάνων: and besides, Plato is not yet describing the particular city which we are ποιεῖν λόγῳ (infra line 19), but laying down the law as to the γένεσις of cities in general. For the anacoluthon see Engelhardt Anac. Pl. Spec. III p. 40.

τῶν περὶ τὸ σῶμα : neuter, not masculine; otherwise Plato would have written θεραπευτῶν (as in q and some other MSS).

ἀναγκαιοτάτη πόλις . Referring to this passage, Aristotle ( Pol. Δ 4. 1291^{a} 10—19) attacks Plato for making the end of his city not τὸ καλόν, but τὰ ἀναγκαῖα. No doubt, the end of this ‘first city’— so Aristotle calls it—is primarily τὰ ἀναγκαῖα; but Plato would reply that the cities of the farmers, the auxiliaries, and the rulers, are in reality one city, γινομένη μὲν τοῦ ζῆν ἕνεκεν, οὐσα δὲ τοῦ εὖ ζῆν (Arist. Pol. A 2. 1252^{b} 29. Cf. Laws 828 D δεῖ δὲ αὐτὴν καθάπερ ἕνα ἄνθρωπον ζῆν εὖ).

ἕνα ἕκαστον κτλ. Cf. Charm. 161 E δοκεῖ ἄν σοι πόλις εὖ οἰκεῖσθαι ὑπὸ τούτου τοῦ· νόμου τοῦ κελεύοντος τὸ ἑαυτοῦ ἱμάτιον ἕκαστον ὑφαίνειν καὶ πλύνειν, καὶ ὑποδήματα σκυτοτομεῖν, καὶ λήκυθον καὶ στλεγγίδα καὶ τἄλλα πάντα κατὰ τὸν αὐτὸν λόγον κτλ.;

οὕτω ῥᾷον ἢ )κείνως . οὕτω refers to the alternative which is more familiar, although mentioned first: cf. (with Ast) Xen. Mem. I 3. 13 τοῦτο τὸ θηρίον—τοσούτῳ δεινότερόν ἐστι τῶν φαλαγγίων ὅσῳ ἐκεῖνα μὲν ἁψάμενα, τοῦτο δὲ οὐδ’ ἁπτόμενον—ἐνίησί τι. On the corruption ῥᾴδιον for /ῥᾷον (also in Men. 94 E ) see Introd. § 5.

φύεται strikes the keynote of the City of Books II—IV. The first critic to lay sufficient stress on this point was Krohn: see Pl. St. pp. 59—62, where he collects the references to φύσις throughout Books I—IV. The City of II—IV is a κατὰ φύσιν οἰκισθεῖσα πόλις. What is meant by φύσις? Not inorganic Nature, but the ‘nature’ of a πόλις or aggregate of πολῖται, i.e. (as the unit in a city is the man) human nature, in other words, the nature of the human soul, which, according to Plato and Socrates, constitutes a man's true and proper individuality. It is not however human nature as it is, but as it ought to be, which is the foundation on which the Platonic State is built; so that, although the doctrine of transcendent Ideas is excluded from the first four books (see on III 402 C), Idealism at all events is present. See also Krohn Plat. Frage pp. 8—11, and (for the connotation of φύσις) Benn's article on ‘The Idea of Nature in Plato’ in Archiv f. Gesch. d. Phil. IX pp. 24 —49 and Pöhlmann l.c. pp. 110 ff.

ὅταν ‐‐ εἷς μίαν . This principle—the cardinal principle of the Republic, reiterated also with great emphasis in Laws 846 D—847 B—is deduced by Plato from φύσις, whose rule is specialization: cf. 370 C ὅταν εἷς ἓν κατὰ φύσιν—πράττῃ. Plato (as usual in the Republic) is thinking of Man's nature, one man being naturally fitted for one pursuit, another for another: cf. III 395 B, IV 433 A, 434 A, B. The principle of specialization had already been enunciated by Socrates: see e.g. Xen. Mem. III 9. 3 , 15, Cyrop. VIII 2. 5, 6. Aristotle widens it into a general law of Nature: οὐθὲν γὰρ ἡ φύσις ποιεῖ τοιοῦτον οἷον οἱ χαλκοτύποι τὴν Δελφικὴν μάχαιραν πενιχρῶς, ἀλλ’ ἓν πρὸς ἕν ( Pol. A 2. 1252^{b} 1 ff.). In its application to politics, the principle becomes in Plato's hands a weapon for attacking the foundations of Athenian democracy (see Gorg. 455 A —C), to which, in this respect, his own Ideal City was a kind of counterblast.

κάλλιον . Did Plato write καλλίω? κάλλιον γίγνεται may no doubt mean ‘are better made,’ which is fairly satisfactory in point of sense, but καλλίω forms a better balance to πλείω τε, and is more suited to καλόν just below. With ῥᾷον immediately following, the corruption would be easy. On the other hand the collocation καλλίω καὶ ῥᾷον is unpleasing, and it is probably safer to adhere to the MSS.

ἐπὶ τὸ ἀροῦν . See on 372 B.

αὐτὴν τὴν πόλιν : ipsam urbem: the city as opposed to the inhabitants ( τέκτονες, χαλκῆς etc.). Cf. 360 D note It is not necessary to adopt Hermann's conjecture αὖ for αὐτήν, or (with Hartman) to eject τὴν πόλιν.

σχεδόν τι ἀδύνατον . Plato nevertheless endeavours to secure this advantage in the Laws: see 704 A—705 B. Cf. Arist. Pol. H 5. 1326^{b} 26 ff.

ὧν ἐκεῖνοι δέονται . All exchange with foreign cities is to be in kind: money is used only for transactions within the city: see infra 371 C ff. Here again Plato is constructing his city κατὰ φύσιν: cf. Arist. Pol. A 9. 1257^{a} 28 ἡ μὲν οὐν τοιαύτη μεταβλητικὴ οὔτε παρὰ φύσιν οὔτε χρηματιστικῆς ἐστὶν εἷδος οὐδέν.

ὧν ἂν αὐτοῖς χρεία . αὐτοῖς is of course emphatic ( ipsis). For the rare omission of ᾗ cf. III 416 D and Schanz Nov. Comm. Pl. p. 33 with Cope's Rhetoric of Aristotle Vol. II p. 328.

ὧν ἂν δέωνται . ὧν is masculine in spite of ὧν ἐκεῖνοι δέονται just above. The reading of q ἐκείνοις ἄξουσιν, οἳ μεταδώσουσιν ὧν ἂν δέωνται is a free correction (after 371 B) intended to make ὧν neuter.

τῆς ‐‐ ἐργασίας is not the work of a seaman (as Jowett seems to suppose), but a special department of ἐμπορία, viz. ναυκληρία: see Arist. Pol. A 11. 1258^{b} 21 ff. The ναύκληρος owned a ship and conveyed passengers and cargo for payment (cf. Gorg. 511 D , E): he is frequently mentioned along with the ἔμπορος, e.g. Pol. 290 A ἐμπόρους καὶ ναυκλήρους καὶ καπήλους, Laws 831 E, Xen. Vect. 3. 4, 5. 3.

ὧν δὴ ἕνεκα . ὧν can hardly (as J. and C. suppose) refer to μεταδώσουσιν: it must denote the same objects as the previous ὧν. The meaning is ‘for the sake of which things we established the principle of community and founded a city.’ Cf. 369 C κοινωνοὺς—μεταδίδωσι δὴ ἄλλος ἄλλῳ κτλ.

νόμισμα ‐‐ ἕνεκα . Cf. Laws 742 A νόμισμα δ’ ἕνεκα ἀλλαγῆς τῆς καθ’ ἡμέραν. See also 370 E note Plato regards coined money as a necessary evil— the offspring, not of φύσις, but of νόμος (cf. Arist. Eth. Nic. V 8. 1133^{a} 30 ff. διὰ τοῦτο τοὔνομα ἔχει νόμισμα, ὅτι οὐ φύσει ἀλλὰ νόμῳ ἐστί and Pol. A 9. 1257^{b} 10 ff.), a mere conventional symbol, the private possession of which is denied to the highest classes of the State (III 416 D ff.).

οἱ ἀσθενέστατοι κτλ. Cf. Laws 918 A—920 C, where καπηλεία is confined by Plato to those ὧν διαφθειρομένων οὐκ ἂν γίγνοιτο μεγάλη λύμη τῇ πόλει (919 C).

καπήλους ‐‐ ἐμπόρους . Soph. 223 D τῆς μεταβλητικῆς οὐχ ἡ μὲν κατὰ πόλιν ἀλλαγή, σχεδὸν αὐτῆς ἥμισυ μέρος ὄν, καπηλικὴ προσαγορεύεται; Ναί. Τὸ δέ γε ἐξ ἄλλης εἰς ἄλλην πόλιν διαλλαττόμενον ὠνῇ καὶ πράσει ἐμπορική; Τί δ) οὕ;

ἀξιοκοινώνητοι : worthy of being admitted into the κοινωνία of our city. This explanation (Schneider's) is better than ‘worthy of one's society’ (L. and S.).

τὴν τιμὴν ταύτην . ταύτην is idiomatic for ταύτης: see I 333 B note

μισθωτοί . Plato does not admit slave labour in his city, unless perhaps in the persons of barbarians. The exclusion of slaves is also a touch of ‘Nature’: cf. Arist. Pol. A 3. 1253^{b} 20 τοῖς δὲ παρὰ φύσιν (sc. δοκεῖ) τὸ δεσπόζειν with Suse mihl and Hicks ad loc. If barbarians may be enslaved, it is because they are φύσει δοῦλοι: cf. V 469 B ff., with 470 C and Arist. Pol. A 2. 1252^{b} 9 ταὐτὸ φύσει βάρβαρον καὶ δοῦλον.

ἐν αὐτῶν ‐‐ ἀλλήλους . The reply is to the first question, not to the second: see on V 465 E. In so far as δικαιοσύνη can be said to exist in so elementary a state, Plato would have identified it with the performance by each class (farmers, artisans, etc.) of their own work and no more. This is the first view of δικαιοσύνη in the Republic: for the second see IV 432 ff., 441 D ff., and for the third or metaphysical VI 504 B note

ὑποδήματα . I have placed the mark of interrogation after ὑποδήματα, as it is only the present participles which belong to διαιτήσονται. ‘And when they have built themselves houses’ marks a fresh start, no longer interrogative, for which reason I have also departed from the usual punctuation after ἱκανῶς (in B) and πόλεμον (in C).

θρέψονται κτλ. The picture which Plato proceeds to draw represents the working of well-regulated ἐπιθυμία or appetite—the psychological groundwork of the third or lowest order in Plato's city. τὰ μέν is the wheaten meal ( ἄλευρα), τὰ δέ the barley-meal ( ἀλφιτα). Only the wheaten meal was (as a rule) baked ( πέσσειν or ὀπτᾶν) into loaves ( ἄρτοι): the barley-meal was “kneaded into a simple dough ( μάσσειν, whence μᾶζα), dried in a mould, and afterwards moistened with water and eaten” (Blümner, Gr. Privatalt. p. 218). μᾶζαι made of barley meal was the staple food of the common Greek: the wheaten loaf was a luxury. The double chiasmus ἄλφιτα, μάξαντες, μάζας )( ἄλευρα, πέψαντες, ἄρτους is noticeable: cf. Crito 47 C.

τὰ μὲν πέψαντες κτλ. The asyndeton (as usual) is ampliative. The punctuation in the text avoids the difficulty of the two verbs θρέψονται and εὐωχήσονται. Schneider places the colon before μάζας, but this is much less natural. For μάζας γενναίας, ‘noble bannocks’ (J. and C.), cf. (with Stallbaum) Laws 844 E τὰ γενναῖα σῦκα ἐπονομαζόμενα. κάλαμον is not ‘a mat of reeds’ (Jowett, with L. and S.), which would be much too artistic, but ‘reeds,’ κάλαμον being collective as in Arist. Hist. An. IX 36. 620^{a} 35; and τινα is contemptuous (cf. II 363 D note).

παραβαλλόμενοι is also contemptuous for the παρατιθέμενοι of civilised society: it suggests throwing food before animals (cf. 372 D).

στιβάδων : not ‘mattresses’ (L. and S.): why should they ‘strew’ mattresses? The whole point of the passage is that instead of reclining on manufactured couches they lie on natural ones of bryony and myrtle boughs: contrast 372 D. στρωννύναι στιβάδας is simply ‘to make couches of leaves’: cf. στορέσαι λέχος. The word μῖλαξ means bryony (as Schneider saw): cf. Sandys on Eur. Bacch. 107 χλοήρει μίλακι καλλικάρπῳ. The ‘yew’ of the English translators would make a sombre and lugubrious couch.

ἐπιπίνοντες . ἐπί means ‘after’: cf. Xen. Cyr. VI 2 28 μετὰ δὲ τὸν σῖτον εἰ οἶνον ἐπιπίνοιμεν. In Greek banquets there was little or no drinking during dinner. The conjecture ὑποπίνοντες (Stephanus-Hase Thes. s. v. ἐπιπίνω) is unnecessary.

ἢ πόλεμον . The origin of war is over-population (373 D).

ἄνευ ὄψου κτλ. ὄψον is meant by Glauco in its narrower sense of animal food (whether fish or flesh); Socrates on the other hand uses the word in its wider sense of anything eaten in addition to, or along with, bread, e.g. vegetables (see Blümner Gr. Privatalt. p. 223). A spirited and athletic Athenian like Glauco cannot tolerate a vegetarian diet: cf. 372 D.

ἑστιωμένους : sarcastic, with reference to εὐωχήσονται: ‘you call it feasting when they have nothing but dry bread!’ (J. and C.).

ἅλας ‐‐ ἑψήσονται . ‘Of course they will make salt and olives and cheese and vegetables whether wild’ ( βολβούς) ‘or cultivated’ ( λάχανα) ‘into such boiled dishes as can be prepared in the country.’ ἕψημα is not ‘something for boiling,’ but something boiled; and ἑψήσονται is used with two accusatives, one external ( ἅλας, &c.) and the other internal ( ἑψήματα). Plato hints that cookery in the country ( ἐν ἀγροῖς, cf. κατ’ ἀγρούς III 399 D) is inferior to that in the town. For the kind of dishes in question cf. Ath. II 64 E περὶ δὲ τῆς τῶν βολβῶν σκευασίας Φιλήμων φησὶ τὸν βολβόν, εἰ βούλει, σκόπει | ὅσα δαπανήσας εὐδοκιμεῖ, τυρὸν μέλι | σήσαμον ἔλαιον κρόμμυον ὄξος σίλφιον· | αὐτὸς δ’ ἐφ’ αὑτοῦ )στὶν πονηρὸς καὶ πικρός .

φηγούς : ‘acorns,’ not ‘beech-nuts’ (D. and V.): see Blaydes on Ar. Peace 1137 .

ὑποπίνοντες . Wine was sipped during dessert. ὑπο‐ in ὑποπίνοντες emphasizes the moderation already expressed in μετρίως: cf. Lys. 223 B ὑποπεπωκότες ἐν τοῖς Ἑρμαίοις. Dr Jackson connects πρὸς τὸ πῦρ with ὑπο πίνοντες, comparing IV 420 E, Ar. Ach. 751 al. This may be right, but the ordinary view seems to me somewhat more natural.

ὑῶν . The city of Pigs is supposed by Zeller^{4} II 1 pp. 325, 893, and Dümmler Antisthenica pp. 5 ff., Proleg. zur Pl. Staat p. 61, to be a contemptuous allusion to Antisthenes' ideal commonwealth (on which see Susemihl in Fl. Jahrb. 1887 pp. 207—214). This conjecture requires us to interpret Plato's first sketch of a State as wholly ironical and intended ‘to warn us against the false ideal of a Nature-City’ (Zeller l. c.). I agree with Henkel ( Stud. zur Gesch. d. Gr. Lehre vom Staat pp. 8 f.) in thinking that there is no solid ground for Zeller's theory. The πρώτη πόλις is not of course Plato's ideal republic, and his description of it is plentifully bestrewn with irony, but it is nevertheless the foundation on which his city is built, and, in point of fact, although some of its features are implicitly corrected or superseded in the sequel, it still remains on the whole, and as far as it goes, a not unpleasing picture of the life of the lowest stratum in Plato's city, and it is nowhere expressly cancelled or abolished. See also on 369 B and 372 E. The εὐχερὴς βίος ( Pol. 266 C ) of the πρώτη πόλις is fitly compared to that of pigs, the εὐχερέστατον γένος τῶν ὄντων (ib.); and it is appropriate that Glauco, who is nothing if not θυμοειδής ( Introd. § 2), should thus express his contempt for a life which hardly if at all rises above the level of ἐπιθυμία.

καὶ οἱ νῦν ἔχουσι : e.g. fish, flesh, fowl: see on 372 C. The words ἅπερ—ἔχουσι are to be taken with τραγήματα as well as with ὄψα. Glauco is thinking of delicacies like the preserved sorb-apples ( ὄα τεταριχευμένα) alluded to in Symp. 190 D . See Blümner Gr. Privatalt. p. 222 note 2.

τρυφῶσαν πόλιν . Krohn ( Pl. St. pp. 34, 72) thinks that Plato originally meant to look for ἀδικία in this τρυφῶσα πόλις: but see on 369 A.

ἀληθινὴ ‐‐ φλεγμαίνουσαν . There is a vein of irony in ἀληθινή: for the πρώτη τόλις is not the final form of Plato's city. The epithets τρυφῶσαν, φλεγμαίνουσαν are not however ironical (as Dümmler seems to hold Proleg. p. 62): see III 399 E.

εἰδ)αὖ ‐‐ ἀποκωλύει . I have adopted Richards' suggestion, and printed a comma after βούλεσθε, a colon before οὐδέν. The meaning is: ‘but if you wish it, let us contemplate also’ etc. The scribe in Paris A must have understood καὶ θεωρήσωμεν in the same way, for he assigns the words οὐδὲν ἀποκωλύει to Glauco. We are hardly justified in making θεωρήσωμεν the subjunctive after βούλεσθε, in the absence of other examples in which the subjunctive follows a dependent βούλει ( βούλεσθε). A possible view would be to take θεωρήσωμεν as= δεῖ θεωρῆσαι and construe ‘but if you wish it and we are to contemplate’ etc., cf. Crat. 425 D εἰ μὴ ἄρα δἠ (MSS δεῖ)— καὶ ἡμεῖς— ἀπαλλαγῶμεν (‘unless we too are to get quit’), and Postgate in Transactions of the Camb. Philol. Soc. III Pt. I pp. 50—55. But Richards' proposal is a better one.

ταῦτα ‐‐ τισιν . γάρ is introductory and means not ‘for’ but ‘well.’ τισιν contains a sly allusion to Glauco: cf. V 465 E, VI 504 C.

καὶ ὄψα δή . For δή see 367 C note

ἑταῖραι . G. W. Nitzsch ( Rhein. Mus. 1857, pp. 471 f.), Richter ( Fl. Jahrb. 1867, p. 141), Madvig, and Stallbaum take offence at the juxtaposition of ἑταῖραι and πέμματα and suggest respectively ἀθῆραι (apparently an error for ἀθάραι, cf. Ar. Plut. 673 ), ἐραῖα (= ἑψήματα in Schol. on 445 C), ἐσχαρῖται ‘panes delicati,’ and ἕτερα (with the following καί deleted),—conjectures which are altogether needless and refute one another. The text is successfully defended by Hug ( Hermes 1876, p. 254), who cites an exact parallel in Ar. Ach. 1090 —1092 κλῖναι, τράπεζαι, προσκεφάλαια, στρώματα, | στέφανοι, μύρον, τραγήμαθ’ , αἱ πόρναι πάρα, | ἄμυλοι πλακοῦντες, σησαμοῦντες, ἴτρια | (varieties of πέμματα). Cf. also Amphis ap. Ath. XIV 642 A οἶνος ἡδύς, ᾠά, σησαμαῖ, | μύρον, στέφανος, αὐλητρίς and infra III 404 D, IX 573 D note From these passages it may fairly be doubted whether Plato's mention of ἑταῖραι is in any way even παρὰ προσδοκίαν (as the Oxford editors suggest): for αὐλητρίδες were almost as common a feature at dessert as the cakes ( πέμματα) etc. which accompany them here: see e.g. Xen. Mem. I 5. 4 , Symp. 2. 1, Pl. Symp. 176 E , Prot. 347 D . Vahlen ( Index Lect. per sem. hib. 1875—6 Berol.) quotes also Catullus'

cenabis bene

si tecum attuleris bonam atque magnam

cenam non sine candida puella

et vino et sale et omnibus cachinnis

(13. 1 ff.).

καὶ τὴν ποικιλίαν . ποικιλία means variety of colour as e.g. in embroidery: cf. 378 C, III 401 A, Euthyph. 6 C . On the omission in A see Introd. § 5.

χρυσὸν καὶ ἐλέφαντα : with reference to chryselephantine statuary. Note that (according to Plato) the demand for decorative arts does not arise till the physical necessities of man are satisfied. Cf. Nettleship Lectures and Remains, II p. 73.

μείζονά τε αὖ τήν . τέ is ἀνακόλουθον (Hoefer de part. Pl. p. 14): for other instances in the Republic see V 463 D, VII 522 B, IX 575 A. In this passage Richter would change τε αὖ τήν into τοιαύτην, comparing 372 E; but the text is sound, and τοιαύτην would be quite wrong. αὐτὴν τὴν πόλιν (cf. 370 E), conjectured by Heller instead of αὖ τὴν πόλιν, is neat but needless.

πλήθους ἅ : i.e. πλήθους τούτων ἅ, as Ficinus understood the words. Stallbaum's alternative suggestion (that ἅ refers directly to ὄγκου and πλήθους) gives a poor sense. Cf. infra 373 E note

θηρευταὶ πάντες . The addition of πάντες shews that θηρευταί is used in a wide sense, including every variety of fishing as well as hunting: Laws 823 B θήρα γὰρ πάμπολύ τι πρᾶγμά ἐστί, περιειλημμένον ὀνόματι νῦν σχεδὸν ἑνί. πολλὴ μὲν γὰρ ἡ τῶν ἐνύδρων, πολλὴ δὲ ἡ τῶν πτηνῶν, πάμπολυ δὲ καὶ τὸ περὶ τὰ πεζὰ θηρεύματα. In Euthyd. 290 B —D, Soph. 219 E ff., and Laws (l.c.), Plato makes θηρευτική include ‘fishing for men’ e.g. in war, or by Sophists etc. This wider meaning clearly rests upon a Platonic—or rather Socratic (see Xen. Mem. II 6. 29 , quoted by J. and C.)—metaphor, and is not intended here. Cf. Benseler in Fl. Jahrb. 1881, pp. 236 ff. Aristotle on the other hand regards hunting as characteristic of the most primitive society ( Pol. A 8. 1256^{a} 35 ff.), and so too Plato himself in Laws 679 A.

ῥαψῳδοί ‐‐ ἐργολάβοι are the poet's servants. In Athens and elsewhere they formed regular guilds or σύνοδοι τῶν περὶ τὸν Διόνυσον τεχνιτῶν: cf. Arist. Probl. XXX 10. 956^{b} 11 οἱ Διονυσιακοὶ τεχνῖται. The ἐργολάβος contracted with the poet for the performance of his play, acting as a kind of financial agent or middleman between him and the σύνοδος to which he belonged. See Müller Bühnenalterthümer, pp. 392—414.

παιδαγωγῶν ‐‐ κουρέων . We infer that in the ‘healthy’ State fathers were παιδαγωγοί, mothers suckled ( τιτθῶν) and nursed ( τροφῶν) their own children, and the professional hair-dresser was unknown.

συβωτῶν . See on 372 B.

χρείαις . Cobet's χρείᾳ is not, I think, necessary. The plural (for which cf. 369 D al.) refers to the different occasions when we may require the help of doctors.

λέγομεν . λέγωμεν may be right, but the first hand of A was apt to err in these subjunctive forms ( Introd. § 5), and the Indicative is somewhat more natural here: cf. (with Schneider) 377 E ἀλλὰ πῶς δὴ λέγομεν καὶ ποῖα;

πολεμήσομεν . Stallbaum adds δή after πολεμήσομεν with some inferior MSS. The effect of its omission is to lay special stress on the first mention of πόλεμος in πολεμήσομεν, which should be pronounced with emphasis. Cf. IV 432 C, IX 583 C.

πολέμου ‐‐ γένεσιν . War then arises from the acquisition of territory and wealth: cf. Phaed. 66 C διὰ γὰρ τὴν τῶν χρημάτων κτῆσιν πάντες οἱ πόλεμοι ἡμῖν γίγνονται, where war is farther traced to the body and its desires, to satisfy which we seek to multiply our possessions. Cf. Arist. Pol. A 8. 1256^{b} 23 ἡ πολεμικὴ φύσει κτητική πως ἔσται.

ἐξ ὧν ‐‐ γίγνηται defines γένεσιν. War comes ἐξ ὧν i.e. ἐκ τούτων ὧν κτλ. ( ὧν for ἐξ ὧν, according to the usual Greek idiom, cf. Euthyph. 10 C , and III 402 A ἐν ἅπασιν οἷς ἔστι al.), from that which involves both cities and individuals in calamities, viz. from the desire of money. Cf. 373 B note and (for the sentiment Laws 870 A ff. ἡ τῶν χρημάτων τῆς ἀπλήστου καὶ ἀπείρου κτήσεως ἔρωτας μυρίους ἐντίκτουσα δύναμις διὰ φύσιν τε καὶ ἀπαιδευσίαν τὴν κακὴν κτλ. The love of money —so Plato held—is the root of all evil. This explanation is due to Schleiermacher; others (Schneider, Stallbaum, J. and C. as an alternative) refer ἐξ ὧν to war and the like=‘ex cuiusmodi rebus’ (Stallbaum). It is an objection to such a view that it makes Plato say that evils come from War (and the like), directly after he has declined to say anything of the sort ( μηδέν γέ πω — ἐργάζεται). Further, if ὧν referred to war, the sentiment would in itself be a platitude and almost deserve to be expunged from the text, as it is by Herwerden. On the other hand ἐξ ὧν—γίγνηται is on Schleiermacher's view quite consistent with μηδέν γέ πω— <*>ργάζεται, for although war arises from that which harms a State, in itself it may (and does) actually do good. Good in other words may come out of evil; which is exactly the principle on which Plato evolves his ideal city out of the τρυφῶσα πόλις. ὅταν γίγνηται (sc. κακά) is equivalent (as J. and C. remark) to ἑκάστοτε: cf. Phaed. 68 D φόβῳ μειζόνων κακῶν ὑπομένουσιν αὐτῶν οἱ ἀνδρεῖοι τὸν θάνατον <*>ταν ὑπομένωσιν.

ὅλῳ . Herwerden's conjecture μεγάλῳ seems to shew that he connected σμικρῷ with στρατοπέδῳ, but the meaning is ‘not by a small amount, but by a whole army.’ For the datives cf. IX 579 C note

αὐτοὶ οὐχ ἱκανοί ; Glauco speaks as an Athenian citizen-soldier. In making war a profession, and citizens synonymous with soldiers, Plato is laconizing. The language which Isocrates ( Archid. 81) applies to Sparta might in point of fact be used of Plato's State: τῶν Ἑλλήνων διενηνόχαμεν οὐ τῷ μεγέθε<*> τῆς πόλεως, οὐδὲ τῷ πλήθει τῶν ἀνθρώπων, ἀλλ) ὅτι τὴν πολιτείαν ὁμοίαν κατεστησάμεθα στρατοπέδῳ καλῶς διοικουμένῳ καὶ πειθαρχεῖν <*>θέλοντι τοῖς ἄρχουσιν. Cf. Grote Plato III pp. 176, 209.

ὡμολογοῦμεν : without εἶναι as in X 610 C ἀθανάτους τὰς ψυχὰς ὁμολογεῖν, and Soph. 246 E . The analogy of these cases shews that ἀδύνατον here is not neuter but masculine, agreeing with ἕνα. The reference is to 370 B.

ἀλλ’ ἄρα . As διεκωλύομεν is certainly interrogative, Ast conjectured ἆρα for ἄρα, but ἄρα ( nimirum) is regularly present in a fortiori arguments of this kind, either in the δέ clause ( Ap. 34 C , 37 C, D, Crito 46 D) or in both ( Crito 50 E, Prot. 325 B , C). In place of the second ἄρα is here written δὴ ( τὰ δὲ δὴ περὶ τὸν πόλεμον κτλ.). For the combination ἀλλ’ ἄρα cf. Soph. 243 E ἀλλ) ἄρα τὰ ἄμφω βούλεσθε καλεῖν ὄν; Ἴσως.

ἀλλὰ σκυτοτόμον . See cr. n. and Introd. § 5. The homoioteleuton as well as the presence of the clause ἵνα—γίγνοιτο is in favour of the genuineness of these words: and the construction itself, which requires ἐκελεύομεν or the like to be supplied out of διεκωλύομεν (see Heindorf on Gorg. 457 C and Kühner Gr. Gr. II p. 1072), is too idiomatic to have been readily invented by a scribe.

ἐφ’ ᾧ : with σχολὴν ἄγων (Schneider): cf. Ap. 36 D . The phraseology here recalls 370 B and C.

ἢ οὕτω ῥᾴδιον : singular in spite of the plural τὰ περί. Cf. (with Schneider) Hipp. Maior 299 A, Laws 708 D.

τε καὶ ὀργάνων is ejected by Herwerden, who is also inclined to denounce τὰ ὄργανα below. But it is just these words which “point the analogy: the weapons of the warrior are his tools.” (J. and C.) On similarly inadequate grounds τῶν νέων has been condemned in Euthyph. 3 A τοὺς τῶν νέων τὰς βλάστας διαφθείροντας: see my note ad loc. τινος ἄλλης μάχης below refers for example to ψιλοί or πελτασταί; the ἀσπίς (it should be remembered) was worn by the ὁπλίτης (whence ἀσπίδα μὲν λαβὼν—ὁπλιτικῆς). The necessity of special knowledge and training for success in war is insisted on by the historical Socrates in Xen. Mem. III 1.

φυλάκων . This is the first occurrence of φύλακες in the technical sense which it bears throughout the Republic. It is important to remember that the name includes not only the soldiers, but also—after they have been introduced— the rulers; when it becomes necessary to distinguish between the two classes, the former are called ἐπίκουροι (first named in III 414 B), the latter φύλακες παντελεῖς (III 414 B), τέλεοι φύλακες (IV 428 D) or the like, or more commonly ἄρχοντες (first alluded to in III 389 B, but not expressly separated off until 412 B ff., and finally and fully described only in Books VI and VII).

ὅσον γ’ ἂν δύναμις παρείκῃ . The phrase is not found elsewhere in Plato, although παρείκει is found with a personal subject ( ὁ θεός, θεοί) again in Theaet. 150 D , Laws 934 C. Herwerden would eject δύναμις (cf. Symp. 187 E καθ’ ὅσον παρείκει), but such a word is very unlikely to have been interpolated. δύναμις is simply ‘our powers’: the article is omitted as in the idiomatic κατὰ δύναμιν, εἰς δύναμιν.

σκύλακος . A play on σκύλαξ and φύλαξ is intended. Analogies from the animal kingdom were freely employed by the historical Socrates: for the dog in particular cf. Xen. Mem. IV 1. 3 καὶ τῶν κυνῶν τῶν εὐφυεστάτων, φιλοπόνων τε οὐσῶν καὶ ἐπιθετικῶν τοῖς θηρίοις, τὰς μὲν καλῶς ἀχθείσας ἀρίστας γίγνεσθαι—, ἀναγώγους δὲ γιγνομένας μᾳταίους τε καὶ μανιώδεις καὶ δυσπειθεστάτας. Cf. note on φύεται 370 A.

αἰσθανόμενον : ‘the moment he perceives.’ The present (where one might expect the aorist) emphasizes the rapidity with which pursuit follows upon sight.

ἀνδρεῖος . For ἀνδρεῖος applied to beasts cf. Isocr. 15. 211 εἰ περὶ τοὺς ἵππους καὶ τοὺς κύνας καὶ τὰ πλεῖστα τῶν ζῴων ὁρῶντες τέχνας ἔχοντάς τινας, αἷς τὰ μὲν ἀνδρειότερα, τὰ δὲ πραότερα, τὰ δὲ φρονιμώτερα ποιοῦσι, περὶ τὴν τῶν ἀνθρώπων φύσιν μηδεμίαν οἴονται τοιαύτην ηὑρῆσθαι παιδείαν κτλ. See also Lach. 196 D — 197 B and Arist. Eth. Nic. III 11. 1116^{b} 33 ff.

θυμοειδής . The technical term θυμοειδής is here for the first time used in the Republic. Plato probably inherited the word from Socrates (see Xen. Mem. IV 1. 3 τῶν τε ἵππων τοὺς εὐφυεστάτους, θυμοειδεῖς τε καὶ σφοδροὺς ὄντας κτλ.): in practice he employs it as the adjective corresponding to θυμός (see e.g. III 411 A, B), as ἐπιθυμητικός corresponds to ἐπιθυμία. The usual translation ‘spirited’ probably expresses the meaning as nearly as can be done by a single word. For a full discussion of the word reference may be made to P. Meyer ὁ θυμὸς ap. Arist. Platonemque (1876), whose conclusion (p. 65) is “ τὸν θυμὸν esse eam naturalem vim, qua ductus suam quisque propriam naturam explere studeat, quaque incitatus, quaecunque hanc naturam ipsi propriam tollere vel laedere conentur, fugiat, quae contra perfectiorem reddere possint, adpetat.” See also on IV 439 E.

ἄμαχόν ‐‐ ἀνίκητον . Ast may be right in supposing that Plato has in view the words of Heraclitus, often referred to in antiquity, θυμῷ μάχεσθαι χαλεπόν· ὅ τι γὰρ ἂν χρηίζῃ γίνεσθαι, ψυχῆς ὠνέεται ( Fr. 105 B ywater).

οὗ παρόντος ‐‐ ἀήττητος . Cf. Arist. Eth. Nic. III 11. 1116^{b} 26 ἰτητικὠτατον γὰρ ὁ θυμὸς πρὸς τοὺς κινδύνους.

ἐναντία γὰρ ‐‐ φύσις . Plato regarded this opposition as the fundamental antithesis of human character, and thought it a statesman's foremost duty to blend the θυμοειδές and πρᾶον harmoniously together: see Pol. 306 C — 311 C, infra III 410 B ff., VI 503 C, Theaet. 144 A , B.

ταῦτα ‐‐ ἔοικεν . Van Heusde ( Initia Phil. Plat. p. 471 note 1) somewhat hastily declares these words to be corrupt, and supplies ἀμφότερα ἔχειν after ταῦτα δέ. ταῦτα refers like τούτων simply to the two qualities πρᾶον and μεγαλόθυμον: ‘these’—meaning the combination of these as opposed to one of them—‘are apparently unattainable’: cf. VI 499 D οὐ γὰρ ἀδύνατος γενέσθαι, οὐδ’ ἡμεῖς ἀδύνατα λέγομεν.

ἐνοήσαμεν ‐‐ φύσεις . ἐνενοήσαμεν (with q) is read by most of the editors, quite unnecessarily, as Schneider shews. νοεῖν is not ‘putare,’ nor— I think—‘perpendere,’ but simply ‘animadvertere,’ ‘notice,’ as often. Such a meaning is peculiarly appropriate with ἴδοι following. Presently ἄρα is not ‘then’ (J. and C.), but ‘after all.’

τῷ φύλακι : not τῷ σκύλακι, as Groen van Prinsterer conjectured ( Plat. Prosop. p. 209). τῷ φύλακι of course depends on παρεβάλλομεν, and ἐν ᾧ is for ἐν τούτῳ ὃ.

αὐτῶν τὸ ἦθος . With αὐτῶν (unnecessary, but welcome, after τῶν γενναίων κυνῶν) cf. IV 428 A note

πρὸς μὲν ‐‐ τοὐναντίον . In Od. XVI 4—10 the dogs of Eumaeus do not bark at Telemachus, and Odysseus remarks (8, 9) Εὔμαἰ, ἦ μάλα τίς τοι ἐλεύσεται ἐνθάδ’ ἑταῖρος | ἢ καὶ γνώριμος ἄλλος, ἐπεὶ κύνες οὐχ ὑλάουσιν | ἀλλὰ περισσαίνουσι. See also Od. XIV 30, where they bark at the stranger Odysseus, and cf. Heracl. 115 (Bywater) κύνες καὶ βαύζουσι ὃν ἂν μὴ γινώσκωσι. In Aristotle similar characteristics are attributed to the lion: see Physiogn. 5. 809^{b} 34—36 μεγαλόψυχον καὶ φιλόνικον, καὶ πραΰ καὶ δίκαιον καὶ φιλόστοργον πρὸς ἃ ἂν ὁμιλήσῃ, and Hist. An. IX 44. 629^{b} 10—12.

πρὸς τῷ θυμοειδεῖ κτλ. There seems to be no other example in good Greek of προσγενέσθαι meaning ‘to become in addition’: but we may compare προσέσονται II 373 A, προσέχειν VII 521 D, προσείπωμεν X 607 B, and similar instances with other verbs. I formerly wrote φιλόσοφον for φιλόσοφος (‘that to the element of spirit nature should have added’— προσγενέσθαι, i.q. accessisse, cf. I 346 D —‘a philosophical temperament’). The accusative with infinitive has however a harsh effect. Herwerden cuts the knot by deleting the προς‐ of προσγενέσθαι.

ὅτι ‐‐ προπεπονθώς . Schneider justly observes that ὅτι is not likely to be an interpolation, and might easily have disappeared before ὅν, as it has in A (see cr. n.). In itself the presence of ὅτι is an improvement. For οὐδὲν δή v (supported also by Stobaeus Flor. 43. 149) reads οὐδέν, which may be right. Cobet's οὐδὲ ἕν is too emphatic.

οὐ πάνυ ‐‐ τὸν νοῦν : ‘I have hardly thought of the matter till now.’ μέχρι δεῦρο is more idiomatic than μέχρι τούτου in this sense, but Xen. Cyr. VIII 8. 9 and Dem. de Cor. 48 are closely analogous instances. The alternative rendering ‘my observation has hardly extended so far’ is (in view of οὔπ ω τοῦτο ἐθαύμασας;) less suitable.

ὡς ἀληθῶς φιλόσοφον . ὡς ἀληθῶς indicates that φιλόσοφον is to be taken in its etymological sense: cf. I 343 C note The dog shews ‘a love of knowledge’ because he loves the known, and hates the unknown. Brandt ( Zur Entwick. d. Pl. Lehr. v. d. Seelentheilen p. 10) ingeniously takes φιλόσοφον as= σοφὀν τοὺς φίλους: but the other interpretation is more natural and relevant. There is perhaps an allusion to the Cynics: see Schol. in Arist. ed. Brandis (Berlin 1836) 23^{b} 16 ff. τετάρτη δὲ (SC. αἰτία τοῦ κληθῆναι Κυνικοὺς) ὅτι διακριτικὸν ζᾧον ὁ κύων γνώσει καὶ ἀγνοίᾳ τὸν φίλον καὶ τὸν ἀλλότριον ὁρίζον· ὃν γὰρ γιγνώσκει, νομίζει φίλον εἶναι καὶ εἰ ῥόπαλον ἐπιφέροιτο, ὃν δὲ ἀγνοεῖ ἐχθρόν, καὶ εἰ δέλεαρ ἐπιφερόμενος εἴη. οὕτως οὗν καὶ οὗτοι τοὺς μὲν ἐπιτηδείους πρὸς φιλοσοφίαν φίλους ἐνόμιζον καὶ εὐμενεῖς ἐδέχοντο, τοὺς δὲ ἀνεπιτηδείους ἀπήλαυνον δίκην κυνῶν κατ’ αὐτῶν ὑλακτοῦντες, and Philoponus ib. 35^{a} 5—12. The Cynics were themselves very fond of pointing the moral from the lower animals to man (Dümmler Proleg. p. 58 note 2), and Plato here paints them not unkindly in colours of their own. It should be noted that throughout II—IV Plato uses φιλόσοφος and φιλοσοφία with less of an intellectual than of a moral connotation. In the earlier books the word is for the most part connected with a gentle considerate disposition or character, whether naturally implanted or the result of culture (cf III 410 E, 411 C, 411 E): in 407 C the sense is somewhat different. See Nettleship in Hellenica pp. 77—79, and Krohn Pl. St. p. 71. It is not until the latter part of Book V (473 B ff.) where Plato is proposing to enter on the third and final stage of his ideal city, viz. the κατάστασις τῶν ἀρχόντων, that the intellectual aspect of the word begins to predominate over the moral. Cf. IV 439 D note

φύσει is better taken with φιλόσοφον than with πρᾶος. Cf. 375 B.

ἆρά τι προὔργου κτλ. See on 368 E.

ἵνα μὴ ‐‐ διεξίωμεν . See cr. n. The omission in the text of A may be accidental (see Introd. § 5), but the sentence is certainly a difficult one. If the MSS are right, the meaning must be “For we do not want to be tedious,”— but συχνός is rather ‘lengthy’—“and we do not want to leave unsaid what is required for completeness” (J. and C., comparing for συχνός Theaet. 185 E , Phil. 23 B al.). The conjectures of Teuffel ( Rhein. Mus. 1850 p. 469) and Herwerden ( Mnem. N. S. XI p. 339)— ἵνα ἢ (so q) ἐῶμεν συχνὸν (so v) λόγον ἢ ἱκανὸν (so v) διεξίωμεν and ἵνα μὴ ἢ ἐῶμεν συχνὸν λόγον ἢ οὐχ ἱκανὸν διεξίωμεν— improve the antithesis, but are much too violent. It is safest to retain the MS reading until a thoroughly satisfactory emendation appears. Dr Jackson suggests ἵνα μὴ ἐῶμεν ἱκανὸν λόγον ἢ οὐχ ἱκανὸν διεξίωμεν.

τίς οὖν ἡ παιδεία; κτλ. The educational scheme contained in Books II and III contributes to the purgation of the τρυφῶσα πόλις, and thereby helps to complete Plato's second picture of an ideal city: see on 372 D ff. For the correct understanding of these regulations it is well to bear in mind (1) that Plato's object in this preliminary discipline is to train the character rather than the intellect (cf. IV 430 C note), and (2) that all the guardians have to pass through this curriculum. The higher scheme of education (in Book VII), on the other hand, is confined to those guardians who are to be made Rulers in the State, and its express aim is to educate the intellect rather than the will. See especially VI 502 E, VII 521 D—522 A notes The best discussion on Plato's theory of education in its broader aspects is still, I think, Nettleship's Essay in Hellenica pp. 67—180. Platon's Erziehungstheorie n. s. Schrift. dargestellt von Dr A. Drygas Schneidemühl 1880 is a useful summary. For Plato's criticism of poetry, we may refer in particular to Heine's excellent dissertation De rat. quae Platoni c. poet. Gr. intercedit etc. Vratislaviae 1880, and to Reber's Plato und die Poesie Leipzig, 1864.

ἔστιν δέ που — μουσική . The usual Greek view (see for example Isocr. 15. 180—185), corrected by Plato in III 410 C ff.

εἶπον . Richter ( Fl. Jahrb. 1867 p. 141) revives Muretus' conjecture εἶδος: but εἶπον is alone satisfactory. The confusion of ο and ω occurs in Inscriptions from the third century B.C. onwards (Meisterhans^{3} p. 24 note 128). See also Introd. § 5.

λόγων δὲ ‐‐ ἕτερον . The word ‘lies’ is here used by Plato in its popular sense of that which is false in fact: his own definition of the ‘veritable lie’ is different: see 382 B note ‘Lies’ are necessary—so Plato holds—in education: only they must be moral lies. Under ‘lies’ he includes stories ( μῦθοι) about the gods, about the daemons and heroes long since dead, about a future life—all of them subjects where the alleged facts cannot be verified. The ἀληθεῖς λόγοι are concerned with men, and are passed over by Plato, because he could not state his view without anticipating the conclusion which the Republic is intended to prove (see III 392 A—C). This point is missed by Krohn ( Pl. St. p. 12).

ἀληθῆ : i.e. truths of fact or history, not yet with reference to moral truth, for nothing has been said to change the connotation of ψευδής or its opposite ἀληθής. In Plato's view legend contains some elements of historical truth.

ἀρχὴ ‐‐ μέγιστον : semi - proverbial, with reference to ἀρχὴ ἥμισυ παντός: cf. Laws 753 E, and (for the application of the sentiment) ib. 765 E.

μάλιστα ‐‐ τύπον . See cr. n. To τύπος there are two objections: (1) the subject of πλάττεται and ἐνδύεται should be the same; but the subject of πλάττεται is not τύπος, but the νέῳ καὶ ἁπαλῷ ὁτῳοῦν, cf. πλάττειν τὰς ψυχὰς in C below: (2) it is more natural and correct to say that an object which ‘is being moulded’ ‘puts on’ a τύπος, than to say that the τύπος sinks into it. Reading τύπον we obtain the proper contrast between ἐνδύεται and ἐνσημήνασθαι: the youth puts on whatever impression or type the educator desires to stamp him with. The metaphor becomes more explicit in Plutarch De lib. educ. 3 F καθάπερ γὰρ σφραγῖδες τοῖς ἁπαλοῖς ἐναπομάττονται κηροῖς, οὕτως αἱ μαθήσεις ταῖς τῶν ἔτι παιδίων ψυχαῖς ἐναποτυποῦνται. Cf. also Theaet. 191 D and Hor. Epp. II 2. 8 argilla quidvis imitaberis uda.

ῥᾳδίως οὕτω : ‘carelessly, without more ado’: cf. 378 A and I 331 C. This idiomatic οὕτω is common with adverbs like ῥᾳδίως, εἰκῇ, ἁπλῶς, νῦν, ἐξαίφνης: for examples see Blaydes on Ar. Wasps 461 .

καλὸν : sc. μῦθον, which some MSS (including II) insert. For μῦθον understood from μυθοποιοῖς cf. III 399 D, where τοῦτο i.e. αὐλός is understood from αὐλοποιούς, 410 A, where αὐτοί (i.e. ἰατροί) follows ἰατρική, IV 421 E, and (with Schneider) Laws 886 C θεογονίαν διεξέρχονται, γενόμενοί τε (sc. οἱ θεοὶ) ὡς πρὸς ἀλλήλους ὡμίλησαν.

πλάττειν κτλ. Mothers and nurses practised massage on the bodies of infants: cf. Laws 789 E τιθέντες νόμους τὴν μὲν κύουσαν περιπατεῖν, τὸ γενόμενον δὲ πλάττειν τε οἷον κήρινον ἕως ὑγρόν, καὶ μέχρι δυοῖν ἐτοῖν σπαργανᾶν, and Alc. I 121 D. A trace of massage practised for medical purposes appears in Zeno Fr. 180 (ed. Pearson).

ἐλεγέτην . The dual links together Homer and Hesiod as jointly responsible for Greek theology: see on 363 A. Among the first to rebel against their authority were Pythagoras, Xenophanes, and Heraclitus (D. L. VIII 21, IX 18, IX I). Xenophanes' protest was particularly famous in antiquity: see Sext. Emp. adv. Math. I 289 and IX 193 ap. Ritter and Preller Hist. Philos. Gr.^{7} pp. 76, 77. Plato's attack on the Olympian theology in this and the succeeding book was perhaps the severest blow that Paganism received before the Christian era, and pointed the way for those exaggerated diatribes against the heathen gods in which it afterwards became the fashion of early Christian apologists to indulge, beginning with the Apology of Aristides (cc. 8—11). Cf. X 607 B note

ὅπερ ‐‐ ψεύδηται . ὅπερ is τὸ εἰκάζειν κακῶς περἰ θεῶν etc. A distinction is drawn between mere lies and the lie which is in itself οὐ καλόν, unbeautiful and immoral in tendency, e.g. the story of Uranus and Cronus ( ὁ εἰπὼν οὐ καλῶς ἐψεύσατο in E below). Such legends not merely misrepresent the gods, but also corrupt mankind.

τῶν μεγίστων : masculine, not neuter: cf. 378 B.

Ἡσίοδος . Theog. 154—181.

τὰ δὲ δὴ κτλ . δή emphasizes the case of Cronus as the most important (cf. Prot. 311 D , 312 E): it is so because the delinquent is Zeus, the reigning king of gods and men. The example set by Zeus on this occasion was no doubt sometimes used to justify wrong-doing: see for example Aesch. Eum. 640 , 641, Ar. Clouds 904 —906 πῶς δῆτα δίκης οὔσης ὁ Ζεὺς | οὐκ ἀπόλωλεν τὸν πατέρ’ αὑτοῦ | δήσας ; ib. 1079 ff., Eur. H. F. 1317 — 1319, and especially Pl. Euthyph. 5 E — 6 A, where Euthyphro urges the analogy in all seriousness to justify his vexatious prosecution of his own father. The pernicious effect of such legends on human conduct is again pointed out in Laws 886 C, 941 B: cf. also Isocr. Bus. 38—43, Luc. Men. 3, and Grote Plato III p. 194 note

ῥᾳδίως οὕτω : 377 B note

θυσαμένους ‐‐ ἀκοῦσαι . ἀπορρήτων suggests the mysteries, whence the allusion to the ‘mystic pig’ ( Ar. Ach. 764 ). For ἄπορον, ‘unprocurable’ (Jowett), ἄπνρον has been suggested, absurdly enough. ἄπορον is further explained by ὅπως— ἀκοῦσαι. It should be noted that ὅπως with a past tense of the indicative in clauses of this kind is rare in Plato: it occurs again only in Laws 830 B, 959 C (where ἄν should be expunged). Cf. Weber in Schanz's Beiträge zur hist. Synt. d. Gr. Sprache II 2, p. 64.

οὐδ’ αὖ has been needlessly doubted by Richter ( Fl. Jahrb. 1867 p. 142), who suggests οὐδέν. The words ἀδικῶν—ποιοῖ correspond to the conduct of Uranus and Cronus towards their children: οὐδ’ αὖ—τρόπῳ to Cronus' treatment of Uranus, and Zeus' of Cronus. Cf Euthyph. 5 E —6 A. The Euthyphro presents so many parallels to § 378 that some have—erroneously, no doubt—supposed it to be a spurious elaboration of that section: see my edition of the dialogue p. xxix.

πολλοῦ δεῖ ‐‐ ποικιλτέον . πολλοῦ δεῖ is not adverbial (like ἥκιστα), as J. and C assert: otherwise δεῖ would be δεῖν (so Herwerden would read Mnem. N. S. XI p. 339). The asyndeton is justified by emphasis and the ampliative character of the sentence. The verbals are best explained (with Stallbaum) by supposing an ellipse of εἶναι: cf. Schanz Nov. Comm. Pl. p. 33.

ποικιλτέον . ποικίλλειν is used of depicting in a variety of colours (VIII 557 C), not necessarily by embroidery. Cf. 373 A note There is probably a special reference here to the πέπλος. At the greater, if not also at the lesser, Panathenaic festival, a robe woven by Athenian maidens and representing the triumph of Athena and the Olympians over the giants, together with other celestial fights, was carried in procession to the Acropolis, and presented to the statue of the goddess in the Erechtheum: cf. Euthyph. 6 B , C and Mommsen Feste d. Stadt Athen pp. 107 ff. The subject was depicted on the Parthenon frieze: see Baumeister Denkm. d. kl. Alterth. II p. 1185. The allusion to the ceremony is the more appropriate in this connexion, if, as appears to be probable, the action of the dialogue takes place just before the great Panathenaea of 410 B.C. See Introd. § 3.

ὡς οὐδεὶς κτλ. Plato desires to obtain a religious sanction for his institutions, as in the myth III 414 B ff. The best δημηγόρος, according to Socrates, is ὁ στάσεις τε παύων καὶ ὁμόνοιαν ἐμποιῶν ( Xen. Mem. IV 6. 14 ): and the Platonic State may from this point of view be regarded as “an attempt to determine the ways and means of securing political ὁμόνοια” (Krohn Pl. St. p. 369).

λεκτέα —see cr. n.—cannot be dispensed with. Madvig's suggestion, that μᾶλλον is corrupt for φατέον or ᾀστέον or the like, and Liebhold's μελητέον for μᾶλλον, are much less probable than the accidental omission of λεκτέα in A. See Introd. § 5. Vermehren ( Pl. Stud. p. 92), rejecting λεκτέα, would carry on μυθολογητέον or the like; but this solution is much too difficult.

καὶ πρεσβυτέροις γιγνομένοις . The dative goes with λογοποιεῖν (‘to make tales for them as they grow older’), and καί before τοὺς ποιητάς means etiam. This explanation was proposed by Richter ( Fl. Jahrb. 1867 p. 138) and Vermehren (l. c. p. 91), and is probably right. Cf. Ar. Frogs 1054 f. Others connect the words with καὶ γέρουσι καὶ γραυσί: old men, old women, and the boys themselves as they grow older, must tell such stories πρὸς τὰ παιδία εὐθύς. But it is difficult to understand τοῖς παιδίοις with γιγνομένοις unless πρεσβυτέροις γιγνομένοις is construed with λογοποιεῖν.

ὑέος . Hephaestus. Διός is a false reading derived from a mistaken reference to Il. XV 18 ff. The story (according to Clement ap. Suid. s. vv. Ἥρας δὲ δεσμοὺς ὑπὸ υἱέος) was in Pindar: παρὰ Πινδάρῳ γὰρ ὑπὸ Ἡφαίστου δεσμεύεται ἐν τῷ ὑπ’ αὐτοῦ κατασκευασθέντι θρόνῳ—καί φασι δεθῆναι αὐτὴν ἐπιβουλεύσασαν Ἡρακλεῖ. Cf. Paus. I 20. 3.

Ἡφαίστου ῥίψεις . Il. I 586—594.

θεομαχίας ‐‐ οὐ παραδεκτέον . Homer Il. XX 1—74, XXI 385—513. Cf. Xenophanes Fr. 1. 19—22 (Bergk) and Pind. Ol. IX 43, 44 μή νυν λαλάγει τὰ τοιαῦτ’ · ἔα πόλεμον μάχαν τε πᾶσαν χωρὶς ἀθανάτων.

ἐν ὑπονοίαις : adverbial, like ἐν φαρμάκου εἴδει III 389 B (J. and C.). The allegorical interpretation of Homer probably originated in the desire to save his character for piety and morality: πάντη γὰρ ἠσέβησεν (says Heraclides Alleg. Hom. ad init.), εἰ μηδὲν ἠλληγόρησεν. Before the time of Plato it was practised by Theagenes of Rhegium, Anaxagoras, Metrodorus of Lampsacus, Stesimbrotos of Thasos and others: see Wolf Proleg. ad Homerum pp. 161—166 and Jebb's Homer p. 89. In Plato's day the Cynics were the chief exponents of this school of criticism, especially Antisthenes: examples may be found in Winckelmann's Antisth. Frag. pp. 16, 23—28: cf. also Dümmler Antisthenica pp. 16 ff. Dümmler, many of whose combinations are highly speculative, regards the present passage as directed against Antisthenes, whose rivalry with Plato is well known: but there is nothing to suggest any personal reference. The historical Socrates occasionally played with the same weapons, as appears from Xen. Symp. 3. 6, and Mem. I 3. 7: so also does Plato, but seldom, if ever, without irony, e.g. Rep. 1 332 B ἠνίξατο— ὁ Σιμωνίδης ποιητικῶς: cf. also Theaet. 194 C , Alc. II 147 B—D al. Plato's attacks upon Homer lent a great impetus to this method of exegesis—the only method, as it was thought, by which his animadversions could be met: cf. Schow's Heraclides pp. 223—234.

οἷος τυγχάνει ‐‐ ἐν μέλεσιν . τυγχάνει ὤν=‘really is’: cf. I 337 B note On the omission of ἐάν τε ἐν μέλεσιν in A see Introd. § 5.

ἀλλὰ μὴν κτλ. It is first proved that good is not the cause of evil ( ἀλλὰ μὴν—πῶς γάρ;), and next that good is the cause of εὐπραγία ( τί δέ;— ναί): the conclusions are then stated in the reverse order. The step by which each conclusion is reached—the identification of ἀγαθόν and ὠφέλιμον—is Socratic (cf. Xen. Mem. IV 6. 8 ); but it is doubtful if the historical Socrates ever went so far as to deny that God is sometimes the cause of real evil or adversity to man, in spite of his belief in Providence ( Mem. I 4 and IV 3; yet I 4. 16 οἴει δ’ ἂν τοὺς θεοὺς τοῖς ἀνθρώποις δόξαν ἐμφῦσαι, ὡς ἱκανοί εἰσιν εὖ καὶ κακῶς ποιεῖν, εἰ μὴ δυνατοὶ ἦσαν;). The moral goodness of the Deity himself was proclaimed before Socrates and Plato by Xenophanes, Pindar, and the dramatists, but the inference, that God, because He is good, is never the cause of evil, is probably due to Plato. Bacchylides expresses a kindred sentiment in Fr. 29 (Bergk) Ζεὺς ὑψιμέδων, δ(\ς ἅπαντα δέρκεται, | οὐκ αἴτιος θνατοῖς μεγάλων ἀχέων. Read in the light of Book VI, the theology of this and the succeeding chapters gains, no doubt, in significance and depth; yet it is illegitimate to argue on this account (as Susemihl does Genet. Entwick. II p. 121) that the existence of the Idea of Good is already presupposed, unless it is shewn that Plato could not have purified his theology except by metaphysics. In point of fact, Plato might have written the end of Book III even if he had never thought of the Ideas at all.

οὐδ) ἄρα ‐‐ πάντων . Contrast Aesch. Ag. 1485 , 1486 Διὸς παναιτίου πανεργέτα. | τί γὰρ βροτοῖς ἄνευ Διὸς τελεῖται ; Suppl. 822—824 and many other examples in Nägelsbach Hom. Theol. pp. 26, 51 ff., and Nachhom. Theol. pp. 16, 18, 60 ff., 73 ff.

πολὺ γὰρ ‐‐ ἡμῖν . An old saying, as appears from Pind. Pyth. 3. 81 ff. μανθάνων οἶσθα προτέρων· | ἓν παρ’ ἐσλὸν πήματα σύνδυο δαίονται βροτοῖς | ἀθάνατοι, and Eur. Suppl. 196 , 7: cf. also Hom. Il. XXIV 527 ff., Philem. Fr. Inc. 65 (ed. Meineke). Plato and Aristotle make room for it in their philosophies: see e.g. Pol. 273 D , Laws 906 A, and Arist. Probl. X 45. 895^{b} 39 ff. ἡ φύσις φαῦλα μὲν πάντα ποιεῖ, καὶ πλείους καὶ πλείω, σπουδαῖα δ’ ἐλάττω, καὶ οὐ πάντα δύναται. The counterpart in the sphere of morals is Bias's οἱ πολλοὶ κακοί: with which may be compared Rep. IV 428 E, 431 A, 442 A, C, IX 588 D. It is a melancholy cry born of the age of iron: in the golden age—so Plato tells us Pol. 273 C —the balance was the other way.

ἄλλ) ἄττα ‐‐ τὰ αἴτια . The dualism should not be taken too seriously, in spite of the good and evil souls in Laws 896 E. Plato is not now constructing a philosophy, but casting moulds for theology and poetry.

δοιοὶ πίθοι . See Il. XXIV 527—532 δοιοὶ γάρ τε πίθοι κατακείαται ἐν Διὸς οὔδει | δώρων οἷα δίδωσι κακῶν, ἕτερος δὲ ἑαων· | ᾦ μέν κ’ ἀμμίξας δώη Ζεὺς τερπικέραυνος, | ἄλλοτε μέν τε κακῷ ὅ γε κύρεται, ἄλλοτε δ’ ἐσθλῷ· | ᾧ δέ κε τῶν λυγρῶν δώῃ, λωβητὸν ἔθηκεν | καί ἑ κακὴ βούβρωστις ἐπὶ χθόνα δῖαν ἐλαύνει. In our Homer there is apparently only one jar of good to two of evil (see Leaf ad loc. and cf. 379 C note): in Plato there is one of each. So great a difference is not likely to be due to Plato: it is easier to believe that he used a different recension from the Alexandrian. The use of κῆρες unpersonihed was apparently not admitted by the Alexandrian critics. Cf. Wolf Proleg. p. 37, and Howes in Harvard Studies in Cl. Phil. VI p. 204.

ἀγαθῶν ‐‐ τέτυκται is either from a lost line of Homer, or from some other poet (as Schneider inclines to think): note οὔτ’ ἄλλου ποιητοῦ just above. There can hardly be any reference to Il. IV 84 Ζεύς, ὅς τ’ ἀνθρώπων ταμίης πολέμοιο τέτυκται, as Howes imagines (l. c. p. 196). The sentiment is common: cf. e.g. Hes. O. D. 669 and Pind. Isthm. IV 52, 53 Ζεὺς τά τε καὶ τὰ νέμει, Ζεὺς ὁ πάντων κύριος.

σπονδῶν σύγχυσιν . Il. IV 69 ff.

θεῶν ἔριν τε καὶ κρίσιν . This is usually explained as referring to the Theomachy ( Il. XX 1—74), which was caused by Zeus and Themis in the sense that Zeus sent Themis to summon the gods to the council at which it was sanctioned ( v. 4). But (1) Themis' part in causing the Theomachy is very small, (2) the simplest and most natural meaning of κρίσις is not ‘contention,’ but ‘judgment’ or ‘decision,’ and (3) the Theomachy in Homer is not productive of evil to men, but only to the gods themselves: its citation here would therefore be quite irrelevant. W. R. Hardie (in Cl. Rev. IV p. 182) is, I believe, right in supposing that the strife of the goddesses three and Paris' judgment is meant. ἔρις and κρίσις are regularly thus used: e.g. Eur. I. A. 1307 κρίσιν—στυγνὰν ἔριν τε καλλονᾶς; cf. ib. 581, Hel. 708, Troad. 924. Hec. 644 f. Κρίσις was the name of Sophocles' play on the judgment of Paris ( Fr. 330). The poem referred to by Plato is the Cypria (so also Wilamowitz Hom. Unters. p. 367 note 46), which traced the war of Troy to the judgment of Paris, and that to Zeus' deliberations with Themis ( Ζεὺς βουλεύεται μετὰ τῆς Θέμιδος περὶ τοῦ Τρωικοῦ πολέμου Kinkel Epic. Graec. Fr. p. 17. Θέμιδος is Heyne's emendation for Θέτιδος: but it is scarcely open to doubt: for the marriage of Peleus and Thetis, at which the three goddesses quarrelled, was an episode of the poem, and Thetis could hardly therefore have been privy to the plot. See Kinkel l. c. pp. 20, 22 and Jebb's Homer p. 153). Themis was Zeus' ἀρχαία ἄλοχος (Pind. Fr. 30 B ergk), and still appears as one of the Olympians in Il. XV 87. The Cypria is quoted again by Plato in Euthyph. 12 A . We may fairly suppose that θεῶν ἔρις τε καὶ κρίσις was the heading of one of the introductory episodes in the poem: to this also the omission of the article with ἔριν τε καὶ κρίσιν seems to point. Mr Hardie thinks Plato may have attributed the poem to Homer; but Euthyph. l. c. ( ὁ ποιητὴς ὁ ποιήσας) does not favour this view.

θεὸς μὲν ‐‐ θέλῃ : Aesch. Fr. 160 . For other examples of this familiar Greek idea see Nägelsbach Hom. Theol. p. 321 and Nachhom. Theol. pp. 54 ff.

ἐν οἷς ‐‐ ἔνεστιν . I have left these words in the text, although they are certainly open to suspicion, and have been condemned by Platt ( Cl. Rev. III p. 72). The antecedent to οἷς is apparently τὰ τῆς Νιόβης πάθη: but the play was not called ‘The sufferings of Niobe’ but ‘Niobe,’ and the relative can hardly precede its antecedent in sentences of this kind. If οἷς is referred to ταῦτα understood after ποιῇ, then ἐν is difficult: ‘if any one puts into poetry topics in which these iambics occur’ gives no good sense. Unless Plato is writing very inaccurately, we must pronounce the clause a marginal gloss on τὰ—πάθη .

ὠνίναντο κολαζόμενοι . An earlier generation looked upon punishment as retributory— δράσαντι παθεῖν. This view appears in Hes. Fr. 217, ed. Goettling, and especially in Aeschylus, e.g. Ag. 1563 f., Choeph. 309—314, 400 —404, 886, 927: in Sophocles and Euripides it is rarer ( Ant. 1074—1076, El. 1411 f., 1495 f., Andr. 438, Suppl. 614 —616), and Euripides expressly argues against it in Or. 508 ff. Traces of a milder theory were however contained in the doctrine πάθος μάθος ( Ag. 176 ff.), as well as in the use of words like σωφρονίζειν, δικαιοῦν, εὐθύνειν, for ‘punish.’ In Plato punishment is remedial. Ignorance or vice is in the soul what disease is in the body (IV 444 C, cf. IX 591 A, B), and the judge is the soul's physician (III 409 E ff., Gorg. 478 D ): hence ( Gorg. 480 B ff.) the sinner should go before the judge as a patient visits his doctor, and we should even prosecute our guilty friends and relations. See also Laws 854 D, 862 E, 934 A, 944 D τὸν γὰρ κακὸν ἀεὶ δεῖ κολάζειν, ἵν’ ἀμείνων ᾖ. The punishment, again, which awaits the wicked after death is intended to cure their souls, unless they are incurable: and such as are themselves incurable, help to cure others by their deterrent example (X 616 A): so that in its deepest relations this doctrine reaches to the very roots of Plato's philosophy, with all due deference to Mr W. S. Lilly, who with much intemperance of language denounces those who attribute such a view to Plato ( Fortnightly Review N.S. XLVI p. 116).

ἐν τῇ αὑτοῦ πόλει : ‘in one's own city,’ with reference to the subject of διαμαχετέον, not to τινα. Plato implies that the preachers of such theology must be suppressed in his ideal city. In all this Teichmüller ( Lit. Fehd. I p. 114) detects an assault upon Isocrates, but his evidence is of the slightest.

μυθολογοῦντα is rejected by Herwerden: Ast suggested μυθολογούμενα. The choice of the participle is determined by λέγειν, which is more important than ἀκούειν: for without saying hearing is impossible. μήτε νεώτερον μήτε πρεσβύτερον belongs both to λέγειν and to ἀκούειν.

νόμων τε καὶ τύπων . All laws are in Plato's view only moulds or outlines, within which our actions should fall. Cf. infra 383 C and especially Pol. 294 A ff.

ἆρα γόητα κτλ. Although the gods are constantly represented as deceivers in Greek poetry and legend, Plato was by no means the first to uphold the opposite view. In Pindar ( Ol. 10. 4) Truth is the daughter of Zeus, and the dramatists often teach a similar doctrine: see Nägelsbach Nachhom. Theol. p. 46. There is a close imitation of Plato's argument throughout this passage in Arist. Fr. 15. 1476^{b} 14 ff. ed. Rose.

αὐτόν is emphatic: the contrast is between actual and apparent transformations of the Deity. After αὐτόν, Herwerden would insert παντοδαπόν, comparing 381 E; before it, Richards adds ἄλλον, by which Benedictus and Ast replace αὐτόν. Hartman proposes <τι> γιγνόμενον. It has apparently escaped notice that γιγνόμενον, as well as ἀλλάττοντα τὸ αὑτοῦ εἶδος, belongs to εἰς πολλὰς μορφάς in the sense of ‘passing into’: cf. Tim. 57 A εἰς ἄλλο τι γιγνόμενον, infra III 400 B εἰς βραχύ τε καὶ μακρὸν γιγνόμενον, IX 588 C, and the frequent idiom γένεσις εἰς e.g. Phaed. 71 B , 71 E, Phil. 26 D , Tim. 49 C , 54 B.

ἁπλοῦν : one of the watchwords of Plato's State (370 B, C, 374 A—D al.): his citizens are to be nothing if not ἁπλοῖ. In making the gods a reflection of the type of human character which he desired to foster, Plato is acting strictly in accordance with the method of Greek theology, whose Olympus is an image of human society. The end of human action is ὁμοίωσις θεῷ κατὰ τὸ δυνατόν ( Theaet. 176 B ); and Plato's God, changeless and with ‘no shadow of turning,’ furnished the citizens of his ideal city with an abiding standard of human conduct. Cf. 383 C.

τί δὲ τόδε ; Steinhart ( Platon's Werke V p. 680) justly observes that the method of reasoning employed here— the disproof of each of the two members of the opposite alternative—recalls the arguments by which Parmenides established the attributes of Being (see RP.^{7} §§ 95, 98); but the resemblance is not close enough to suggest that Plato was thinking of Parmenides when he wrote this chapter. Although the unchangeableness of God was taught by Xenophanes and the Eleatics, there are few if any traces of such a doctrine outside the philosophers before Plato.

ὑπὸ μὲν ἄλλου κτλ . μέν has its counterpart in ἀλλ’ ἆρα αὐτὸς αὑτὸν κτλ. 381 B.

κινεῖται : a more general word for change than ἀλλοιοῦται: cf. Theaet. 181 D δύο δὴ—εἴδη κινήσεως, ἀλλοίωσιν, τὴν δὲ περιφοράν. The doctrine of the permanence and immutability of good enunciated here foreshadows, but does not presuppose, the metaphysical predominance of the Good in Book VI.

καὶ ἀμφιέσματα . See cr. n. and Introd. § 5.

ἀνάγκη : sc. ἐστίν. For ἔμοιγε δοκεῖ without ὡς see on I 332 E. Hartman needlessly suggests ἔμοιγε δοκεῖν.

θεοὶ ‐‐ πόληας . Od. XVII 485 f. Cf. Nägelsbach Hom. Theol. pp. 166—168.

Πρωτέως καὶ Θέτιδος . For Proteus see Od. IV 456—458. Aeschylus also wrote a satyric drama called Proteus: Fragg. 208—213. The transformations of Thetis to escape marrying Peleus had been celebrated by Pindar ( Nem. IV 62 ff.), Sophocles ( Fr. 548), perhaps also (as Stallbaum thinks) by Hesiod in his ἐπιθαλάμιον εἰς Πηλέα καὶ Θέτιν (see Goettling's Hesiod pp. XLIX and 304).

ὡς ἱέρειαν ‐‐ βιοδώροις : from Aesch. Ξαντρίαι (Schol. on Ar. Frogs 1344 ). Dindorf ( Aesch. Fr. 170 ) restores as follows: ὀρεσσιγόνοισι | Νύμφαις κρηνιάσιν κυδραῖσι θεαῖσιν ἀγείρω, | Ἰνάχου Ἀργείου ποταμοῦ παισὶν βιοδώροις. Herwerden's βιοδώρου is a wanton change: the sons of the river-god are his tributaries, and lifegiving like himself. It is not clear why Hera was disguised as a priestess. The incident in Inachus' history most suited to dramatic treatment was the persecution of his daughter Io by Hera in consequence of her intrigue with Zeus. As Io was a priestess of Hera, Hera may have disguised herself as another priestess in order to discover her husband's unfaithfulness: see Apollod. Bibl. II 1. 3 φωραθεὶς δὲ (sc. ὁ Ζεὺς) ὑφ’ Ἥρας, τῆς μὲν κόρης ἁψάμενος εἰς βοῦν μετεμόρφωσε λευκήν, αὐτὴν δὲ ἀπωμόσατο μὴ συνελθεῖν. The subject seems to have been treated by Sophocles in his satyric drama Inachus ( Fragg. 255—278). With ὡς ἱέρειαν ἀγείρουσαν cf. ἀγύρται in 364 B and note ad loc.

τοιαῦτα πολλά . For examples see Heyne's Virgil II pp. 146— 152 (cited by Ast on 381 D). πολλὰ ψεύδονται ἀοιδοί. said the proverb.

κακῶς : like οὐ καλῶς 377 E.

ὡς ‐‐ ἰνδαλλόμενοι . ἄρα expresses incredulity (358 C note) and τινες contempt. Plato is thinking, inter alia, of the bugbears of the nursery—Lamia, Mormo, and Empusa, whose power of self-transformation was unlimited: see Blaydes on Ar. Frogs 293 . ξένοις need not here be limited to the masculine gender. Cf. Strab. I 19 παισὶ προσφέρομεν—εἰς ἀποτροπὴν—τοὺς φοβεροὺς ( μύθους). ἥ τε γὰρ Λαμία μῦθός ἐστι καὶ ἡ Γοργὼ καὶ ὁ Ἐφιάλτης καὶ ἡ Μορμολύκη.

φάντασμα is said with reference to φαίνεσθαι just above, and should be taken both with λόγῳ and ἔργῳ. The φάντασμα λόγῳ is the spoken lie: an example of the φάντασμα ἔργῳ is a φαντασία or unreal appearance (382 E). The words ἔργῳ φάντασμα προτείνων must not be understood of actual self-transformations of the gods.

τό γε ὡς ἀληθῶς ψεῦδος κτλ. Cf. τοῦ ἀληθῶς ψεύδους Theaet. 189 C , and (for the sentiment) Laws 730 C.

οὐδεὶς ἑκὼν κτλ. With Plato, as with Socrates, vice is ignorance, and involuntary. The doctrine reappears below in III 413 A, IX 589 C: it is further implied by the entire scheme of education in Books VI and VII. For other assertions of this view in Plato see Simson der Begriff d. Seele bei Pl. p. 125 note 359. Cf. also Soph. Fr. 663 ἡ δὲ μωρία | μάλιστ’ ἀδελφὴ τῆς πονηρίας ἔφυ. The identification of ignorance and vice is in harmony with popular Greek psychology, in which the intellect was not clearly distinguished from the will; it can be traced in the moral connotation of words like ἀμαθής, ἀπαίδευτος, ἀγνώμων. In close connexion with this conception of vice is Plato's view of punishment as remedial: see 380 B note

τὰ ὄντα κτλ. τὰ ὄντα =‘the truth.’ The contrast between the act and state in ψεύδεσθαί τε καὶ ἐψεῦσθαι resembles I 351 B: ἐψεῦσθαι, moreover, suitably bridges the distance between ψεύδεσθαι and ἀμαθῆ εἶναι. ἔχειν τὸ ψεῦδος corresponds to ψεύδεσθαι, κεκτῆσθαι τὸ ψεῦδος to ἐψεῦσθαι: the contrast is between ‘holding, ready for use, that which is already possessed,’ and permanent possession: cf. Soph. Ant. 1278 and Jebb ad loc. The words ἐν τῷ τοιούτῳ, ‘in such a case’ (i.e. ἐν τῷ ἐψεῦσθαι τῇ ψυχῇ περὶ τὰ ὄντα), are quite satisfactory (cf. III 393 C), and ought not to have caused Herwerden difficulty.

μίμημά τι ‐‐ ψεῦδος . τοῦ ἐν τῇ ψυχῇ παθήματος must not be explained (with Bosanquet Companion p. 93) as the state of mind of him who tells a lie: for that is knowledge, and the spoken lie certainly is not an imitation of knowledge. They refer to the ‘true lie,’ which is a certain πάθημα in the soul of the ‘true liar,’ viz. ignorance, and of which the spoken lie is an imitation. It is a tolerably accurate definition of a lie to call it ‘an imitation of ignorance in the soul’: cf. IV 443 C note The spoken lie is ‘not a wholly unmixed lie,’ because it implies that the speaker knows the truth: in a certain sense therefore it is mixed with truth. It is ὕστερον γεγονός, because the spoken lie cannot be uttered until the truth is known. Inasmuch as the spoken lie is mixed with truth, it is better than the ‘veritable lie.’ We have here nothing but a special application of the old Socratic paradox ὁ ἑκὼν ἁμαρτάνων ἀμείνων (see on I 334 A). I have placed a comma after γεγονός, to mark the antithesis between εἴδωλον and ἄκρατον ψεῦδος, and because εἴδωλον is not so much to be taken with τοῦ ἐν τῇ ψυχῇ παθήματος: rather it stands for εἴδωλον ψεύδους, as οὐ πάνυ ἄκρατον ψεῦδος shews. The distinction between veritable and spoken lies savours, no doubt, of idealism: but it enables Plato to call his ideal archons ideally truthful, even when practically they tell lies, and it is with this object in view that the distinction is introduced. See III 389 B.

πότε ‐‐ μίσους ; τῷ is masculine: it is presently shewn that the spoken lie is useless to God. Plato does not permit a man to lie in his own interest. Ordinary Greek morality, in spite of Achilles' ἐχθρὸς γάρ μοι κεῖνος <*>μῶς Ἀΐδαο πύλῃσιν etc., probably did. The saying of Democritus ἀληθομυθεύειν χρεών, ὅπου λώιον (Stob. Flor. 12. 13) leaves us to infer that we may also lie ὅπου λώιον. Cf. Soph. Fr. 323 καλὸν μὲν οὖν οὐκ ἔστι τὰ ψεύδη λέγειν· | ὅτῳ δ’ ὄλεθρον δεινὸν ἁλήθεἰ ἄγει, | συγγνωστὸν εἰπεῖν ἐστὶ καὶ τὸ μὴ καλόν. The cynical immorality of Hdt. III 72 exceeds what Greek public opinion would have tolerated: cf. Arist. Eth. Nic. IV ch. 13. See also on III 389 B and Nägelsbach Nachhom. Theol. pp. 240 ff.

ἆῤ οὐ ‐‐ πολεμίους κτλ. Cf. I 331 E— 332 B.

τῶν καλουμένων φίλων depends on ἀποτροπῆς. If ὅταν διὰ μανίαν—τότε had been omitted, the construction would be quite clear: as it is, some difficulty has been felt. Schneider understands τινες as subject to ἐπιχειρῶσι: by Hermann ὅταν is changed to οἳ ἂν: by Herwerden ὅταν to οἳ ἂν and τότε to τοῦτο: while Stallbaum resorts to an anacoluthon, as if Plato had intended to say τῶν καλουμένων φίλων ἕνεκα. None of these expedients is so simple as to connect ἀποτροπῆς with φίλων. The clause ὅταν—πράττειν cancels out with τότε and does not affect the construction. καλουμένων, ‘so-called,’ involves a theory of friendship, viz. that no one who is ἀνόητος καὶ μαινόμενος can be a friend to man (any more than to God: cf. 382 E).

μυθολογίαις κτλ. Plato seems to have supposed that ancient history and mythology could be manufactured to order. Cf. Arist. Pol. B 9. 1269^{b} 28 and Susemihl ad loc. He attempts the task himself in III 414 B ff., Prot. 320 C —322 D (unless this is really an extract from one of Protagoras' own works), Pol. 269 A —274 E, Tim. 21 A — 25 D, Critias, and Laws 676 B—682 D.

εἰδέναι . The omniscience of the gods was no new doctrine: see Nägelsbach Hom. Theol. p. 23, Nachhom. Theol. pp. 23 ff.

ποιητὴς ‐‐ ἔνι . ‘There is nothing of the lying poet in God.’ Cf. 365 C note I can see no point in Stallbaum's notion that there is a play on the two senses of ποιητής—‘poet’ and ‘creator.’

ψεύδοιτο . ἄν is carried on: cf. I 352 E note

μαινομένων . Phaedr. 265 A μανίας δέ γε εἴο=η δύο, τὴν μὲν ὑπὸ νοσημάτων ἀνθρωπίνων, τὴν δὲ ὑπὸ θείας ἐξαλλαγῆς τῶν εἰωθότων νομίμων γιγνομένην. Plato refers here only to the first variety: the second is discussed in Phaedr. 265 B ff.

κομιδῇ ἄρα κτλ. The words ἁπλοῦν, οὔτε αὐτὸς μεθίσταται sum up 380 D—381 E (see on ἁπλοῦν in 380 D), the rest 382 A—D.

οὔτε κατὰ φαντασίας . See cr. n. and Introd. § 5. φαίνεσθαι and ἔργῳ φάντασμα προτείνων in 381 E, 382 A favour the view that these words are genuine.

ὕπαρ οὐδ’ ὄναρ . See cr. n. ὕπαρ οὐδ’ ὄναρ is not co-ordinate with οὔτε κατὰ φαντασίας etc., but subordinate to them: for φαντασίαι, λόγοι, and especially σημείων πομπαί might be vouchsafed either in waking moments or in dreams: see Stengel and Oehmichen in Iwan Müller's Handbuch V 3 pp. 37—47. For the doctrine cf. Xen. Mem. I 3. 4.

παράγειν . παράγοντας (conjectured by Richards) would be easier, but the slip, if such it be, is excusable. ὡς—ὄντας is not the accusative absolute: if it were, ὡς would express the reason, and here it does not. We are defining the τύπος: and the construction is (they must ποιεῖν) ὡς μήτε αὐτοὺς γόητας ὄντας, ‘represent the gods as neither themselves being sorcerers,’ etc. In παράγειν the construction is changed, but the change is natural, for our rule applies both to λόγος and ποίησις ( καὶ λέγειν καὶ ποιεῖν), and λέγειν takes the accusative and infinitive. Both λέγειν and ποιεῖν affect the construction, which involves a sort of chiasmus. Cf. III 390 B note

τοῦ ἐνυπνίου πομπήν . Il. II 1— 34.

ἡ Θέτις κτλ. The verses are perhaps, as Schneider conjectures, from Aeschylus' Ὅπλων κρίσις, in which Thetis was one of the characters (Schol. on Ar. Ach. 883 ). Apollo with his harp ( ἔχων φόρμιγγα) appears as present at the marriage of Thetis also in Homer ( Il. XXIV 62, 63). Plato accommodates the beginning of the quotation to his own sentence: in Aeschylus perhaps it ran ὁ δ’ ἐνεδατεῖτο τὰς ἐμὰς εὐπαιδίας (so Butler, quoted by Schneider). ἐνδατεῖσθαι, ‘to dwell upon or emphasize,’ is elsewhere always used in an ominous sense (see Jebb on Soph. O. T. 205 ): and here too, perhaps, it strikes a foreboding note. The words μακραίωνας βίους were doubted by Stephanus, who suggested μακραίωνος βίου (so Euseb. Praep. Ev. XIII 3. 35) or μακραίωνας βίου: but Apollo's prophecies did not refer to Achilles only, so that the plural is justified. ἀπείρους should be taken not with εὐπαιδίας, but with βίους, which is in apposition to εὐπαιδίας. In the next line θεοφιλεῖς ἐμὰς τύχας depends on the compound expression παιῶν’ ἐπηυφήμησεν—a construction frequent in Aeschylus, especially with verbs which denote singing, celebrating, etc. ( Ag. 174, 175 al.): after enumerating all the blessings in store for Thetis ( ξύμπαντά τ’ εἰπών) Apollo raised a paean over her θεοφιλεῖς τύχας. This explanation— Schneider's—is much better than to connect ξύμπαντα adverbially with θεοφιλεῖς.

ἤλπιζον : ‘fancied,’ not ‘hoped’: cf. V 451 A, IX 573 C, and ἐλπίς in VII 517 B. This idiomatic usage is illustrated by Rutherford on Babrius 9. 2.

θεῖοι—οἷόν τε . The object of all worship and all religion, as of human action in general, is assimilation to God: cf. X 613 A note

πλούτου δὲ οὐδέν and Liebhold ( Fl. Jahrb. 1888, p. 107)

κόσμου δὲ ἄλλο μὲν <ἔχοντ)> οὐδέν for τοῦτον δὲ ἄλλο μὲν οὐδέν. Neither of these proposals has any plausibility, and it is best to regard this as one of the places where we owe the right reading to Π .

τὰ μὲν δὴ περὶ θεοὺς κτλ. Rettig ( Proleg. pp. 61 ff.) and others suppose that the virtue of ὁσιότης is alluded to here—a virtue which in the earlier dialogues is sometimes placed by the side of the four cardinal virtues ( Prot. 329 C , Men. 78 D , Gorg. 507 B ). But ὁσιότης is not specifically named (in spite of II 380 C), and it is clear from the words καὶ γονέας—ποιησομένοις that Plato is thinking at least as much of duty to man as of duty to gods: cf. II 378 B C, 381 E, 383 C. See also App. I.

ἀνδρεῖοι . Plato has in view chiefly courage in war: hence the importance which he attaches to removing the fear of death. Cf. Tyrtaeus 10 ( τεθνάμεναι γὰρ καλὸν κτλ.) and 12. 23 — 32. The poems of Tyrtaeus are not open to Plato's censure in this connexion. Pfleiderer ( Zur Lösung der Pl. Fr. p. 23) wrongly represents the present passage as tantamount (or nearly so) to a denial of the immortality of the soul, which is affirmed in Book X. It is possible to criticise the popular conception of immortality without disbelieving in a higher form of the same doctrine, and this is just what Plato does here.

καὶ περὶ τούτων τῶν μύθων should be taken with ἐπιστατεῖν rather than with λέγειν (sc. αὐτούς, i.e. τοὺς μύθους). Hartman, connecting the words with λέγειν, would expunge τῶν μύθων “cum poetae non de fabulis τὰ ἐν Αἵδου describentibus λέγειν soleant, sed ipsi Orci territamenta narrent”—a just criticism, and conclusive in favour of the construction which Hartman rejects.

λοιδορεῖν . The traditional literary picture of the Greek Hades deserves what Plato says of it (see the quotations in Nägelsbach Hom. Theol. pp. 397 ff., Nachh. Theol. pp. 396—398), although a brighter prospect was held out in the Eleusinian mysteries and the Orphic theology ( Nachh. Theol. pp. 398—407).

ἁπλῶς οὕτως . 11 377 B note

λέγοντας . For the accusative after the dative ἐπιχειροῦσι cf. Euthyph. 5 A , Crito 51 D. Before λέγοντας Ξ^{1} (with a few other MSS) adds ἄν, as if εἰ λοιδοροῖεν should be understood (cf. II 380 C); but we should supply not λοιδοροῖεν, but εἰ λοιδοροῦσι (Schneider).

τοῦδε τοῦ ἔπους κτλ. The singular ἔπος is sometimes used of more than one verse, e.g. Hdt. VII 143. The lines are addressed by the shade of Achilles to Odysseus: Od. XI 489—491. On the omission of ᾧ μὴ βίοτος πολὺς εἴη see Introd. § 5.

οἰκία ‐‐ θεοί περ . Il. XX 64, 65. The words in Homer are under the construction of δείσας—μή.

ὢ πόποι . The exclamation of Achilles when the ghost of Patroclus eludes his embrace: Il. XXIII 103, 104. On φρένες as the “physical basis of life” in Homer see Leaf ad loc.

οἴῳ ‐‐ ἀΐσσουσι . Tiresias retained in the other world something of the physical reality of his earthly existence: Od. X 493—495 τοῦ τε φρένες ἔμπεδοί εἰσιν· | τῷ καὶ τεθνηῶτι νόον πόρε Περσεφόνεια | οἴῳ πεπνῦσθαι· τοὶ δὲ σκιαὶ ἀΐσσουσιν. Plato allows the force of attraction to alter τοί to ταί: cf. Men. 100 A οἶος πέπνυται τῶν ἐν Ἅιδου, αἱ δὲ σκιαὶ ἀΐσσουσι.

ψυχὴ ‐‐ ἥβην . Il. XVI 856, 857. ῥεθέων, explained by the ancients as μέλη τοῦ σώματος (Hesych. s.v.), more probably denotes the mouth (as part of the face): cf. Leaf ad loc. and Il. IX 409. Leaf plausibly suggests that ἀν in ἀνδροτῆτα, ‘manhood’—found in all but two MSS of the Iliad—was only the written sign of the nasalis sonans, and counted as a short vowel.

ψυχὴ δὲ ‐‐ τετριγυῖα . Il. XXIII 100. “The voice,” says Leaf, “is as weak a copy of the living voice as is the εἴδωλον of the αὐτός”: whence τετριγυῖα and τετριγυῖαι again just below.

ὡς δ) ὅτε ‐‐ ᾔεσαν . Said of the souls of the suitors following Hermes down to Hades: Od. XXIV 6—9. Possibly we should read ἤισαν for ᾔεσαν (with Howes, Harvard Studies in Cl. Philol. VI p. 190).

ἐνέρους καὶ ἀλίβαντας . The Scholiast writes: ἐνέρους τοὺς νεκρούς, ἀπὸ τοῦ ἐν τῇ ἔρᾳ ( ὅ ἐστι γῇ) κεῖσθαι. Cf. ἔραζε. Early psychology scarcely separated the dead body from the surviving spirit: the latter still lived where the body lay ‘within the ground.’ Hence ‘those within the ground’ (opposed to the ἐπιχθόνιοι or living) became an expression for the spirits of the departed, and the denizens of the lower world in general: see Il. XV 188, XX 61. The Scholiast's derivation is more probable than that of Brugmann, who ( Grundriss II p. 180) derives the word from ἐν and a nominal suffix ‐ερο. Plato at any rate would have preferred the Scholiast. On ἀλίβαντας (not found in Homer or Hesiod) see Plut. Quaest. Symp. VIII 736 A (cited by Ast) ὁ δὲ ἀλίβας καὶ ὁ σκελετὸς ἐπὶ τοῖς νεκροῖς γέγονε, λοιδορουμένης ὀνόματα ξηρότητος. The ancients derived the word from ἀ and the root of λείβω λίψ etc., calling the dead ‘sapless’ διὰ τὴν τῆς λιβάδος ἀμεθεξίαν (Schol.). L. and S. object that the ἀ is long, relying perhaps on the line of Callimachus in Et. M. 63, 51 ἔβηξαν οἷον ἀλίβαντα πίνοντες (where ἀλίβαντα = ὄξος). There, however, the right reading may be ἁλίβαντα, i.e. οἱ ἀλίβαντα. But in Sophocles Fr. 751 ed. Dindorf the α is certainly long, unless the text is corrupt. Possibly the word is connected with ἠλίβατος; cf. Hesych. s. v. ἠλίβατον, where we are told that Στησίχορος Τάρταρον ἠλίβατον τὸν βαθὺν λέγει.

τούτου τοῦ τύπου . Instead of writing ἄλλα ὀνόματα ὅσα τούτου τοῦ τύπου ὄντα Plato writes ἄλλα ὄσα τούτου τοῦ τύπου ὀνομαζόμενα, with precisely the same meaning: τούτου τοῦ τύπου therefore depends on the copula involved in ὀνομαζόμενα. Stallbaum takes ὀνομαζόμενα as “quum pronuntiantur”; but this is pointless. The words mean simply ‘other names of this type which make all who hear them shudder’ etc.

φρίττειν δὴ ποιεῖ . The remark ὡς οἴεται, which appears in the best MSS— see cr. n.—after ποιεῖ gives no sense, and is admittedly corrupt. ὡς οἷόν τε, found in four inferior MSS besides q, is a rare phrase, occurring, I believe, nowhere else in Plato (except of course in combination with superlatives, e.g. III 412 B, VI 484 C), though found in Aristotle ( Pol. E 11. 1313^{a} 39, where Bekker conjectured οἴονται); but ‘to shiver as much as possible’ is painfully frigid. No emendation at all satisfactory has yet been proposed—neither Winckelmann's οἰκέτας, nor Hermann's ὅσα <*>τη (with reference to recitations of the rhapsodists!), nor Madvig's ὡς οἰητέα, nor Campbell's ὡς ἐτεά. Hertz ( Fl. Jahrb. 1872 p. 852) supposes the words to be a gloss by some Christian reader, meaning ‘as he’ (i.e Plato) ‘imagines.’ The author of the gloss wished to indicate that he at least could hear such tales without shivering. After ὡς οἴεται found its way into the text, it was probably altered to οἴονται (to suit the plural ἀκούοντας), from which οἷόν τε is a corruption: cf. 11 358 E, where q has οἴονται as against οἷόν τε of the best MSS. See also on VI 504 E.

καὶ ἴσως ‐‐ ἄλλο τι : “videlicet ad suavitatem et delectationem: v. p. 387 B, 390 A, 397 D, 398 A al.” (Stallbaum).

μὴ ἐκ ‐‐ ἡμῖν . φρίκη is a cold shiver, sometimes followed by sweat, whence ἐκ τῆς τοιαύτης φρίκης θερμότεροι. Cf. (with Hartman) Phaedr. 251 A ἰδόντα δὲ αὐτόν, οἷον ἐκ τῆς φρίκης, μεταβολή τε καὶ ἱδρὼς καὶ θερμότης ἀήθης λαμβάνει, where Thompson remarks that φρίκη is used by Hippocrates of the ‘cold fit of a fever.’ In θερμότεροι καὶ μαλακώτεροι Plato is thinking of the softening effect of heat upon iron: cf. (with J. and C.) infra 411 B ὥσπερ σίδηρον ἐμάλαξε, Laws 666 C, 671 B καθάπερ τινὰ σίδηρον τὰς ψυχὰς τῶν πινόντων διαπύρους γιγνομένας μαλθακωτέρας γίγνεσθαι; see also Il. XVIII 468—477 and Whitelaw on Soph. Ajax 651 in Cl. Rev. v pp. 66, 230. In so far as it associates heat with cowardice, the comparison breaks down, for heat meant courage to the Greeks. For this reason Stephanus conjectured ἀθερμότεροι and Ast ἀθυμότεροι, a reading afterwards found in v. Ast's conjecture is thus refuted by Hartman (l.c.): “Astii coniectura inepta est, quum ἀθυμία vitium sit, non vero iusta ac temperata μαλακία (dixit enim μαλακώτεροι τοῦ δέοντος).” In the next sentence Hartman expunges φοβούμεθα without sufficient cause.

τῶν ἐλλογίμων ἀνδρῶν : a subjective, not an objective genitive: see E below, and 388 E, 390 D εἴ πού τινες— καρτερίαι—καὶ λέγονται καὶ πράττονται ὑπὸ ἐλλογίμων ἀνδρῶν with X 605 D.

ὁ ἐπιεικὴς ἀνὴρ ‐‐ προσδεῖται . This passage is full of Socratic colouring. οὖπερ καὶ ἑταῖρός ἐστι contains a suggestion that only good men can be comrades: cf. Xen. Mem. II 6. 19 , 20 and Pl. Lys. 214 C . That death has no terrors for the good man is laid down in Ap. 41 C ff. The self-sufficiency of virtue was illustrated in the person of Socrates himself ( Mem. 1 2. 14. IV 8. 11), and continually preached by him ( Mem. II 6. 2, cf. IV 7. 1). Steinhart appears to me to exaggerate the force of αὐτάρκης when he characterises the doctrine of this passage as anti-christian ( Einleitung p. 160).

ὑέος . The fortitude of Pericles on receiving the news of the death of his two sons was a case in point, and may have been known to Plato. It is commemorated in a fine fragment of Protagoras preserved by Plut. Consol. ad Apoll. 33. 118 E, F.

ὀδύρεται, φέρει . See cr. n. The infinitives ὀδύρεσθαι and φέρειν are explained by Stallbaum as dependent on λέγομεν, but this is too harsh. The rhetorical repetition of ἥκιστ’ ἄρα proves that like στερηθῆναι they should be under the government either of δεινόν itself, or of some notion supplied out of δεινόν. As the former alternative gives the wrong sense we must, if the text is sound, take refuge in the latter. Hartman by a tour de force resolves ἥκιστα δεινόν into ἥκιστα εἰκὸς αὐτὸν δεδιέναι, and carries on the εἰκός. It would be somewhat easier, I think, though still very harsh, to supply δεινός out of δεινόν, δεινός being used as in δεινὸς καταράσασθαι τῷ λίθῳ (Theophr. Char. 15, cf. infra 395 C): but it is difficult not to believe that the text is corrupt. In q, καί has been corrected to χρή, and the insertion of δεῖ before καί is suggested by Hartman. The question however is not what the good man ought to do, but what he actually does, and for this reason Richards' ἔοικε after ὀδύρεσθαι is better, although otherwise unlikely. Stallbaum's alternative proposal to read ὀδύρεται, φέρει δέ seems to me far the best both in point of sense, and because it might easily pass into ὀδύρεσθαι, φέρειν δέ under the influence of στερηθῆναι. For these reasons I have printed it in the text. Cf. Introd. § 5.

ἄλλοτ) ‐‐ ἀτρυγέτοιο . The picture of Achilles sorrowing for Patroclus in Iliad XXIV 10—12. Plato accommodates the Homeric narrative to his own ποιεῖν, and reads πλωΐζοντ’ — ἀτρυγέτοιο instead of δινεύεσκ’ ἀλύων παρὰ θῖν’ ἁλός, which appears in our Homer. πλωίζω elsewhere is always used of sail ing in the literal sense (yet ἐκ τοῦ νοῦ ἐκπλώειν in Hdt. VI 12), but it cannot bear such a meaning here. If the MSS are right, πλωΐζοντ) must be regarded (with Schneider) as a metaphor, the agitated movements of Achilles being compared to the unsteady motion of a ship upon the sea. Achilles is so to speak ‘at sea’ and shews it in his gait; cf. the metaphorical sense of χειμάζομαι. The picture savours of the burlesque, and Howes suggests that πλωίζων may be a deliberate parody on Plato's part ( Harvard Studies etc. VI p. 202). As no other example of such a use of πλωίζω has been adduced, the word is perhaps corrupt. Heyne's πρωΐζοντ) “matutinum se agentem” ( οὐδέ μιν ἠὼς | φαινομένη λήθεσκεν ὑπεὶρ ἅλα, says Homer) will never command a wide assent: still less πλώϊσοντ) (Benedictus), πρὼ| ἴοντ) (Ast), whose quantity is not above suspicion, or πρὼ| ἰύζοντ) (Liebhold Fl. Jahrb. 1888, p. 108). αἰάζοντ) (Herwerden and Naber) is better in point of sense, but the alteration is too great. I have thought of πόλλ’ ᾤζοντ) ( ᾤζεις ‘cry ὤ’ and not ὤζεις is the spelling of the Codex Mediceus in Aesch. Eum. 124 ), or ἀφλοίζοντ) (cf. ἀφλοισμός in Il. XV 607). Perhaps, however, πλωΐζοντ) conceals some word meaning ‘to rush wildly from his tent,’ ἐπὶ θῖν) being probably for ἐπὶ θῖνα, not for ἐπὶ θινί. There is apparently a contrast between Achilles' anguish within his tent and without, and some word is needed to mark his exit. Nothing can be made of the variant πλάζοντ) (in a few inferior MSS). In default of anything better we must (I suppose) provisionally acquiesce in Schneider's interpretation.

μηδὲ ‐‐ κεφαλῆς . Il. XVIII 23, 24.

ἐκεῖνος . Homer.

ἐγγὺς θεῶν . Zeus was Priam's seventh ancestor (Apollod. III 12). The phrase has a dash of old-world romance about it: cf. 391 E infra and Stallbaum on Phil. 16 C οἱ μὲν παλαιοί, κρείττονες ἡμῶν καὶ ἐγγυτέρω θεῶν οἰκοῦντες.

λιτανεύοντά τε ‐‐ ἕκαστον . Il. XXII 414, 415.

ὤμοι κτλ. Said by Thetis in Il. XVIII 54.

ὢ πόποι . Il. XXII 168, 169. The words are uttered by Zeus with reference to Hector. For ἄστυ our Homer has τεῖχος.

αἲ αἲ ‐‐ δαμῆναι . Il. XVI 433, 434. The only variant is ὤμοι for αἲ αἴ.

εἰ καὶ ἐπίοι αὐτῷ . καί is not ‘even’ (J. and C.), otherwise there would be too much emphasis on ἐπίοι, but ‘also’: ‘if it should also occur to himself’ (sc. as Homer says it occurs to gods). The emphatic word is αὐτῷ. For τοιοῦτον Hartman requires either τοιοῦτόν τι or τὸ τοιοῦτον; but cf. 416 B, IV 426 B, 429 E, IX 590 E and II 368 A note

σμικροῖσιν . See on I 330 B. ἐπὶ σμικροῖσι παθήμασι has a poetical rhythm, and may possibly be from a hexameter.

ἕως ἄν τις ‐‐ πείσῃ . Cf. Phaed. 85 C , D, Gorg. 527 A .

ἐφιῇ ‐‐ τοιοῦτον . See cr. n. The present ἐφιῇ is slightly better than ἐφῇ: for τὸ τοιοῦτον denotes the state or condition rather than the act. ἐφῇ comes rather nearer to the reading of A and II, and is preferred by Baiter and Hartman. For ζητεῖ H. Wolf conjectured ποιεῖ, Herwerden τίκτει or ἐντίκτει, in both cases needlessly: cf. with J. and C. ἐθέλει in 11 370 B. The sentiment is generalised in VIII 563 E.

οὔτε ἄρα . οὔτε followed by δέ is rare (examples in Kühner Gr. Gr. II p. 832) but δέ follows τε very often, especially in πολὺ δέ, μέγιστον δέ etc.: see II 367 C note Cobet's οὔτἆρα i.q. οὔτοι ἄρα, though approved by Hartman, is therefore unnecessary.

οὐκοῦν ‐‐ λόγον . The lines are Il. I 599, 600. Hermann wished to read οὔκουν and reject ἀποδεξόμεθα περὶ θεῶν, placing τὰ τοιαῦτα under the government of ἀποδεκτέον. οὔκουν may be right, but the change is not necessary. τὰ τοιαῦτα does not refer specifically to the verses, but means τὸ κρατεῖσθαι ὑπὸ γέλωτος and the like; while the two verses are themselves the object of ἀποδεκτέον. I have accordingly placed a colon after θεῶν and removed the pause after ποιπνύοντα; a remedy which removes, I think, the objections felt by Hermann to ἀποδεξόμεθα περὶ θεῶν, and by Herwerden to περὶ θεῶν. The asyndeton in ἄσβεστος δ’ ἄῤ etc. is common in ampliative and illustrative sentences.

ἀλλὰ μὴν καὶ ἀλήθειαν ff. On the place of this section in the general plan of the Republic see App. I.

θεοῖσι μὲν ‐‐ εἴδει . For the dative θεοῖσι see I 330 B note ἐν φαρμάκου εἴδει (cf. 11 382 C, D) implies the usual Socratic analogy between body and soul: see on II 380 B.

οὐχ ἁπτέον κτλ. Cf. Laws 916 E ff.

τοιούτους is omitted by Hartman, and is certainly open to doubt. The balance of MS evidence is in its favour, although a few inferior MSS and one MS of Stobaeus ( Flor. 46. 95), agree with A^{1} in omitting it. It must either mean rulers who act ἐπ’ ὠφελίᾳ τῆς πόλεως, or else such rulers as Plato's. The former alternative is not altogether satisfactory, and it is difficult not to believe that Plato was in reality referring to his own rulers. The serious objection to this view is that we have not yet heard anything of Plato's rulers: they are not described till 412 B. I think the solution may be that the present section on truth is a later addition made by Plato after he had written his first account of the rulers in Book III. See also App. I.

λέγοντι has caused difficulty, and Madvig would expunge the word. The explanation is simple enough. μὴ τἀληθῆ λέγειν should be repeated between ἤ and πρός, and μὴ τὰ ὄντα λέγοντι ὅπως taken closely together, ‘or to lie’ ( μὴ τἀληθῆ λέγειν understood) ‘to a pilot about the ship and its crew by misrepresenting the facts about one's own condition etc.’ One MS of Stobaeus (l.c.) has λέγοντα, which is also possible, and could only be explained in this way. I have removed the comma usually printed after λέγοντι.

λαμβάνῃ : sc. ὁ ἄρχων. Cf. 1347 A note λαμβάνῃς (Ficinus and Benedictus) gives a wrong sense.

τῶν οἳ ‐‐ δούρων . Od. XVII 383, 384. κακῶν is of course neuter. If Schneider could shew that this quotation refers to a case in which a chieftain in Homer did or did not punish a δημιουργός for lying, he would make out a prima facie case for his view that Plato is here prescribing canons for poetical representations, but there is nothing of this in Homer; and we must suppose that Plato is speaking here of his own citizens. See App. I.

ἐάν γε ‐‐ τελῆται does not mean ‘if our theory is carried out’ (J. and C.) or ‘if our ideal city is ever realised’ (Rettig). Such a remark would be frigid and superfluous. The meaning is merely that the ruler will first use words, but, if these fail, he will afterwards proceed to deeds i.e. κολάσει. The first γε assents: the second enters a caveat. ἔργα τελῆται= ἔργων τέλος γίγνηται.

σωφροσύνης δὲ ‐‐ μέγιστα : ‘for the mass of men, are not the cardinal points of temperance such as these?’ (Jebb on Soph. O. C. 20 μακρὰν γὰρ ὡς γέροντι προὐστάλης ὁδόν—a precise parallel). There is no authority for interpreting these words (with Stallbaum, Hartman etc.) as ‘plerumque’ ‘in universum.’ Plato is warning us not to regard his account of σωφροσύνη here as scientifically accurate and complete. It is the most obvious and conspicuous aspects of self-control which poets should chiefly impress upon the multitude, and to these Plato confines his attention. On the Greek conception of σωφροσύνη see the passages collected by Nägelsbach, Nachhom. Theol. pp. 227 ff.

Ὁμήρῳ . For this Ξ and a few other MSS read παρ’ Ὁμήρῳ. Schneider successfully defends Ὁμήρῳ by Arist. Pol. Θ 5. 1339^{b} 7 οὐ γὰρ ὁ Ζεὺς αὐτὸς ᾄδει καὶ κιθαρίζει τοῖς ποιηταῖς. The line is addressed by Diomede to Sthenelus in Il. IV 412.

τὰ τούτων ἐχόμενα . The two verses which Plato here quotes do not follow τέττα, σιωπῇ κτλ., and do not even occur together in our Homer. ἴσαν— Ἀχαιοί is from Il. III 8 ( οἱ δ’ ἄρ’ ἴσαν σιγῇ μένεα πνείοντες Ἀχαιοί), σιγῇ—σημάντορας from IV 431. Some editors bracket the first verse, but (as Hartman points out) it is not likely that a scribe should have interpolated a line from Il. III before one from Il. IV. Plato may be guilty of ‘contamination,’ or the lines may really have occurred together in his text of Homer. J. and C. suggest that Plato perhaps did not mean the lines to be connected. The objection to this view is that σιγῇ (as in our text of Homer, though there it is in a different place) goes best with ἴσαν, and that ἴσαν μένεα πνείοντες Ἀχαιοί is not by itself an illustration of obedience to rulers, and therefore would not be relevant here. See on the whole subject of Platonic quotations from Homer, Howes in Harvard Studies etc. VI pp. 153—237, with whose conclusions (p. 210) I heartily agree.

οἰνοβαρὲς κτλ. Achilles to Aga memnon in Il. I 225. The point of this illustration is not in the abusive epithets, but in the insubordination which they and the rest of the speech ( τὰ τούτων ἑξῆς) express.

νεανιεύματα . See cr. n. The spelling seems established by the verb νεανιεύεσθαι: e.g. Gorg. 482 C . νεανισκεύματα has however some authority, for νεανισκεύομαι was used (Photius s.v.). νεανικεύματα, to say the least, is doubtful, nor is νεανικοῦν (Photius s.v.) enough to justify such a form, in spite of Schneider ( Addit. p. 19).

παρὰ πλέαι ‐‐ δεπάεσσι . Odysseus in Od. IX 8—10. Our text of Homer has παρὰ δἐ πλήθωσι. I have written παρὰ πλέαι for παράπλειαι or παραπλεῖαι of nearly all the MSS. Vat. r and Vind. B have ε ι περιπλεῖαι, Cesenas M παράπλειαι (sic). παράπλειαι (which Howes l.c. p. 205 thinks Plato found in his text of Homer) is in reality a vox nihili; even if it did occur, it could not mean ‘almost full,’ as L. and S. say: and such a meaning would be ludicrously inappropriate here. With παρὰ δἐ πλέαι cf. Anacr. 94. 1 ed. Bergk κρητῆρι παρὰ πλέῳ οἰνοποτάζων. See my article in Cl. Rev. XI p. 349.

λιμῷ δ) ‐‐ ἐπισπεῖν . Od. XII 342.

ἢ Δία ‐‐ ἐπιλανθανόμενον . μόνος ἐγρηγορώς refers to Il. II 1—4: the incident itself is narrated in Il. XIV 294 ff. For the postponement of the relative ἅ cf. IV 425 C. The effect is to throw emphasis on μόνος ἐγρηγορώς—that Zeus should forget what he had purposely kept awake to devise makes the scandal all the worse —and brings it into sharper contrast with καθενδόντων—ἀνθρώπων. ὡς must be taken with ἐπιλανθανόμενον, the construction being ἢ ποιεῖν Δία ὡς ἐπιλανθανόμενον: cf. II 383 A. Stallbaum explains ὡς μόνος ἐγρηγορώς as “ut solus vigil”: while J. and C. supply ἀκούειν after ἤ. Neither view seems to me at all satisfactory. The text has been often called in question. Instead of ὡς Hermann reads καί: Herwerden and Richards suggest ὅσα (dropping ἅ before ἐβουλεύσατο). The best emendation is perhaps Jackson's εἷς for ὡς ( Journal of Phil. IV p. 147), but I see no good reason why ὡς cannot be construed with ἐπιλανθανόμενον. The pause which on this view is necessary after ὡς helps still further to increase the stress on μόνος ἐγρηγορώς, which Plato certainly intended to emphasize.

βουλόμενον ‐‐ τοκῆας . βουλόμενον is not otiose after ἐθέλειν (as Hartman alleges): ‘to wish’ ( βούλεσθαι) and ‘to be willing’ ( ἐθέλειν) are different ideas. The same critic also rejects καί before λέγοντα “quia ea verba excusationem τοῦ ἐθέλειν humi consuescere continent”; but it is more effective to represent so gross an utterance as an additional part of the picture. For φοιτᾶν πρός cf. Lys. I 15, 19. where the meaning is the same. Herwerden should not have wished to replace the preposition by παρά. In Homer the line εἰς εὐνὴν φοιτῶντε φίλους λήθοντε τοκῆας ( Il. XIV 296) is not said by Zeus, as Plato—doubtless intentionally, to increase the effect—makes it appear to be.

Ἄρεως ‐‐ δεσμόν . Od. VIII 266 ff. δεσμόν is still under the government of πο<*>εῖν.

καὶ λέγονται καὶ πράττονται κτλ. : ‘are either described or done by famous men’ etc.: described e.g. in poetry by Homer's heroes, or done in actual life before our eyes. θεατέον refers to πράττονται, ἀκουστέον to λέγονται by the usual chiasmus. J. and C. translate “performed by famous men or told concerning them,” understanding περὶ ἐλλογίμων ἀνδρῶν with λέγονται, but this cannot be right.

στῆθος δὲ ‐‐ ἔτλης . Odysseus in Od. XX 17, 18.

δωροδόκους κτλ. The excessive love of money is a sign of ἀκράτεια: so that its mention here is relevant enough, although the vice was not specifically named in 389 D.

δῶρα ‐‐ βασιλῆας : an old saying attributed by some to Hesiod ( οἱ μὲν Ἡσιόδειον οἴονται τὸν στίχον Suidas s. vv. δῶρα κτλ.). It is referred to by Eur. Med. 964 πείθειν δῶρα καὶ θεοὺς λόγος. Cf. Nägelsbach Nachhom. Theol. II p. 64.

συμβουλεύων . Il. IX 515 ff. The gentitive μήνιος, for which a few MSS read μήνιδος, is natural in paraphrasing Homer. Cf. the form Θάλεω in X 600 A.

οὐδ’ ὁμολογήσομεν . “Dele futile interpretamentum” exclaims Hartman. The words are genuine, and add a new point: cf. 391 A φάναι καὶ ἄλλων λεγόντων πείθεσθαι.

δῶρα λαβεῖν . Il. XIX 278 ff. Plato is unjust to Achilles: see ib. 147 ff. (J. and C.).

τιμὴν κτλ. Il. XXIV 502, 555, 594.

ἄλλως ‐‐ ἐθέλειν is again unfair: see Il. l.c. 560.

ἔβλαψας ‐‐ παρείη . Il. XXII 15, 20.

ποταμόν . Scamander: Il. XXI 130—132, 212—226, 233 ff.

καὶ αὖ κτλ . ὡς should be repeated with ἔφη (J. and C.). Herwerden rejects both τοῦ and Σπερχειοῦ, the former because he thinks the article would suggest the Simois. Why should it not specify the other river towards which Achilles (according to Plato) shewed insubordination? Plato (as Hartman remarks) has just as much right to mention the river's name as that of Achilles' tutor (390 E). The reference is to Il. XXIII 140—151. Although the locks were ‘sacred to Spercheius,’ the vow was nevertheless conditional on Achilles' safe return, which he knew was hopeless. This is the reason which Achilles gives for offering his locks to the shade of Patroclus rather than to Spercheius: ib. 150. ὀπάσαιμι—‘suffer me to give’—is in reality a prayer to the Spercheius.

Ἕκτορος ἕλξεις . Il. XXIV 14 ff.

σφαγάς Il. XXIII 175 ff.

τρίτου ἀπὸ Διός . Peleus' father, Aeacus, was son of Zeus.

μηδὲ ‐‐ μηδέ . Bekker read μήτε— μήτε; but μηδὲ τάδε is of course ne haec quidem.

ὥρμησαν ‐‐ ἁρπαγάς . Pi rithous assisted Theseus to abduct Helen: and Theseus Pirithous in his attempt to carry off Persephone from the lower world. οὕτως belongs to δεινάς: the order is regular and idiomatic: cf. Ap. 36 A , Symp. 192 C al. Sophocles and Euripides each wrote a play called ‘Theseus’: but Plato is probably alluding to some epic Theseis. Cf. Kinkel Epic. Gr. Frag. p. 217.

αὐτά is censured by Heller, who conjectures τοιαῦτα, while Hartman keeps αὐτά but rejects ἔργα. Stallbaum says we should expect ταῦτα for αὐτά: but ταῦτα would be too precise. αὐτά means simply ‘the actions in question.’ Cf. I 339 E note The turn of the sentence recalls II 380 A ἢ οὐ θεοῦ ἔργα ἐατέον αὐτὰ λέγειν ἢ κτλ. Cf. also infra 408 C.

κακά . Hartman approves Cobet's conjecture κακούς, “cum γεννᾶν hic translaticiam vim non obtineat.” Why not? Cf. κακὰ γίγνεσθαι just below. κακούς would be extremely tame and commonplace.

ἐν τοῖς πρόσθεν . II 378 B, 380 C.

ἄρα : II 358 C note

οἱ ‐‐ δαιμόνων . From Aeschylus' Niobe: see Dindorf Fr. 155. The passage is also quoted in part by Strabo (XII 8. 21), from whom it appears that Niobe is the speaker, and that οἱ θεῶν ἀγχίσποροι are her father Tantalus and his kindred ( οἱ περὶ Τάνταλον). ὧν—αἰθέρι means ‘whose is the altar to ancestral Zeus on Mount Ida high in heaven,’ i.e. their θεὸς πατρῷος is Zeus (who was Tantalus' father), and they worship him on the heights of Ida. Tantalus' territory extended to Ida: see Strabo l.c. ὁ Τάνταλος λέγει σπείρω δ’ ἄρουραν δώδεχ’ ἡμερῶν ὁδόν, | Βερέκυντα χῶρον, ἔνθ’ Ἀδραστείας ἕδος | Ἴδη τε μυκηθμοῖσι καὶ βρυχήμασιν | πρέπουσι μήλων. For ὧν κατ’ Ἰδαῖον πάγον Strabo has οἷς ἐν Ἰδαίῳ πάγῳ, a much inferior reading. καί before οὔπω may be Plato's (so Stallbaum and others),—in which case the last line is from a different part of the play,—but is much more likely to come from Aeschylus, the resolution of κοὔπω being due to Plato. The line follows naturally on the others, and is not sufficiently important to have been selected from a different context. The verses are complete in themselves, and present a stately picture of the sons of the gods, which is the only reason why they are cited here.

τί οὖν κτλ. This is the ἀληθὲς εἶδος λόγων. Plato has prescribed canons for the ψευδεῖς λόγοι or legends about gods etc.; but rules for ἀληθεῖς λόγοι, i.e. λόγοι relating to men and human affairs, cannot be drawn up without begging the conclusion which the Republic seeks to establish. See also on II 376 E.

ἡμῖν . See cr. n. Without ἡμῖν, we should have τοῖς λόγων πέρι ὁριζομένοις. I agree with Hartman and the majority of editors in retaining the word. See Introd. § 5.

ἀδύνατον δή . For δή Stallbaum approves Ast's conjecture δέ. δέ would be too weak, if the meaning were adversative, but it is not. δή is only ‘well’: cf. II 368 A (Schneider).

καὶ ποιηταὶ καὶ λογοποιοί . On λογοποιοί see II 365 E note; and for the statement itself Laws 660 E ff., 662 B.

ἀλλότριον ‐‐ ἀγαθόν . I 343 C note

ζητοῦμεν . Stallbaum's conjecture —see cr. n.—is now generally accepted. ἐζητοῦμεν would imply that the discussion had changed, but it has not. Cf. IV 420 C ὃ πάλαι ζητοῦμεν.

τότε διομολογησόμεθα κτλ. This is not “an ironical or fanciful excuse for varying the order of the subject” (J. and C.), for if Socrates declared at this stage that justice is a good for its possessor he would in point of fact be presupposing the results of the whole investigation. See IX 588 B—592 B. Others (e.g. Hirzel der Dialog p. 237 note) have taken τότε διομολογησόμεθα as a hint of the additional discussion on Poetry in Book X: but there is nothing either here or in that book to justify any such interpretation. Cf. X 595 A note What Plato's regulations about λόγοι περὶ ἀνθρώπων would have been may be easily gathered from the end of Book IX and X 608 C ff., although the subject is nowhere specifically and expressly resumed in the Republic. Cf. I 347 E note

τὸ δὲ λέξεως . Hartman approves the variant τὰ δὲ λέξεως: but the subject of λέξις is better treated as a unity until it has been subdivided.

μυθολόγων ἢ ποιητῶν . μυθολόγων is said so as to include writers of μῦθοι in prose: cf. 394 B and II 365 E note

ὥσπερ οὖν κτλ. Plato means that poor speakers cannot grapple with an abstract notion, but use a part of it, i.e. a concrete example. οὐ κατὰ ὅλον κτλ. may be illustrated from Symp. 205 B , C.

καὶ ἐλίσσετο ‐‐ λαῶν . Il. I 15, 16. Leaf. reads λίσσετο because “ λίσσομαι apparently had a second initial consonant, and is never preceded by a short vowel.” The word had probably been Atticised by Plato's time.

δοκεῖν ‐‐ ὄντα . δοκεῖν is here ‘to fancy’ not ‘to seem.’ Contrast II 381 E ἡμῖν δὲ ποιοῦσι δοκεῖν σφᾶς παντοδαποὺς φαίνεσθαι—a passage which is cited by Hartman to justify ποιῆσαι as against the variant πεῖσαι.

περί τε τῶν ‐‐ παθημάτων . This clause is rejected by Herwerden. The difficulty—which lies in the collocation of Ἰθάκῃ the place and Ὀδυσσείᾳ the poem—is no doubt lessened by reading (with Richards) καὶ ἐν or κἀν before ὅλῃ, but does not wholly disappear. Possibly the last twelve books of the Odyssey, in which the scene is Ithaca, were sometimes known collectively as Ἰθάκη.

ὅτι ἦλθεν ‐‐ βασιλέων paraphrases Il. I 12—16.

ὡς Χρύσης γενόμενος : ‘as if he had been transformed into Chryses,’ not merely ‘in the person of Chryses’ (Jowett). In ‘simple narrative’ he is Homer: when Chryses begins to speak, he becomes Chryses. Cf. 393 B ὥσπερ αὐτὸς ὢν ὁ Χρύσης (‘as if he himself were Chryses’).

ἐλθὼν ‐‐ αἰδεσθέντας . Il. I 17—21. The emphatic αὐτοὺς accurately represents Homer's ὑμῖν μέν. For λῦσαι H. Wolf conjectured ἀπολῦσαι; but Plato is closely following Homer, who has λύσαιτε. τὸν θεόν is Apollo.

ταῦτα δὲ ‐‐ βέλεσιν . Il. I 22— 42. The paraphrasis is accurate, and Plato leaves nothing essential out. There is no sign that his text differed from ours in this passage.

μὴ ‐‐ οὐκ ἐπαρκέσοι . ἐπαρκέσοι presupposes ἐπαρκέσει in the narratio recta: Homer has μή νύ τοι οὐ χραίσμῃ σκῆπτρον καὶ στέμμα θεοῖο. It is usual to regard this sentence as final: if so, it is the solitary instance in Plato where the future after a final μή must be admitted. See Weber in Schanz's Beiträge II 2, p. 60 and Goodwin MT. pp. 45, 91. The nearest parallel is Euthyph. 15 D ἀλλὰ καὶ τοὺς θεοὺς ἂν ἔδεισας παρακινδυνεύειν, μὴ οὐκ ὀρθῶς αὐτὸ ποιήσοις, where μή depends on a verb of fearing. It is better, both in point of grammar and of sense, to regard this sentence also as expressing apprehension (‘for fear lest’), although no verb of fearing is present. It is not final in any proper sense of the word. Bekker read ἐπαρκέσειε, saying that Θ has ἐπαρκέσειεν.

μὴ ἐρεθίζειν . Valckenaer's conjecture μή ἑ ἐρεθιζειν ( μή μ’ ἐρέθιζε in Homer) is attractive in view of τὰ ἃ δάκρυα in 394 A for Homer's ἐμὰ δάκρυα, and because it provides an object for ἐρεθίζειν. Plato uses the pronoun tolerably often (e.g. in I 327 B, X 617 E, Symp. 175 C , 223 B): other Attic writers seldom, if ever (Kühner-Blass Gr. d. Gr. Spr. I p. 592). It is not however clear that ἐρεθίζειν could not be used without an object expressed, and I therefore revert to the MS reading.

ἐν ναῶν οἰκοδομήσεσιν shews that Plato understood Homer's ἔρεψα ( εἴ ποτέ τοι χαρίεντ’ ἐπὶ νηὸν ἔρεψα) of building. According to Leaf, ἔρεψα seems to denote the most primitive form of temple —“a mere roof to protect the image of a god standing in a grove.”

τεῖσαι ‐‐ βέλεσιν . Ἀχαιούς is of course the subject to τεῖσαι (‘pay for,’ ‘expiate’): in Homer it is τίσειαν Δαναοὶ ἐμὰ δάκρυα σοῖσι βέλεσσιν. The translation ‘that he would avenge his tears upon the Achaeans’ (D. and V.) is wrong. ἅ is apparently a solitary instance of ὅς=‘suus’ in Attic prose (Kühner-Blass l.c. I 1, p. 602). Plato chooses the word because it expresses Homer's ἐμά briefly and neatly, rather than from any conscious desire to make the paraphrase archaic.

τραγῳδίας . Adimantus quotes a single concrete instance—‘tragedies’—to shew that he now apprehends the meaning of μίμησις. Socrates, out of politeness and because he wishes to make progress, interprets this as a recognition of the imitative character of Tragedy and Comedy in general ( ὥσπερ σὺ λέγεις τραγῳδία τε καὶ κωμῳδία), as in point of fact it virtually is. ὥσπερ σὺ λέγεις is not true in the beggarly literal sense of λέγειν, but it is sufficiently so for polite conversation. To insert—with Herwerden and Hartman— τε καὶ κωμῳδίας after τραγῳδίας seems to me unnecessary and pedantic.

εὕροις δ) ἂν ‐‐ διθυράμβοις . The dithyramb was at first purely narrative or nearly so; it afterwards became mimetic (Arist. Probl. XIX 15. 918^{b} 19). Only one of Pindar's dithyrambic fragments appears to be ‘mimetic’ ( Frag. 74). On the growth and decline of the Dithyramb see Smyth Greek Melic Poets pp. xliii—lviii.

τε ‐‐ δὲ καί . II 367 C note

εἴ μοι μανθάνεις : ‘if I can make you understand,’ with reference to μανθάνω in 392 C, 394 B, C. Heindorf's εἴ μου μανθάνεις (as in Phil. 51 C ) is attractive, but the corruption is not easy to explain, and the MS reading is sufficiently defended by I 343 A ὅς γε αὐτῇ οὐδὲ πρόβατα—γιγνώσκεις (so also Hartman).

τοῦτο ‐‐ αὐτό refers to ὅτι χρείη— μιμεῖσθαι, and ἔλεγον is ‘was saying’ i.e. ‘was trying to say,’ viz. when I digressed.

εἴτε παραδεξόμεθα κτλ. Krohn ( Pl. St. p. 13) declares this passage to be inconsistent with II 373 B, where ὑποκριταί, χορευταί, ἐργολάβοι are admitted. He forgets or ignores the fact that in § 373 Plato is describing the τρυφῶσα πόλις, which he is now engaged in ‘purging’ (399 E). See II 372 D note

ἴσως δὲ ‐‐ τούτων . In this remark J. and C. find “an anticipation of the condemnation of epic poetry in Book X.” I cannot see that it does more than prepare the way for ἀλλ’ ὅπῃ ἂν—ἰτέον. See on X 595 A.

πότερον μιμητικοὺς κτλ. The question is not ‘Are our guardians to become dramatic poets?’ but ‘Are they to have the imitative habit of mind?’ The answer is in the negative, and the drama is banished because it fosters this habit in spectators. Cf. 395 D note

ὅτι ‐‐ πολλὰ δ) οὔ explains τοῖς ἔμπροσθεν, as Hartman points out, and not τοῦτο, as D. and V. translate. ἔμπροσθεν refers to II 370 B.

πολλῶν κτλ. suggests, perhaps intentionally, πόλλ’ ἠπίστατο ἔργα, κακῶς δ’ ἠπίστατο πάντα. The words ὥστ’ — ἐλλόγιμος—equivalent to a neuter accusative—are undeservedly cancelled by Herwerden and Hartman. Translate ‘he will fail in all of them to attain creditable distinction’: cf. the adverb κακῶς in κακῶς δ’ ἠπίστατο πάντα.

οὐκοῦν κτλ. The reasoning is a fortiori: if two or more departments of merely imitative art cannot be represented by the same person, still less can imitation be combined with any serious pursuit ( σχολῇ ἄρα κτλ.).

οὐδὲ τὰ δοκοῦντα ‐‐ ποιοῦντες . The reverse is affirmed by Socrates in Symp. 223 D τοῦ αὐτοῦ ἀνδρὸς εἶναι κωμῳδίαν καὶ τραγῳδίαν ἐπίστασθαι ποιεῖν, καὶ τὸν τέχνῃ τραγῳδοποιὸν ὄντα καὶ κωμῳδοποιὸν εἶναι. The solution is that in the Symposium Socrates is applying to the drama the Socratic principle μία ἐπιστήμη s. δύναμις τῶν ἐναντίων: theoretically, therefore, and ideally, the tragedian is also capable of writing a comedy. In the Republic, on the other hand, he is describing Greek dramatic art as he found it: for which reason he writes δύνανται and not δύναιντ’ ἄν (a corruption in v, wrongly adopted by Stallbaum). Cf. Ion 534 C. Aristophanes did not write tragedy, nor the tragedians comedy. The passage in the Symposium is interesting as an unconscious prophecy of the Shakespearian drama. Cf. Reber Plato u. d. Poesie p. 11.

μιμήματα . See cr. n. Former editors variously read μιμήματα or μιμήματε. Either is admissible, so far as concerns the Greek, but the plural was perhaps— owing to the proximity of τούτω—somewhat more likely to be corrupted to the dual in this instance than vice versâ. Cf. X 614 C δύο—χάσματα ἐχομένω ἀλλήλοιν with note ad loc. The reading μιμήματά τε represents the correction μιμήματα. This is, I think, a somewhat simpler view than to suppose that an original μιμήματε became μιμήματέ τε by dittography, and τέ was afterwards changed to τά. Roeper, however, pronounces in favour of the dual ( de dual. usu Pl. p. 14), and it must be admitted that duals are peculiarly liable to corruption in the MSS of the Republic. See Introd. § 5.

ῥαψῳδοί ‐‐ ὑποκριταί . Even ῥαψῳδοί seem to have generally confined themselves to a particular poet: see Ion 531 C, 536 B.

ἀλλ) οὐδὲ ‐‐ οἱ αὐτοί . This was true without exception till comparatively late times: see Müller Gr. Bühnenalt. pp. 185—188. κωμῳδοῖς and τραγῳδοῖς (literally ‘at the tragedians’ etc.) are local—almost adverbial—datives, regularly used to denote the exhibitions of comedies and tragedies: see e.g. Arist. Eth. Nic. IV 6. 1123^{a}23, Aesch. in Ctes. 36, and cf. the Latin use of ‘gladiatoribus’ for ‘at a gladiatorial show.’

ἢ αὐτὰ ἐκεῖνα πράττειν . καλῶς should be repeated with πράττειν, and ἤ is simply ‘or,’ not ‘or else.’ The alternative rendering given by J. and C. ‘or else—if able to imitate—is not able to do the things themselves,’ does violence to both grammar and sense.

δημιουργοὺς ἐλευθερίας . An artificial and somewhat strained expression, selected in order at once to compare and contrast the guardians with other artists. They too are artists, and their ἔργον is Freedom. To ἐλευθερία Plato attaches his own meaning: true freedom lies in the subordination of the lower to the higher, both in private conduct and in political life: cf. Xen. Mem. I 2. 5 , 6 and infra IX 577 D, E, X 617 E notes It is in this sense that ἐλευθέρους is used below.

τούτοις : viz. τοῖς δημιουργοῖς ἐλευθερίας τῆς πόλεως.

ἵνα μὴ ‐‐ ἀπολαύσωσιν reveals the object of this attack upon the drama: cf. II 383 C and infra 401 B. An admirable illustration of the sentiment is quoted by Susemihl from Plut. Sol. 29. 6 μετὰ δὲ τὴν θέαν προσαγορεύσας (sc. ὁ Σόλων ) αὐτὸν (viz. τὸν Θέσπιν) ἠρώτησεν, εἰ τοσούτων ἐναντίον οὐκ αἰσχύνεται τηλικαῦτα ψευδόμενος. φήσαντος δὲ τοῦ Θέσπιδος μὴ δεινὸν εἷναι τὸ μετὰ παιδιᾶς λέγειν τοιαῦτα καὶ πράσσειν, σφόδρα τῇ βακτηρίᾳ τὴν γῆν ὁ Σόλων πατάξας Ταχὺ μέντοι τὴν παιδιάν, ἔφη, ταύτην ἐπαινοῦντες καὶ τιμῶντες εὑρήσομεν ἐν τοῖς συμβολαίοις. To omit μή (with A and a few other MSS), and govern ἵνα by μιμεῖσθαι above is grammatically difficult, and gives an unsatisfactory sense. The genitive τοῦ εἶναι has been called in question by Hartman (following Ast) on the ground that “qui τοῦ εἶναι (sc. αἰσχροὶ) ἀπολαύσωσιν iam sunt turpitudine infecti.” This would be true, if Plato had written the present ἀπολαύωσιν, but the aorist is ingressive, and τοῦ εἶναι ἀπολαύσωσιν is virtually equivalent to γένωνται τοῦθ) <*> μιμοῦνται. Few will acquiesce in Ast's conjecture τὸ εἶναι, or in Stallbaum's view that τοῦ εἶναι is a partitive genitive.

καὶ κατὰ σῶμα ‐‐ διάνοιαν . For σῶμα Stallbaum conjectured σχῆμα, but Plato would surely have said σχήματα, as in 397 B. Hartman boldly ejects κατὰ φωνάς and reads καὶ κατὰ <τὸ> σῶμα καὶ κατὰ τὴν διάνοιαν, remarking that κατὰ τὸ σῶμα by itself includes “gestus, habitus, vocem, vultum, similia.” This is in a sense true, but there is no reason why one particular instance of physical resemblance should not be selected for special remark. Plato differentiates the external from the internal characteristics by combining σῶμα and φωνάς under a single preposition, and repeating κατά before τὴν διάνοιαν.

αὐτούς . For αὐτούς following ὧν see on II 357 B. The rule against the repetition of the relative in such cases is sometimes dispensed with for the sake of rhetorical emphasis, e.g. in II 374 B and perhaps Theaet. 192 B .

μιμεῖσθαι . In what sense can the guardians be said to ‘imitate’ in such a case, or in those specified in 396 A, B? Not as actors, but as spectators. Acting involves three elements—the character, the actor, and the spectator. In good acting the spectator identifies himself with the actor through sympathy; and as the actor ‘imitates,’ so does he. Such is Plato's theory, though merely glanced at here. Cf. X 605 C ff., Ion 533 D ff., and see the excellent remarks of Nettleship Lectures and Remains II pp. 100—104.

ἢ ἀνδρὶ κτλ. ἀνδρί is of course ‘husband,’ not simply ‘a man’ (D. and V.). Contemporary comedy doubtless furnished abundant illustrations. In πρὸς θεοὺς ἐρίζουσαν κτλ. Plato may be thinking of Aeschylus' Niobe (see on II 380 A). The emphasis on οἰομένην should be noted: cf. I 336 A note

κάμνουσαν ‐‐ ὠδίνουσαν glances at Euripides and his school: cf. Ar. Frogs 1043 , 1044 and 1080, with the Scholiast's remark on 1080 ἔγραψε γὰρ ( ὁ Εὐριπίδης) τὴν Αὔγην ὠδίνουσαν ἐν ἱερῷ. Plato's strictures throughout this passage tell much more heavily against Euripides than against the other two dramatists.

ἢ καὶ ἄλλα . ἄλλα must be coordinated with αἰσχρολογοῦντας, not with νήφοντας, so that Hartman's correction ( καί for ἢ καί), though scarcely necessary, is an improvement, and may be right.

μαινομένοις . As in the Eumenides, Ajax, Hercules Furens.

γνωστέον κτλ. cf. 409 A.

μιμητέον . See on μιμεῖσθαι 395 D.

ἵππους ‐‐ βροντάς . The reference is probably to stage machinery and musical effects etc. in dramatic poetry generally, as well as in the later and degenerate form of the dithyramb (see on 394 C). Cf. (with Nettleship Lect. and Rem. II p. 105) Laws 669 C ff. and Ar. Plut. 290 ff. The βροντεῖον and κεραυνοσκοπεῖον for producing thunder and lightning were familiar enough (Müller Gr. Bühnenalt. p. 157 note 2). It is clear, as Nettleship remarks, that “Plato felt strongly that Greek literature and music were declining” in his days: see Laws 659 A ff., 700 A ff., 797 A ff.

ὁ μὲν ‐‐ ἀνήρ . It seems difficult (as Schneider remarked) either to connect ὁ μέν with μέτριος ἀνήρ, or to understand ὁ μέν as ‘the one’ and suppose that μέτριος ἀνήρ is in apposition to it. If the latter alternative is right, we should expect μέτριος < ὢν > ἀνήρ, or < ὁ > μέτριος ἀνήρ, and in view of other cases in which the article is placed at some distance from its noun (e.g. ὁ δέ γε, οἶμαι, ἦν δ’ ἐγώ, καταληφθεὶς θανάτῳ δίδοται VIII 566 C), I still prefer the former view. Some may be inclined to regard μέτριος ἀνήρ as a gloss. I have sometimes been tempted to make μοι δοκεῖ parenthetical (exactly=‘methinks’), in which case ὁ μέν can easily be connected with μέτριος. The idiom occurs in Phaed. 108 D ὁ βίος μοι δοκεῖ ὁ ἐμός—τῷ μήκει τοῦ λόγου οὐκ ἐξαρκεῖ and Menex. 236 B : cf. also Crito 43 D, 50 B, and I 332 E note This solution would involve the change of ἐθελήσειν to ἐθελήσει—so v—and of αἰσχυνεῖσθαι to αἰσχυνεῖται just below, as well as again in D. Such a corruption, once started, ἔρχεται—as Plato might say — ὡς κύκλος αὐξανομένη; but I do not venture to change the text.

καὶ ἧττον is not superfluous with ἐλάττω. ἐλάττω means ‘in fewer respects,’ and ἧττον ‘to a less degree.’

σπουδῇ . Cf. ὅ τι μὴ παιδιᾶς χάριν in E and σπουδῇ 397 A.

ὅ τι μὴ παιδιᾶς χάριν . Cf. VII 518 B.

οἵᾳ . According to Van Cleef ( de Attract. usu Plat. p. 36), οἷος is not elsewhere attracted in Plato.

τῆς ἁπλῆς . See cr. n. The reading of the MSS τῆς ἄλλης ought strictly speaking to mean ‘the rest of διήγησις,’ i.e. besides μίμησις. A reference to 392 D will shew that the rest of διήγησις includes (1) simple διήγησις, (2) the mixed style. If the text is sound, Plato therefore says that the good man's λέξις will resemble Homer's in partaking of all three varieties. This is a cumbrous and unnecessary elaboration: for if style partakes both in μίμησις and in simple διήγησις, it is already ipso facto ‘mixed.’ To take ἄλλης as ‘besides’ may be admissible, but in any case it is desirable to define the kind of διήγησις meant. I believe that Plato wrote ἁπλῆς. The good man's style will resemble Homer's, which has already been said to partake of μίμησις (393 C) and of ἁπλῆ διήγησις (394 B). The corruption—common in uncial MSS—is illustrated by Bast Comment. Palaeogr. p. 730. Cf. my article in Cl. Rev. X pp. 384 f.

μέρος (as Schneider points out) depends on μετέχουσα: cf. Euthyd. 306 A ὦν ἀμφοτέρων μέρος μετέχουσι.

μιμήσεται . See cr. n. The choice of reading lies between this and Madvig's emendation <μιμήσεται ἢ> διηγήσεται. In favour of μιμήσεται is μᾶλλον, which correlates with ὅσῳ ἂν φαυλότερος ᾖ. The corruption doubtless arose from a misinterpretation of μᾶλλον. Thinking that an ἤ clause was needed to explain it, a scribe added ἢ διηγήσεται in the margin, and διηγήσεται was afterwards taken as a variant and ousted μιμήσεται. These arguments, which are Hartman's, seem to me conclusive in favour of μιμήσεται, which Schneider first restored.

σπουδῇ τε καὶ ἐναντίον πολλῶν : like the professional dramatist or actor.

τροχιλιῶν κτλ. Cf. supra 396 B note

σχήμασιν ‘gestures.’

ἔλεγον . 396 B, C.

πρὸς τὴν αὐτήν : sc. ἁρμονίαν, as Schneider saw. To supply λέξιν with Stallbaum, Hartman, and others is not satisfactory, nor is it easy to understand χορδήν (with Campbell). On the other hand ἁρμονίαν may be readily supplied in view of ἐν μιᾷ ἁρμονίᾳ following. ὁ λόγος qualifies τὴν αὐτήν. The somewhat vague expression πρὸς τὴν αὐτήν, where the musical sense of προς may be illustrated by πρὸς Λιβὺν λακεῖν | αὐλόν ( Eur. Alc. 346 ), is afterwards made more explicit and precise by ἐν μιᾷ ἁρμονίᾳ i.e. ‘in one musical mode’ (see on 398 E), as opposed to πασῶν—ἁρμονιῶν in C. μεταβολή was technically used of passing from one ἁρμονία to another: see Cleonid. Isag. Harm. 13 and Bacchius Isag. 53 ed. von Jan. We shall best apprehend the full meaning of the whole passage if we read it in connexion with 399 A, B. The general sentiment may be illustrated from Arist. Eth. Nic. IV 8. 1125^{a} 12 ff. καὶ κίνησις δὲ βραδεῖα τοῦ μεγαλοψύχου δοκεῖ εἶναι, καὶ φωνὴ βαρεῖα, καὶ λέξις στάσιμος, Pl. Charm. 159 B , Dem. 37. 52 and elsewhere.

διὰ τὸ παντοδαπὰς ‐‐ ἔχειν . As the λέξις itself is full of variety, it requires for its proper or appropriate ( οἰκείως) expression every variety of mode and rhythm or musical time. μορφὰς τῶν μεταβολῶν is surely good enough Greek: I cannot see the point of Richards' μορφὰς ἐκ τῶν μεταβολῶν, still less why Hartman should eject τῶν μεταβολῶν or— as an alternative— μορφάς.

ἐπιτυγχ άνουσιν = ‘hit upon,’ ‘stumble upon,’ as if by accident and ἄνευ νοῦ, not ‘succeed,’ as J. B. Mayor is disposed to construe ( Cl. Rev. X p. 109). The same scholar proposes to change ξυγκεραννύντες into ξυγκεκραμένῳ, but the text is much more idiomatic as it stands.

τὸν ἕτερον : ‘one or other.’ Presently τοῦ ἐπιεικοῦς ‘the good man’ is said for ‘the good man's style of speaking’; see 398 B and cf. 399 B note Before ἄκρατον, many editors add τόν (with Ξ^{2}): but the position of ἄκρατον is normal: cf. τὰ ἐν ὕδασι φαντάσματα θεῖα VII 532 C and note ad loc.

παισί ‐‐ τῷ πλείστῳ ὄχλῳ . The expression recurs in Laws 700 C (quoted by J. and C.).

οὐκοῦν διὰ ταῦτα κτλ. There is probably a satirical reference to Athenian democracy: see Prot. 319 D .

αὐτός ‐‐ ἐπιδείξασθαι : ‘anxious to shew himself off together with his poems.’ ἐπιδείξασθαι is intransitive—i. q. ἐπίδειξιν ποιήσασθαι, cf. Lach. 179 E —with αὐτός, but transitive with ποιήματα. This explanation, which is due to Schneider, gives a much better sense than if we regard αὐτός τε καὶ τὰ ποιήματα as subject to ἀφίκοιτο, or translate ‘himself, and wanting to shew his poems’ (J. and C.). A reference to αὐτός τε καὶ τὸν ἀδελφὸν παρακάλει in IV 427 D is therefore hardly to the point.

προσκυνοῖμεν . The insertion of μέν, recommended by Shilleto (Dem. F. L. § 91) and Richards, is unnecessary: cf. I 340 D note For προσκυνεῖν ‘to kiss the hand’ (adorare), as to the image or shrine of a god, see Cope's Rhetoric of Aristotle Vol. I p. 86.

οὔτ) ἔστιν ‐‐ οὔτε θέμις . It is perhaps better to correct οὐκ into οὔτ)—see cr. n.—than the second οὔτε into οὐδέ (with Bekker and the other editors).

μύρον ‐‐ στέψαντες . The idea suggested by προσκυνοῖμεν and ἱερόν, that the poet is a sort of θεός or θεῖος ἀνήρ, is now elaborated with ironical politeness. The images of the gods were anointed, and crowned with garlands, not only on great occasions (cf. Cic. Verr. IV 77), but also at other times, according to Proclus, who remarks on this passage μύρον αὐτῆς (sc. τῆς ποιητικῆς ) καταχέας, ὡς τῶν ἐν τοῖς ἁγιωτάτοις ἱεροῖς ἀγαλμάτων θέμις, καὶ ὡς ἱερὰν στέψας αὐτήν, ὥσπερ καὶ ἐκεινα στέφειν ἦν νόμος ( in remp. p. 42 ed. Kroll). Schneider aptly compares Paus. X 24. 6 τούτου (a sacred stone) καὶ ἔλαιον ὁσημέραι καταχέουσι καὶ κατὰ ἑορτὴν ἑκάστην ἔρια ἐπιτιθέασι τὰ ἀργά. For other illustrations see Frazer on Paus. l.c., and Munro on Lucr. v 1199. Apropos of the present passage, Dio Chrysostom and other ancient writers cited by Ast refer to the anointing of swallows by Greek women: καὶ κελεύει μάλα εἰρωνικῶς (so Ast: MSS εἰρηνικῶς) στέψαντας αὐτὸν ἐρίῳ καὶ μύρῳ καταχέαντας ἀφιέναι παρ’ ἄλλους· τοῦτο δὲ αἱ γυναῖκες ἐπὶ τῶν χελιδόν ων ποιοῦσι (Dio Chr. Or. 53 p. 276 ed. Reiske). To this custom Ast supposes that Plato is alluding, the poets being as it were faithless and garrulous swallows (cf. χελιδόνων μουσεῖα), as well as to the Pythagorean precept ‘not to admit swallows into the house’ (Plut. Symp. VIII 727 B ff.), on which see Frazer in Cl. Rev. V pp. 1—3. This explanation lends an additional point to ἀποπέμποιμεν: and προσκυνοῖμεν might fairly be interpreted of the joyful salutations with which the Greeks hailed the advent of the swallow in the spring (see e.g. Baumeister Denk. d. Kl. Alterth. p. 1985). G. B. Hussey ( Proceedings of the American Philol. Association Vol. XXII pp. xliii ff.) thinks that Plato has in his mind the wellknown χελιδονισμός of which we read in Athenaeus (VIII 360 B ff.), remarking that in the swallow song ‘the custom seems to have been to carry some sort of symbolic swallow from house to house.’ It is perhaps more probable (as Mr J. G. Frazer suggests to me) that “the ceremony of anointing the swallows and crowning them with wool was performed on the children who went from door to door in spring, singing the swallow song and apparently personating the swallow.” But the tone of the whole passage, with its air of studiously exaggerated politeness and compliment, as well as the particular expressions προσκυνοῖμεν, ἱερόν, and θαυμαστόν, are strongly in favour of Proclus' interpretation, although Plato's thoughts may have dwelt for a moment on the practices connected with the χελιδονισμός when he wrote the words ἀποπέμποιμεν— στέψαντες.

κατ’ ἀρχάς . II 379 A ff.

τὸ περὶ ᾠδῆς κτλ. The discussion has hitherto confined itself chiefly to tragedy and comedy. It remains to discuss lyrical poetry also on its formal side. Now the chief formal characteristic of lyric poetry is its invariable association with music. It is therefore necessary to lay down canons for musical composition. This is the justification for the sections on ‘harmony’ and rhythm, which are wrongly pronounced to be irrelevant by Krohn ( Pl. St. p. 15).

τρόπου . Hartman suggests τρόπον, in view of τὸ περὶ ῥυθμούς 399 E; but cf. 392 C. τρόπος is not here used in its technical sense, for which see Monro l. c. p. 63.

συμφωνήσειν . The metaphor may be suggested by the subject under discussion: cf. Phaed. 92 C .

λόγου ‐‐ ῥυθμοῦ . In the best period of Greek music, lyric poetry was written only for music, and music only for poetry, the separation of the two being condemned as illegitimate: see Monro l.c. pp. 119, 120. The elements of music are ῥυθμός and ἀρμονία. The former ‘reconciles’ ταχύ and βραδύ by arranging a proper sequence of short and long notes and syllables, the latter ὀξύ and βαρύ by a proper arrangement of notes of higher and lower pitch ( Symp. 187 A —C). In the wider sense, therefore, any ὁμολογία of ὀξύ and βαρὐ is a ἁρμονία, but in practice the word was used specifically of certain scales or modes, and it is in this sense (according to Westphal) that Plato uses it here and in 398 E, where see note.

ὡσαύτως : i.e. ἐν τῇ αὐτῇ λέξει as defined in 396 E, 397 D.

καὶ μὴν κτλ. The poet should be his own musician, and write the music to suit the words, not vice versá. This was another characteristic feature of classical Greek music, although a change set in during the fourth century B.C. See Westphal Gr. Rhythmik p. 1 and Laws 669 D, E, 812 D.

ἁρμονίαι (according to the orthodox view) are ‘musical modes’ and not simply ‘keys.’ They differed from each other both in the arrangement of the intervals (like our major and minor modes) and also in pitch. It must have been the former difference which chiefly —though not perhaps exclusively—accounted for the different effects of different modes upon the character and emotions, just as we are ourselves affected in different ways by music written in major and in minor keys. See H. S. Jones in Cl. Rev. VIII p. 449.

μιξολυδιστὶ κτλ. The omission of the article has been questioned, but in merely naming the scales it can be dispensed with: cf. (with Stallbaum) Arist. Pol. Θ 5. 1340^{b} 1 ( τὴν μιξολυδιστὶ καλουμένην). On the ἁρμονίαι recognized by Plato see App. II.

Ἰαστὶ ‐‐ καλοῦνται : ‘there are also varieties of Lydian and Ionian which are called ‘slack’.’ Jowett and Campbell, reading αἵτινες (see cr. n.), remark that the “indefinite relative suits with Plato's affected ignorance”; but the speaker is Glauco, not Socrates, and Glauco is μουσικός. See note on 399 C. Richards condemns αἵτινες χαλαραὶ καλοῦνται as spurious because αἵτινες “cannot be used in this way in good Attic prose of Plato's date.” With the older and better attested reading αὖ τινες, which I have ventured to restore, everything is plain. The words αὖ τινες establish once for all what Westphal (l.c. p. 198) and von Jan (l.c. p. 816) detected even when αἵτινες was read, viz. that Plato is referring not to Ionian and Lydian, but to slack Ionian and slack Lydian, a point which escaped Monro (l.c. p. 7) but not his reviewer ( Cl. Rev. VIII p. 449). See also my article in Cl. Rev. X pp. 378 f. We learn from Aristotle that certain musical critics censured Plato for rejecting τὰς ἀνειμένας ἁρμονίας and for characterising them as μεθυστικαί, βακχευτικὸν γὰρ ἥ γε μέθη ποιεῖ μᾶλλον ( Pol. Θ 7. 1342^{b} 23—27). It was partly perhaps in deference to these criticisms that Plato altered his view of μέθη in Laws 666 Aff.: see also Grote Plato III p. 328 note

δωριστὶ καὶ φρυγιστί . The absence of the Aeolian mode is remarkable, for it must certainly have been known to Plato (see Pratinas quoted in App. II). Westphal agrees with Bellermann in supposing (l.c. p. 195) that αἰολιστί is included under δωριστί. Aristotle also ignores αἰολιστί, unless indeed (as Westphal holds ib. p. 196) it was identical with ὑποδωριστί. In Lach. 188 D φρυγιστί is excluded (perhaps because the speaker is Laches, whose ideal of courage is military rather than pacific), and Dorian, ‘the only national Greek mode,’ alone recognized.

ἐκείνην τὴν ἁρμονίαν : viz. Dorian, not Phrygian, as Ast seems to have thought.

μιμήσαιτο . Cf. Laws 798 D τὰ περὶ τοὺς ῥυθμοὺς καὶ πᾶσαν μουσικήν ἐστι τρόπων μιμήματα βελτιόνων καὶ χειρόνων ἀνθρώπων and 397 B above.

καὶ ἀποτυχόντος . καί connects ὄντος and ἀμυνομένου. ἀποτυχόντος (which is itself logically subordinate to ἀμυνομένου) has three subordinate alternatives ( ἢ—πεσόντος), all of which are summarised in ἐν πᾶσι τούτοις .

ἄλλην : viz. Phrygian. Aristotle blames Plato for retaining the Phrygian mode, while rejecting the αὐλός, with which it was usually associated: ἄμφω γὰρ ὀργιαστικὰ καὶ παθητικά ( Pol. Θ 7. 1342^{b} 3). Plato, however, rejects the flute, not because it is orgiastic, but because it is πολυαρμόνιον (399 D). In Plato's opinion the Phrygian mode expressed sobriety and resignation: Aristotle thought it ecstatic and purgative (l.c. 1341^{a} 23). The difference of view is interesting and important as shewing that the ethical effect of different modes was a disputed point even among the ancients.

ἢ εὐχῇ ‐‐ ἄνθρωπον is subordinate to πείθοντός τε καὶ δεομένου.

ὑπέχοντα . ἐπέχοντα—see cr. n.— cannot, I think, be right. ἐπέχειν τὴν διάνοιαν ( Laws 926 B) certainly does not justify ἐπέχειν ἑαυτόν, and even if it did, ‘submitting to’ and not merely ‘attending to’ is the sense required. With ὑπέχοντα cf. Gorg. 497 B ὑπόσχες Σωκράτει ἐξελέγξαι ὅπως ἄν βούληται, where the reflexive pronoun is omitted, as often with παρέχειν. Here it is better to take ἑαυτόν with ὑπέχοντα than with μεταπείθοντι. By changing the construction and writing accusatives instead of genitives, Plato makes the man himself rather than his φθόγγοι appear the object of imitation (cf. 397 D note). This is natural enough, because the situations described in ἢ τοὐναντίον—ἀγαπῶντα give less scope for φθόγγοι. Stephanus wished to read the genitive throughout ( ὑπέχοντος, πράξαντος etc.: so also v and two Florentine MSS), but there is also inscriptional evidence for a genitive or dative participle followed by an accusative in the course of a long sentence: see Meisterhans^{3} p. 205.

κατὰ νοῦν : ‘to his liking’: cf. εὐτυχούντων below.

ταύτας ‐‐ λεῖπε . The style is intentionally weighty and formal, as befits a solemn pronouncement: cf. X 617 D, E. After ταύτας there is a slight pause: ‘Just these, two modes and none other.’ The insertion of τάς would impair the effect. besides suggesting that Socrates had in view two of the current modes, which, not being himself μουσικός, he professedly had not. It is Glauco's business to fit the cap (398 E, 399 A); Socrates only makes it. The indefinite αἵτινες (before φθόγγους) is therefore strictly appropriate in the mouth of Socrates, although it would not be in Glauco's. ἁρμονίας is rejected by Herwerden in both places (see cr. n.), but it is almost as indispensable here as it is wrong after ἀνδρείων, although Stallbaum rejects the word here and retains it there. The genitives δυστυχούντων etc. must depend on φθόγγους. For βίαιον, ἑκούσιον (‘one involuntary, one voluntary’), Ast suggests βιαίου, ἑκουσίου, Hartman βιαίων ἑκουσίων. A human being cannot however be called βίαιος because he is engaged ἐν βιαίῳ πράξει, although the mode which imitates his accents may be so described with propriety and even elegance: cf. (with Schneider) such expressions as φόνος ξυγγενής for the slaughter of kindred. The words δυστυχούντων—κάλλιστα simply define the meaning of βίαιον and ἑκούσιον (‘whatever musical modes they be that shall best imitate the accents of’ etc.): the relative is postponed in order to keep the essential marks of the ἁρμονίαι together, but the careful reader will note that Plato begins a chiasmus with δυστυχούντων, as if to separate the genitives from what precedes and prepare us to find their construction in the sequel. Had he written εὐτυχούντων, δυστυχούντων, ἀνδρείων, σωφρόνων the double chiasmus would have compelled us to connect the genitives with δύο ἁρμονίας.

οὐκ ἄλλας ‐‐ ἔλεγον . The Dorian to express ἀνδρεία, the Phrygian σωφροσύνη. These are the two contrasting virtues which Plato's μουσική endeavours to combine (410 E).

παναρμονίου . In Plato the noun παναρμόνιον occurs only here and in 404 D ᾠδῇ τῇ ἐν τῷ παναρμονίῳ καὶ ἐν πᾶσι ῥυθμοῖς πεποιημένη. In the latter passage it certainly does not denote a musical instrument of any kind. Here the word is sometimes understood of a particular and definite musical instrument, but a careful study of the context shews that it does not bear this meaning even here. Plato has decided to admit only two modes, the Dorian and the Phrygian. ‘Consequently,’ he continues, ‘we shall have no need in our songs and melodies of πολυχορδία or παναρμόνιον, and therefore ( ἄρα) we shall dispense with τρίγωνοι, πηκτίδες etc., with all instruments, in short, which are πολύχορδα and πολυαρμόνια.’ The prohibition of certain musical instruments is an inference from the general principle that πολυχορδία and παναρμόνιον are unnecessary, so that παναρμόνιον cannot itself be a particular musical instrument. Probably, as Mr Archer-Hind has suggested to me, the παναρμόνιον was “not a mode or modes, but a style of composition, in which the ‘Tondichter’ passed freely from δωριστί to φρυγιστί and λυδιστί and as many others as he chose. The name may even have been given to well-known compositions in this style—cf. νόμος πολυκέφαλος—the fantasia with many subjects. The effect, I should think, may have been analogous to a series of bold and sudden modulations in modern music.” See also on αὐτὰ τὰ παναρμονία in 399 D.

τριγώνων ‐‐ πηκτίδων . These were foreign instruments of high pitch, and many strings. The τρίγωνον in particular was associated with loose and voluptuous melodies. For an exhaustive account of both see Susemihl and Hicks' Politics of Arist. vol. 1 pp. 632—636 or von Jan's de fidibus Graecorum pp. 29 ff., 33 ff.

αὐλητἀς . The αὐλός resembled the clarinet. It had a “mouthpiece ( ζεῦγος) in which a vibrating reed ( γλῶττα) was fitted,” and was sometimes played in pairs. See Dict. Ant. s.v. tibia. Plato banishes the ‘flute’ and retains the Dorian mode, although Dorian melodies were often played on it, as Milton well knew: see the noble description of the “Dorian mood of flutes and soft recorders” in Par. Lost 1 550 ff. In Boeotia, where the αὐλός was highly esteemed, it was supposed rather to calm than to excite the feelings. See Rhys Roberts The Ancient Boeotians pp. 33— 35.

ἢ οὐ τοῦτο πολυχορδότατον; τοῦτο is that with which αὐλοποιοί and αὐληταί are concerned, viz. the ‘flute’: cf. II 377 C note οὗτος instead of τοῦτο would have been a trifle harsh. πολυχορδότατον has been repeatedly called in question, and there is the usual crop of emendations, intended to obliterate the metaphor. Schneider has however shewn that the MS reading is sound, by citing Pollux IV 67 Πλάτων δὲ καὶ πολύχορδον εἴρηκε τὸν αὐλόν, and Simon. Fr. 46 ὁ καλλιβόας πολύχορδος αὐλός, and comparing expressions like αὐλὸν κρέκειν, ἁρμόζειν, κρούειν. Many other illustrations are given by Smyth, Greek Melic Poets p. 326. Here the metaphor is intended to arrest attention by its boldness and prepare us for the theory of the origin of παναρμόνια in the next clause; but πολυχορδότατον in itself, like πάμφωνος in Pindar ( Pyth. 12. 19 al.), refers only to the number of different notes which the flute, thanks to various contrivances, such as plugs, wax, etc., was capable of producing. See Abdy Williams in Proceedings of the Musical Association 1897—8 p. 135. Plato objects to the multiplicity of strings and notes as admitting and even inviting change and fusion of modes. We are told by Paus. IX 12. 5 (cited by Monro l.c. p. 38: cf. Ath. XIV 631 E) that it was one Pronomus of Thebes who πρῶτος ἐπενόησεν αὐλοὺς ἐς ἅπαν ἁρμονίας εἶδος ἔχοντας ἐπιτηδείως. Down to his day there were three forms of ‘flutes,’ intended for the Dorian, Phrygian and Lydian modes respectively. On the means by which this change was effected see Dict. Ant. s.v. tibia.

αὐτὰ τὰ παναρμόνια : sc. ὄργανα, such as πηκτίδες and τρίγωνοι. Plato means those instruments on which panharmonic melodies could be played (cf. Proclus in remp. p. 63 ed. Kroll): but we must beware of translating (with D. and V.) ‘the panharmonium itself,’ for no single specific instrument is here intended, as some later lexicographers appear to have supposed. The gloss in Hesychius παναρμόνιον· εἶδος ὀργάνου, ἐξ ὅλου τεταγμένον is not quite clear, and may conceivably refer to a whole class of instruments, but Photius apparently thought that there was a special instrument called παναρμόνιον. His note (p. 388, 26 ed. Porson) is as follows: παναρμόνιον· ὄργανον μουσικόν· Ἄλεξις, ἐν ᾧ τὸ παναρμόνιον τὸ καινὸν ἔντεινον τεχνῶν ( Τέχνων Meineke). Photius may of course be right in his interpretation of Alexis' line: but παναρμόνιον in Plato never, I believe, refers to one particular instrument: and even Alexis may mean no more than ‘perform the new panharmonic melody,’ ἐντείνω being used as in τὸ κάλλιστον ἐντείνας μέλος, Dionys. Hal. de admir. vi dicendi in Dem. c. 48.

λύρα ‐‐ κιθάρα . The λύρα was the stringed instrument in common use; the κιθάρα was employed chiefly by professional musicians or κιθαρῳδοί. See Monro in Dict. Ant. s.v. Lyra, where illustrations of the two instruments are given, and von Jan de fid. Gr. pp. 5—26. By admitting the professional κιθάρα, Plato perhaps lends his sanction to musical festivals or contests in the approved modes.

καὶ κατὰ κτλ. After χρήσιμα supply ἐστίν. This is better than to eject καί (with Ast and—according to Bekker— Vat. Θ). Demetrius ( περὶ ἑρμ. § 185, cited by Schneider) finds in the words καὶ αὖ κατ’ ἀγροὺς τοῖς ποιμέσι (sic, not νομεῦσι) σύριγξ ἄν τις εἴη an imitation of the sound of the σύριγξ. “Ceterum Demetrii rationem me non perspicere fateor,” says Schneider. Demetrius' remark is, I believe, correct, and has reference to the sigmatismus in the words of Plato: cf. Laws 700 C τὸ δὲ κῦρος τούτων—οὐ σύριγξ (used for συριγμός) ἦν οὐδέ τινες ἄμουσοι βοαὶ πλήθους, καθάπερ τὰ νῦν. The σύριγξ was either μονοκάλαμος, resembling our flute, or πολυκάλαμος (like Pan's pipe): see Dict. Ant. s.v. The indefinite τις shews that Plato did not wish to specify which variety he intended.

οὐδέν γε ‐‐ ὀργάνων . Plato puts himself in the position of the Muses, who preferred Apollo's performance on the κιθάρα to that of Marsyas on the flute (Apollod. I 4. 2). This is the force of οὐδέν γε καινὸν ποιοῦμεν. The words τὰ τοῦ Ἀπόλλωνος ὄργανα must not be pressed; for although Apollo invented the cithara, the lyre was ascribed to Hermes (Paus. V 14. 8: cf. the Homeric Hymn to Hermes), and the syrinx to Pan. The discovery of the flute was also ascribed to Athena, especially by the Boeotians. A third account represents Marsyas as picking up the instrument after Athena had discovered and discarded it. This legend may be an attempt to reconcile the two conflicting stories, and probably dates from the decline of the flute as an instrument of education in Athens during the fourth century (Arist. Pol. Θ 6. 1341^{a} 32 ff. Cf. Preller Gr. Myth. p. 223). In making Marsyas its discoverer, Plato declares the flute a foreign instrument, and appropriately excludes it from his ‘Greek city’ (V 470 E).

νὴ τὸν κύνα . This peculiarly Socratic oath occurs only once again in the Republic (IX 592 A). In both passages it marks the highest degree of emphasis. On the oath itself see my note on Ap. 21 E and Blaydes on Ar. Wasps 83 .

ἄρτι : II 372 E note

βάσεις . The word βάσις in the technical writers on Rhythm generally means a dipody or combination of two feet under one main ictus: cf. Schol. in Heph. I 3. 1 p. 124 ed. Westphal βάσις δέ ἐστι τὸ ἐκ δύο ποδῶν συνεστηκός, τοῦ μὲν ἄρσει, τοῦ δὲ θέσει παραλαμβανομένου. Such a technical use of the word would be out of place here, especially in the mouth of Socrates; and the word is employed throughout as equivalent simply to ‘step’ or ‘foot.’ Even technical writers sometimes so use it: cf. the Scholiast already cited δέξεται δὲ (sc. the Iambic metre) ἐν μὲν τῇ πρώτῃ βάσει ἴαμβον καὶ σπονδεῖον II 5. p. 151 and Gleditsch l.c. p. 702.

κοσμίου τε καὶ ἀνδρείου recalls 399 C σωφρόνων ἀνδρείων, and would seem to point to the necessity of two kinds of rhythm, one to go with the Phrygian mode and express sobriety and self-control, the other to join the Dorian mode in expressing courage. On the ethical qualities of Greek rhythm in general, consult Westphal Gr. Rhythmik pp. 226—239 and Arist. Rhet. III 8, with Cope's notes.

μὴ λόγον ‐‐ μέλει . See 398 D note

τρἴ ἄττα εἴδη . Arist. Quint. I 34 ed. Meibom γένη τοίνυν ἐστὶ ῥυθμικὰ τρία· τὸ ἴσον (2/2), τὸ ήμιόλιον (3/2), τὸ διπλάσιον (2/1). To the first belong dactyls, spondees, anapaests: the second includes paeons, cretics, and bacchei: under the third fall trcchees, iambics, ionics. See Gleditsch l.c. p. 694.

ὥσπερ ‐‐ ἁρμονίαι . What are the τέτταρα εἴδη? The following answers (among others) have been given: 1. the intervals of the fourth, fifth, octave, and double octave (Ast): 2. the four notes of the tetrachord, which was probably the historical and at all events the ‘theoretical unit of the scale’ (Stallbaum, Jowett and Campbell): 3. “the four ratios which give the primary musical intervals—viz. the ratios 2:1, 3:2, 4:3 and 9:8, which give the octave, fifth, fourth, and tone” (Monro l.c. p. 106 note; cf. also Dict. Ant. II p. 193): 4. the four ἁρμονίαι Φρυγιστί, Λυδιστί, Δωριστί, Λοκριστί (Westphal Rhythmik p. 238). Ast's view cannot be right, unless we suppose that ἁρμονίαι here includes scales of double compass, which is most unlikely. Westphal's explanation is improbable, for Plato has said nothing of Λοκριστί, and (though perhaps no great stress should be laid on this) it is awkward to derive the ἁρμονίαι ( ὅθεν αἱ πᾶσαι ἁρμονίαι) from themselves. If the principle of Westphal's interpretation is right, I should be inclined to substitute Ἰαστί for Λοκριστί, having regard to 398 E, where see note Cf. Cl. Rev. X p. 379. (I have since found that Prantl also took this view: see note 116 in his translation.) I do not think that Stallbaum has hit the truth, for Plato's language is not suggestive of any allusion to the origin of the octave from the combination of two tetrachords, and a single tetrachord cannot produce a ἁρμονία ( ὅθεν αἱ πᾶσαι ἁρμονίαι). Possibly the τέτταρα εἴδη ἐν τοῖς φθόγγοις denote simply the keynote, its octave, and the intervals of a tone and a semitone: for these are as it were the threads out of which all modes ‘are woven’ ( πλέκονται should be repeated with ἁρμονίαι), the difference between the modes depending on the difference in position of the tones and semitones. But Euclid lays the greatest stress upon the ratios 3:2 and 4:3 as the component elements of the octave: see for example Sect. Can. 6 τὸ διπλάσιον διάστημα ἐκ δύο τῶν μεγίστων ἐπιμορίων συνέστηκεν, ἔκ τε τοῦ ἡμιολίου καὶ ἐκ τοῦ ἐπιτρίτου and cf. ib. 8, 12, and for this reason I now believe that Monro's view has most in its favour.

ποῖα δ’ ὁποίου κτλ. On ὁποίου see I 348 B note, and for the error in Paris A Introd. § 5.

Δάμωνος . μετὰ Δάμωνος is almost a formula with Plato: cf. infra C, 424 C, and Lach. 200 B . Suseminl (on Arist. Pol. Θ 5. 1340^{b} 5) thinks that Plato is alluding to a special work by Damon on the ἦθος and πάθος of modes and rhythms. The word ἀκηκοέναι and the general tone of the passage seem rather to refer to an oral demonstration.

τίνας ‐‐ ῥυθμούς . In general, πόδες ἀπὸ ἄρσεως, or feet in which the θέσις (i.e. the syllable bearing the ictus) followed the ἄρσις, were believed to express more energy and life, than πόδες ἀπὸ θέσεως. See Gleditsch p. 694, and for details as to the ἦθος of the different rhythms ib. pp. 713, 721, 725, 730, 739, 744, 766.

οἶμαι δέ με κτλ. Schneider's δέ γε (found in some inferior MSS) is not appropriate here. The superfluous pronoun after οἶμαι is a well-established colloquialism: cf. Charm. 173 A , Symp. 175 E . οἶμαι, ἀκηκοέναι, and οὐ σαφῶς ὀνομάζοντος are just the words one might employ in giving one's recollections of an abstruse and half-understood lecture, and this is just what Plato is either doing or, more probably, affecting to do. A few technical terms and a vague idea ( οὐκ οἶδ’ ὅπως) of some of the processes are all that he remembers.

ἐνόπλιον ‐‐ ήρῷόν γε . ἐνόπλιος ξύνθετος, δάκτυλος, ἡρῷος are expressions from the lecture: in English they would be in inverted commas. The ἐνόπλιος is not ˘¯¯˘˘˘˘ (Proclus in remp. p. 61, if, as appears probable, by παριαμβίς he means the παρίαμβος or pyrrich), nor the cretic (J. and C.), nor, strictly speaking, the anapaestic foot (Hartman), but ᷋<*>˘˘<*>˘˘<*>, a common processional ( προσοδιακός) or marching rhythm, consisting of an ἰωνικὸς ἀπὸ μείζονος and a choriambus (Hephaestion c. 15), or (as the Scholiast on Ar. Clouds 651 measures it) a spondee, pyrrich, trochee and iambus. For examples we may cite Sappho's αὔτα δὲ σὺ Καλλιόπη ( Fr. 82) and Tyrtaeus' ἄγετ’ ὦ Σπάρτας ἔνοπλοι ( Fr. 16). See Gleditsch l.c. pp. 717, 722, and Bacchius Isag. 101 ed. von Jan, whose example is ὁ τὸν πίτυος στέφανον. ξύνθετος probably refers to the composite character of the rhythm. as described, for example, by the Scholiast on the Clouds. The later technical expression for this peculiarity was ἐπισύνθετος (Gleditsch p. 746). δάκτυλον must be understood as a foot, not as a rhythm, although the ἐνόπλιος ξύνθετος certainly, and probably also the ἡρῷος, are rhythms. There is no difficulty about this, provided we remember that Plato is quoting (or pretending to quote) isolated technical expressions from Damon's lecture. The ingenious, though hazardous, proposal of Blaydes, to read καὶ <κατὰ> δάκτυλον (cf. Clouds 651), would confine the instances to rhythms until we reach ἴαμβον. Dr Jackson suggests δακτυλικόν in place of δάκτυλον. It is tempting (with J. and C.) to take ἡρῷος as ‘spondee,’ but there seems to be no authority for such a use of the word. The ἡρῷος πούς is apparently a dactyl; although the ἡρῷος ῥυθμός admits of the spondee. Unless, therefore, we take ἡρῷος as a rhythm, the spondee seems to be altogether excluded. It is unnecessary to do more than allude to Hartman's excision of ξύνθετον καὶ δάκτυλον.

ἴσον ‐‐ τιθέντος . ἄνω and κάτω refer of course to the position of the arsis and thesis (cf. ὁ ἄνω, ὁ κάτω χρόνος said of the notes at which the foot or bâton is raised and brought down respectively), but Westphal's remark that Plato uses τ ὸ ἄνω and τὸ κάτω is misleading ( Rhythmik p. 104). The words must be taken as adverbs, and can only be explained by supposing that when Damon was demonstrating the equality of arsis and thesis he ‘placed’— τιθέντος is not ‘assuming’ as διακοσμοῦντος shews—the former in a diagram above the latter, in some such way as <*><*>. The position of the ictus — ἄνω καὶ κάτω, not κάτω καὶ ἄνω—shews that Plato is speaking of the dactyl and spondee which replace the anapaest in the anapaestic rhythm: for in the dactylic rhythm proper the ictus falls on the first syllable (see Gleditsch p. 693). Now the ἐνόπλιος is also anapaestic, so that it looks as if Damon had taken as the subject of his demonstration some passage like Persae 9, 10 ἤδη | κακόμαντις ἄγαν ὀρσολοπεῖται, and analysed it into an ἐνόπλιος ξύνθἐτος, a dactyl, and a spondee (included, as stated above, under the ἡρῷος ῥυθμός).

εἰς βραχύ ‐‐ γιγνόμενον . These words can only mean ‘passing into a short and a long,’ “mit kurzem und langen Ausgang” (Schneider), “so dass er sowohl in eine kurze als auch in eine lange Silbe auslief” (Prantl): see on II 380 D. The slight inaccuracy involved in saying γιγνόμενον, where τελευτῶντα (cf. VI 511 C) would have been more precise, is perhaps in keeping with the airy nonchalance of Socrates' description. The construction is missed by Westphal ( Rhythmik p. 237) and the English translators and editors. γιγνόμενον agrees with ἡρῷον: the ἡρῷος ῥυθμὸς γίγνεται εἰς βραχύ when it uses a dactyl, εἰς μακρόν when it uses a spondee (or anapaest), the two alternatives being denoted by τε καί. I have sometimes felt disposed to take the words as referring to the iambus, and place them just before ὡς ἐγὼ οἶμαι, translating ‘and when it’ (the rhythm) ‘changed to a short and a long, I think he called it an iambus’: but although this interpretation gives a somewhat better sense to γιγνόμενον, I am not convinced that the MSS are wrong. Hartman also suggests the transposition of καί, but he might have spared his “minime audax coniectura” ἐκ βραχέων τε καὶ μακρῶν γιγνόμενον. See also the next note.

μήκη ‐‐ προσῆπτε . Hartman takes these words as explaining the trochee only, laying emphasis on the precedence given to μήκη; but the use of the plural shews that the iambus is also included. The meaning is simply ‘and he assigned them longs and shorts,’ i.e. to each one long, and one short. This clause is in favour of keeping εἰς βραχύ—γιγνόμενον in its place; if we transpose (as suggested in the last note), the short and long of the iambus will be alluded to twice.

ἀγωγάς . ἀγωγή is tempo (Gleditsch p. 688). The unit of measurement was the χρόνος πρῶτος or ¯: and hence the dactyl. for example, has usually a τετράσημος ἀγωγή, the iambus a τρίσημος, and so on. See Excerpta Neapol. in von Jan's Mus. Script. Gr. § 14. The duration of the χρόνος πρῶτος was of course relative, and not absolute, so that the time occupied in singing or declaiming a foot often varied, and we are told that ἔστιν ὅτε καὶ ἐν δισήμῳ (sc. ἀγωγῇ ) γίνεται δακτυλικὸς πούς ( Exc. Neap. l. c.). But it is clear that in general the ἀγωγαί of the different kinds of feet were different from one another. Hartman ejects τοῦ ποδός, “cum apud Platonem πούς et ῥυθμός non discrepent.” The distinction between πούς and ῥυθμός is not always preserved by writers on metre (e.g. Bacchius Isag. 100 ff. ed. von Jan), but Plato seems to make the πούς differ from the ῥυθμός as the unit from the whole.

ἤτοι . See on I 344 E.

εὐσχημοσύνης : grace or beauty of form in the widest sense. The word is introduced in view of the application of these principles to objects appealing to the eye: see 401 A.

ἀνάρμοστον . The article (which Baiter and Hartman require) is unnecessary. See on I 334 E.

ἀκολουθητέον (i.q. δεῖ ἀκολουθεῖν) has ταῦτα for its subject, as Stallbaum points out: cf. Laws 803 D τί παίζοντα ἐστὶ διαβιωτέον; and infra V 467 C.

τῷ τῆς ψυχῆς ἤθει ἕπεται . Le style c'est l'homme. Conversely, thought is the dialogue of the soul with itself: see Theaet. 189 E (with Wohlrab's note) and Soph. 263 E . Cf. also IV 437 C note and Homer's διελέξατο θυμός.

ὡς εὐήθειαν is expunged by Herwerden; Baiter would omit ὡς. If ὡς belonged to εὐήθειαν (as these critics apparently supposed), it would deserve expulsion; but it goes with οὖσαν understood. The antithesis is between ἄνοιαν and εὐήθειαν: and if the sentence is read so as to lay stress on these two words, it will be seen how easily οὖσαν can be repeated after εὐήθειαν. The sense is: not the εὐήθεια which is really ἄνοια, but which we euphemistically designate as if it were εὐ‐ήθεια (i.e., as before, in the good sense of the word), but εὐήθεια in its true and etymological sense ( ὠς ἀληθῶς)—the εὖ τὸ ἦθος κατεσκευασμένην διάνοιαν. This explanation seems to me better than to regard ὡς εὐήθειαν as attracted for ὡς εὐήθεια (sc. ἐστίν), a construction for which we may compare Prot. 357 D : see my note ad loc. For ὡς ἀληθῶς cf. I 343 C note

τὸ αὑτῶν πράττειν . The principle of ἁπλότης, which is the corner-stone of Plato's city, presents itself in the education of the young, as the pursuit of εὐήθεια.

ἔστιν δέ γέ που κτλ. This lofty conception of ἁρμονία and ῥυθμός—for αὐτῶν shews that these are included no less than εὐσχημοσύνη—stretching throughout the whole domain of art and nature, may have been suggested by Pythagorean teaching: but the view of education as the pursuit and assimilation of all this beauty is due to Plato himself. Cf. 403 C note

ποικιλία . II 378 C note

τὴν τοῦ ἀγαθοῦ κτλ. This famous section describes in glowing language, like that of the Symposium, Plato's ideal of art. He does not desire to banish art, as is sometimes asserted, but rather idealises it by effecting—as he believed—its reconciliation with beauty and truth. Art aspired to be καλόν in his day: Plato wished it to be so in the fullest sense of the word: and his idea of beauty is sufficiently comprehensive to include moral and spiritual beauty as well as physical. Plato was doubtless unfair in the application of his principle to some of the Greek artists and poets, but in itself his ideal—the love of spiritual beauty —is one to which the best and most enduring art—which alone can find a place in an ideal city—consciously or unconsciously ever seeks to conform. See Nettleship Lect. and Rem. II pp. 112— 116.

τοῖς ποιήμασιν κτλ. Cf. Laws 656 D, E. Nettleship ( Hell. pp. 117 f.) remarks on the fact that “Plato in his criticism of Greek art has almost ignored the painters and sculptors, and confined his assaults to the musicians and still more to the poets.” This is true, although the present passage shews that his canons were intended to regulate painting, sculpture, architecture, and the minor arts as well as music and poetry. Among other reasons, Nettleship plausibly suggests that Plato “did not see in the sculptors and architects of his time the signs of degeneracy which drew his attention to the poets and musicians.” Cf. 401 C.

ὁπόθεν ἂν κτλ. No Greek could read these words without thinking of Olympia; no Athenian without recalling the glories of the Acropolis. It was probably in the spirit of this ideal that Epaminondas—himself a man of Platonic sympathies, if not a Platonist— hinted to his countrymen that their city could not be truly great until the Propylaea crowned their citadel (Aesch. περὶ παραπρεσβείας 105. See also Nettleship Hell. pp. 115—123). Partly on grounds of style, and partly for grammatical reasons, I believe that Plato wrote τις and not τι (see cr. n.). ‘Whenever anything strikes on their eyes or ears from fair works of art’ sounds material and gross in a passage so full of poetic feeling; and in the second place ἄγουσα agrees with αὔρα, whereas it should be ἄγον and agree with τι if τι is right. Translate ‘Whensoever from beautiful works of art there smites upon their eyes or ears as it were a salubrious breath from healthful regions.’ In the same way a sort of ἵμερος flows into the soul from beauty, awakening love and admiration ( Phaedr. 251 C ). The melodious current of Plato's rhythmic utterance flows onward like the steady though gentle breeze which it describes. With αὔρα—ὑγίειαν cf. Arist. Probl. I 52. 865^{b} 19 πόλις ὑγιεινὴ καὶ τόπος εὔπνους ( διὸ καὶ ἡ θάλασσα ὑγιεινή). For the syntax of τις—ὥσπερ αὔρα· φερουσα cf. τὰς τῆς γενέσεως ξυγγενεῖς ὥσπερ μολυβδίδας VII 519 B, where a similar corruption occurs in some of the MSS: see note ad loc. Paris A has τί for τίς again in II 360 E.

ἐν μουσικῇ τροφή . The insertion of ἡ before ἐν (suggested by Rückert) is needless: cf. 404 B.

φέροντα : not ‘imparting’ (Jowett), but ‘bearing,’ ‘carrying,’ like φέρουσα in the simile: cf. Symp. 188 A ἥκει φέροντα εὐετηρίαν.

καὶ μὴ καλῶς . Herwerden's conjecture ἤ for καί misses the precise force of παραλειπομένων ‘falling short’: cf. Critias 107 D ὀξέως αἰσθανόμενοι τὸ παραλειπόμενον. The word is explained in καὶ μὴ—φύντων, where the contrast is between imperfections of art and imperfections of nature.

ἐκεῖ : i.e. ἐν μουσικῇ.

ὀρθῶς δὴ κτλ. I formerly (with Baiter and others) adopted Vermehren's proposal ( Pl. Stud. p. 94) to read ὀρθῶς δὴ <χαίρων και> δυσχεραίνων τὰ μὲν καλὰ· ἐπαινοῖ καὶ [ χαίρων καὶ ] καταδεχόμενος κτλ. The correction is certainly an attractive one, in view especially of Laws 653 B, C, where education is defined as μισεῖν μὲν ἃ χρὴ μισεῖν—στέργειν δὲ ἃ χρὴ στέργειν, and 654 1) τὰ μὲν ἀσπαζόμενος ὅσα καλά, τὰ δὲ δυσχεραίνων ὀπόσα μὴ καλά, and Arist. Eth. Nic. II 2. 1104^{b} 11 ff. But the MS reading, though less pointed and pregnant, is in itself satisfactory enough, if δυσχεραίνων be understood with reference to what precedes ( τῶν παραλειπομένων), and we are therefore hardly justified in altering the text. (The omission of χαίρων καί in q should not be used as evidence of dislocation.) Hartman (after Stallbaum) excises καί between χαίρων and καταδεχόμενος, but this too is unnecessary. We may translate (with Jowett) ‘and rejoicing in them’ (as opposed to δυσχεραίνων just before) ‘and receiving them into his soul.’ The preposition κατα‐ in καταδεχόμενος suggests that beauty is an exile coming home again: the return of exiled truth and beauty is indeed with Plato the aim of education and of life. Cf. Phaedr. 250 A —252 A.

τρέφοιτο . For the metaphor cf. Phaedr. 248 B ff.

γραμμάτων . See on II 368 D. The reference in εἴχομεν, however, is not to that passage, but to the actual experience of the speakers.

ἐν ἅπασιν οἷς ἔστιν : i.q. ἐν ἅπασιν ἐν οἷς ἔστι, by a common idiom: see on II 373 E and cf. VII 520 D, IX 590 C.

ὡς οὐ δέοι depends on the idea of thinking involved in ἀτιμάζομεν. Richards suggested δέον, “sine causa,” as Hartman observes.

εἰκόνας γραμμάτων . The reference to letters throughout this part of the Republic is only by way of illustration, and we must beware of reading more into Plato's words than they are capable of meaning in the context where they occur. No doubt it is true, as Dr Jackson remarks, that “this passage makes us acquainted with the relation of copy and model which is to become important later,” but Bosanquet goes too far when he asserts that “the expression ‘images of letters’ points forward to the classification of grades of knowledge, at the end of Book VI, the allegory of the cave at the beginning of Book VII, and the argument of Book X.”

αὐτά is emphatic: ‘the letters themselves’ as opposed to their εἰκόνες. There is of course no allusion to ‘Ideas’ of letters.

τὰ τῆς σωφροσύνης εἴδη κτλ. Are the εἴδη Plato's Ideas? So Zeller (II^{4} 1 p. 560 note), and many other critics, understand the word; nor can it be denied that the language of Plato, if interpreted in the light of Book VII, can bear this meaning. Nevertheless we are bound in the first instance to interpret this passage by itself, and not by Book VII, the more so as the doctrine of transcendent or separate ( χωρισταί) Ideas appears nowhere else in I—IV, and seems to be expressly reserved by Plato for his philosophical, as distinct from his musical education (see IV 435 D and VI 504 B note). What is meant by the words εἰκόνας αὐτῶν? The context shews conclusively that εἰκόνες refers to copies (sc. of the virtues σωφροσύνη etc.) represented in poetry and the fine arts (so also Krohn Pl. Frage p. 47). On any other interpretation the introduction of these εἰκόνες is irrelevant in a discussion on the rules which imitative art must obey. This being so, if εἴδη means the Ideas, Poetry will be a direct imitation of the Ideas, which is inconsistent with X 595 C —598 D. Or does Plato mean to suggest that Poetry and Art in his ideal city are really to imitate the Ideas directly? This is a bold and attractive solution, and there are several hints elsewhere to the same or nearly the same effect, but Plato expressly speaks of the εἴδη here only as immanent, and not transcendent ( ἐνόντα ἐν οἷς ἔνεστιν), and we must therefore suppose that the artist copies from the life (cf. ἐν τῇ ψυχῇ καλὰ ἤθη ἐνόντα D). The word εἴδη is repeatedly used by Plato without reference to transcendent Ideas, as has been amply proved by Krohn ( Pl. St. pp. 65, 66), Pfleiderer ( Zur Lösung etc. p. 17), and Campbell (II pp. 296 ff.). Here it does not mean ‘varieties’ (as if there were more than one variety of σωφροσύνη), but simply ‘forms’ or ‘kinds,’ in the sense in which the immanent reality which every general notion attempts to express is a ‘form’ or ‘kind’—a genus or species—of the totality of things. Cf. IV 435 B note The genitives are genitives of definition. The use of εἴδη in the sense of “immanente Seinsformen” (Krohn) is interesting as a harbinger of the Ideal theory of VI and VII—a sort of half-way house between the Socratic λόγοι and Plato's ideas. It recurs in IV 434 D, 435 B, 437 D. See further Krohn Pl. Frage pp. 54—58, and cf. VI 504 D note But although the separatists have (as I think) made out their claim that transcendent Ideas do not appear in Books I—IV, I agree with Hirmer ( Entst. u. Komp. d. Pl. Pol. p. 645) in thinking their deductions from this fact unwarrantable.

μεγαλοπρεπείας . μεγαλοπρέπεια in Plato is ‘highmindedness,’ not, as in Aristotle, ‘magnificence’: cf. VI 486 A note In like manner Plato's ἐλευθεριότης denotes the virtue proper to an ἐλεύθερος, and is not restricted to liberality in spending money. Contrast Arist. Eth. Nic. IV cc. 2—6.

τῶν δὴ ‐‐ ἀσπάζεσθαι . Cf. Symp. 209 B and 210 B, C. The whole of Diotima's wonderful speech (210 D— 212 A) should be compared with the closing sections of this chapter. In point of language the words κάλλιστον θέαμα— ἐρασμιώτατον closely resemble Tim. 87 D .

ἀξύμφωνος : i.e. (as Glauco's answer shews) strictly speaking one whose soul and body do not harmonise in point of beauty, but the word also suggests “the man who has no music in his soul.” Cf. Symp. 206 C τὰ δὲ ( κύησις καὶ γέννησις) ἐν τῷ ἀναρμόστῳ ἀδύνατον γενέσθαι. With the sentiment in general cf. Tim. 87 D ff.

μανθάνω ‐‐ ὅτι : ‘I understand: (you say so) because’ etc.: see I 322 A note

οὐ προσοιστέον ἄρα . This somewhat extreme example of a common liberty in concord serves to increase the rhetorical emphasis by the energetic repetition of Glauco's ipsissima verba. The emphasis becomes still greater in Glauco's reply οὐ μέντοι, μὰ Δία, προσοιστεόν. The particle μέντοι is especially used in replies when the words of a previous speaker are repeated (Hoefer de part. Plat. p. 32). q and Flor. U have προσοιστέα.

φιλεῖν is ‘kiss’ (as Schneider rightly translates the word): cf. V 468 B and Arist. Pol. B 4. 1262^{a} 32 ff., where χρήσεις (as Hicks observes) means ‘endearments.’

ἅπτεσθαι κτλ. We think of Socrates and the ‘disciple whom he loved’ in the Phaedo: εἰώθει γάρ, ὁπότε τύχοι, παίζειν μου εἰς τὰς τρίχας (89 B).

ὥσπερ ὑέος . Herwerden's conjecture ὡς πατὴρ ὑέος (or ὥσπερ πατὴρ ὑέος) deserves the praise of ingenuity, but Plato's text is better and more expressive, because it represents the object of affection almost as the lover's very son. It should be noted that in Plato's ἔρως it is the elder who loves, and the younger who is loved; and that the aim and purpose of Platonic love is τόκος ἐν καλῷ ( Symp. 206 B )—the bringing to birth of noble thoughts and aspirations from the beautiful soul of youth. Socrates was the embodiment of Plato's ideal in this respect ( Symp. 216 D ff.). Some true and excellent observations on the subject will be found in Dugas L'Amitié Antique pp. 50—53 al.

τῶν καλῶν χάριν . Plato is resolved that Love, as well as Art, shall serve Virtue and not Vice.

τὰ δ) ἄλλα ‐‐ ξυγγίγνεσθαι . σπουδάζειν πρός τινα occurs with the same sense in Gorg. 510 C . Madvig's περὶ ὧν for πρὸς ὅν would give quite a wrong meaning. σπουδάζει has been suggested for σπουδάζοι (Ast, Richards, Hartman), but the optative puts the case more generally: any one in whom one may be interested. Cf. Soph. Ant. 666 ἀλλ’ ὃν πόλις στήσειε, τοῦδε χρὴ κλύειν, with Jebb's note. The previous sentence has told us what the actual relations of the pair of friends must be; and Plato now forbids all conduct likely in any way to occasion scandal or misapprehension: hence δόξει (‘be supposed to’). Such conduct is in bad taste ( ψόγον ἀμουσίας), rather than positively αἰσχρόν or immoral, like actual vice. μαργότερα τούτῳ (Herwerden) instead of μακρότερα τούτων is a singularly gross conjecture.

ὑφέξοντα . “Si ὑφέξοντα non sanum, corrige ὑφέξειν” (Hartman). This catches the point, but, as Hartman admits, the text can be defended as it stands. The participle agrees with the subject of ὁμιλεῖν, εἰ δὲ μή being all but adverbial, and therefore not followed by a main clause. Cf. Prot. 311 D .

δεῖ δέ που κτλ. The love of Beauty is φιλοσοφία ( Symp. 204 B ); so that the famous saying of the Phaedo (61 A) φιλοσοφία μεγίστη μουσική resembles this. I agree with Krohn ( Pl. St. p. 71) in holding that τοῦ καλοῦ is still beauty as it is revealed in Nature and in Art (see on 402 C), the πολὺ πέλαγος τοῦ καλοῦ of Symp. 210 D , and not yet the transcendent Idea of the Beautiful, the contemplation of which demands a still higher flight (ib. 210 D—212 A). But Plato leaves his μουσικός already knocking at the gates ‘of the blest promised Land.’

γυμναστικῇ κτλ. Plato's statements on γυμναστική have been carefully collected and expounded by Kanter Platos Anschauungen über Gymnastik, Graudenz 1886. Admirable remarks on the whole subject will be found in Nettleship Hell. pp. 132—134: cf. also his Lectures and Remains II pp. 123—126. Plato deals here chiefly with the hygienic aspect of gymnastic—a subject which was much discussed in his day: see Dict. Ant. I p. 929, where we are reminded that gymnasia were dedicated to Apollo, father of Asclepius, and himself a god of healing. In his interesting treatise Die Platonischen Dialoge in ihrem Verhältnisse zu den Hippokratischen Schriften (Landshut 1882) Poschenrieder has shewn that Plato was strongly influenced throughout this passage by the views of Hippocrates and his school. See also Häser Lehrb. d. Gesch. d. Med. etc. I pp. 94 ff. The athletics of Gymnastic are treated of in Laws 795 D ff., 833 ff.

ψυχὴ ἀγαθὴ ‐‐ βέλτιστον . No very recondite theory of the relation of body and soul is here involved. Plato simply means that the soul has more power over the body than the body over the soul. (The restriction in ὡς οἷόν τε should be noted.) On this principle some doctors held that to cure the body one should minister to the mind diseased: see the curious passage in Charm. 156 B —157 C. The general sentiment is well illustrated by J. and C. from Democr. Fr. Mor. 128 (Müllach) ἀνθρώποισι ἁρμόδιον ψυχῆς μᾶλλον ἢ σώματος ποιέεσθαι λόγον· ψυχὴ μὲν γὰρ τελεωτάτη σκήνεος μοχθηρίην ὀρθοῖ, σκήνεος δὲ ἰσχὺς ἄνευ λογισμοῦ ψυχὴν οὐδέν τι ἀμείνω ποιεῖ.

εἴπομεν . 398 E.

ἀθληταὶ ‐‐ ἀγῶνος . Cf. Laws 829 E ἀθλητὰς τῶν μεγίστων ἀγώνων, and Lach. 182 A .

τῶνδε means contemporary athletes: cf. IV 425 C note With Plato's strictures on Greek athletics cf. Arist. Pol. Θ. 4. 1338^{b} 10 (with Susemihl and Hicks' note) and especially Eur. Fr. 284 : for his attack on the diet and training of athletes cf. Dict. Ant. I pp. 98, 928 and the authorities there cited.

ἐὰν σμικρὰ ἐκβῶσιν κτλ. Poschenrieder (l.c.) cites the Hippocratean Praedictiones II c. 1 Littré τοὺς ἀθλητὰς γινώσκειν...ἤν τι τοῦ σιτίου ἀπολίπωσιν, ἢ ἑτεροῖόν τι φάγωσιν, ἢ ποτῷ πλέονι χρήσωνται, ἢ τοῦ περιπάτου ἀπολίπωσιν ἢ ἀφροδισίων τι πράξωσι· τούτων πάντων οὐδὲν λανθάνει, οὐδ’ εἰ σμικρόν τι εἴη ἀπειθήσας ὥνθρωπος.

ὥσπερ κύνας . II 375 A.

πολλὰς μεταβολὰς κτλ. Cf. [Hippocr.] de umoribus V p. 496 C. 15 Littré αἱ μεταβολαὶ μάλιστα τίκτουσι νοσηματα καὶ αι μέγισται μάλιστα καὶ ἐν τῇσιν ὥρῃσιν αἱ μεγάλαι μεταλλαγαὶ καὶ ἐν τοῖσι ἄλλοισιν: cf. also Aphorism. IV p. 486 § 1 al. and Pl. Laws 797 D ff. (Poschenrieder l.c. pp. 31 ff.)

ἁπλῆ ‐‐ πόλεμον . The sentence is usually explained by carrying on ἡ βελτίστη γυμναστικὴ ἂν εἴη and regarding ἁπλῆ—γυμναστική as the predicate both to ἡ βελτίστη γυμναστική and to ἡ τῶν περὶ τὸν πόλεμον. Besides its extreme cumbrousness, this view makes Plato say that the best gymnastic is good ( ἐπιεικής is practically synonymous with ἀγαθή), which is, to say the least, unnecessary. It seems to me much simpler and better to make ἐπιεικὴς γυμναστική the subject to ἁπλῆ. The meaning is: will the best course of training be sister to the music we described? How so? ἐπιεικὴς γυμναστική, like ἐπιεικὴς μουσική (this is the force of καί), is ( ἐστὶ understood) ἁπλῆ, and so above all is that of soldiers. Hartman, who saw that the passage must be taken in this way, would write ἡ for καί, and I once preferred καὶ <ἡ>, but the article can be dispensed with (cf. 401 D note), and καί is necessary. As the emphasis is primarily on γυμναστική, some may prefer to read γυμναστικὴ ἐπιεικής or γυμναστικὴ ἡ ἐπίεικής; but if the stress of the voice is laid on γυμναστική, and ἐπιεικὴς γυμναστική treated as a single expression (cf. V 453 A note), I think the text may stand.

οὔτε ἰχθύσιν κτλ. Cf. Eubulus ap. Athen. I 25 C (Jackson).

ἐν Ἑλλησπόντῳ is rejected by Cobet and Hartman; if the Homeric heroes were ἐν Ἑλλησπόντῳ, the fish forsooth would more easily have eaten them than they the fish! This is however so obvious that even Cobet's “scriba sciolus” would have seen it, and avoided the preposition ἐν. The fact is that Ἑλλήσποντος was constantly used to denote the whole coast stretching from the Pontus to the Aegean, including Bosporos and Propontis. See Stein on Hdt. IV 38 and cf. Thuc. II 9. The usage is also found in Inscriptions (Meisterhans^{3} p. 226. 16). An Athenian of Plato's day was much more likely to employ the name Ἑλλήσποντος in this idiomatic sense than a later copyist; and for this reason I have no doubt that the expression is genuine, although the words of Hartman “nihil refert utrum ἐν Ἑλλησπόντῳ an ἐν Αἰγύπτῳ sint” are nearly, if not quite, true. Plato may however intend to remind us that fish were plentiful in the region of the Hellespont: cf. Il. IX 360 and Athen. IV 157 B.

ὡς ἔπος εἰπεῖν . I 341 B note

καὶ ὀρθῶς γε ‐‐ ἀπέχονται . ὀρθῶς must be taken with both verbs: ‘Yes, and they do well in knowing it and in abstaining.’

Συρακοσίαν ‐‐ ὄψου . For δέ (‘autem’) Stallbaum unnecessarily reads δή. The Συρακοσία τράπεζα was proverbial: see Blaydes on Ar. Fr. 206 and the curious account of Syracusan gluttony in Pl. Epp. VII 326 B ff. There is no sufficient basis for Cobet's idea that Plato is here borrowing from some comic poet. Later scandal insinuated that it was the delights of Syracusan living that drew Plato thrice to Sicily (Hermann Gesch. u. System p. 116 note 133, where the authorities are cited).

Κορινθίαν κόρην . Cf. 11 373 A note Κορινθία κόρη is a grisette: see the commentators on Ar. Plut. 149 , and on the general subject Blümner Privatalt. pp. 254—256. φίλην is more refined for ‘mistress’ ( ἑταῖρα). The word κόρην has been doubted: “innocentem puellam eicere ex Platonis republica voluerunt triumviri praestantissimi Buttmannus, Morgensternius, et nuperrime Astius.” So says Stallbaum, her successful champion.

Ἀττικῶν πεμμάτων . The fame of Athenian pastry was as great as its variety: see Athen. XIV cc. 51—58 and other references in Blümner l.c. p. 220.

παναρμονίῳ . See on 399 C.

ἰατρεῖα were both dispensaries and consulting-rooms etc. See Laws 646 C and other references in Blümner l.c. p. 359. In some ἰατρεῖα patients were also housed and treated by doctors (Häser Lehrbuch d. Gesch. d. Med. etc. I pp. 86 ff.), so that in certain cases they resembled a sort of private hospital. For the remedial conception of punishment prevailing in the whole of this section see II 380 B note

δικανική . Cobet calls for δικαστική, and at first sight δικαστῶν just below seems to favour his view. But Plato deliberately selects the less reputable word, meaning by it the arts by which men try to lead the true δικαστής (cf. Ap. 40 A ) astray: see infra B, C. In his own city there is no δικανική, but only δικαστική (409 E, 410 A). It appears from Laws IV 720 C ff. that a doctor's assistants were usually slaves, and that slaves for the most part treated slaves, and freemen freemen, but the rule was not universal (see Blümner l. c. p. 359 note 1). Plato holds that the increase of citizen doctors points to the spread of self-indulgence among the free-born population.

καὶ ἀπορίᾳ οἰκείων has suffered severely at the hands of critics, who have bracketed καί (Ast and others), or read καὶ ἀπορία οἰκείων (Hermann), or δικαίων ἀπορίᾳ οἰκείων (Madvig), or finally denounced the words as a ‘futile interpretamentum.’ Schneider explains καί as “idque” (“und zwar” in his translation), and so also Prantl, and Shilleto (on Dem. F. L. § 101). This interpretation appears to me forced and unnatural. It is simplest to make ἀπορίᾳ as well as τῷ δικαίῳ depend on χρῆσθαι, and regard χρῆσθαι ἀπορίᾳ as equivalent to εἶναι ἄποροι, just as χρῆσθαι ἀμαθίᾳ (for example) means no more than εἶναι ἀμαθεῖς. The plural οἰκείων does not refer to δεσποτῶν, but is the genitive of οἰκεῖα, which means ‘resources of one's own,’ ‘personal resources’) ( ἐπακτῷ παρ’ ἄλλων. Cf. the use of τὰ οἰκεῖα in the literal sense for res familiaris I 343 E al.

ἦ δοκεῖ κτλ. Glauco has said that χρῆσθαι ἐπακτῷ τῷ δικαίῳ is the most disgraceful thing of all. Socrates asks him whether it ( τοῦτο) is more disgraceful than the other case ( τούτου) which he is about to mention; and Glauco's reply is ‘no: this other case is even more disgraceful than the first’ (infra C). The meaning was missed by the critic who (see Rev. de Philol. XV p. 83) ingeniously suggested the insertion of ἢ οὔ; after δικαστοῦ just before Glauco's reply. In what follows the litigiousness of the Athenian nature is satirised.

ἀποστραφῆναι : an expressive and epigrammatic condensation for ἀπολυθῆναι στρεφόμενος. λυγιζόμενος is rightly explained by the Scholiast as στρεφόμενος, καμπτόμενος, ἀπὸ τῶν λύγων· λύγος δέ ἐστι φυτὸν ἱμαντῶδες. The corruption λογιζόμενος (found in all MSS except A and—according to Rostagno —M) was easy and almost inevitable.

παρασχεῖν δίκην . The same phrase appears in Eur. Hipp. 49 , 50, and Herwerden should not have proposed ὑποσχεῖν. Plato's view in the Gorgias is that the guilty should denounce themselves to the judge and be cured by suffering punishment: see II 380 B note

φύσας τε καὶ κατάρρους . The order is chiastic, φύσας referring to πνευμάτων, and κατάρρους to ῥευμάτων. Plato clearly indicates that the medical use of these words was only beginning in his day, and it is the application of these words to diseases which he derides, not the words themselves when used of bellows, blasts, and torrents (see the Lexica). The experiment in language is better preserved by rendering ‘blasts and torrents’ than ‘flatulence and catarrh.’ For φῦσα cf. (with Poschenrieder l. c. p. 47) [Hippocr.] de flatibus VI p. 94 c. 3 Littré πνεύματα δὲ τὰ μὲν ἐν τοῖσι σώμασι φῦσαι καλέονται, τὰ δὲ ἔξω τῶν σωμάτων ἀήρ, and ib. c. 7 ὅταν οὖν τὸ σῶμα σιτίων πλησθῇ, καὶ πνεύματος πλησμονὴ ἐπὶ πλέον γίγνεται τῶν σιτίων χρονιζομένων· χρονίζεται δὲ τὰ σιτία διὰ τὸ πλῆθος οὐ δυνάμενα διελθεῖν· ἐμφραχθείσης δὲ τῆς κάτω κοιλίης, ἐς ὅλον τὸ σῶμα διέδραμον αἱ φῦσαι. Other examples of the use of the term in the Hippocratean corpus are cited by Stephanus-Hase Thes. s.v. With κατάρρους cf. Crat. 440 C ἀτεχνῶς ὥσπερ οἱ κατάρρῳ νοσοῦντες ἄνθρωποι. The word is found in the Hippocratean writings, and denotes “defluxionem aut omnem humoris ex capite ad os et asperam arteriam, atque per eam ad pulmonem, delationem ac descensum” (StephanusHase s. v., where examples are quoted).

τοὺς κομψοὺς Ἀσκληπιάδας . The epithets κομψοί and χαρίεντες were often applied to the more advanced and scientific sort of physicians (Blümner Privatalt. p. 358 note 2). The Ἀσκληπιάδαι were a well-recognised sect or college of physicians, with schools in Cyrene, Rhodes, Cos and Cnidos. See Günther in Iwan Müller's Handbuch V 1 p. 103, and Hug on Symp. 186 E .

καὶ μάλ) ‐‐ ὀνόματα : ‘Yes, indeed, these are truly’ etc. Glauco does not reply to οὐκ αἰσχρὸν δοκεῖ, but simply corroborates what Socrates has said about the new medical terminology. This is simpler than to place (with Schneider) a colon after ἔφη, and take καὶ μάλα with αἰσχρόν. The asyndeton on Schneider's view is too harsh, and would almost require the insertion of καί before ὡς, or (if ὡς ἀληθῶς were taken as ὡς ἀληθῶς αἰσχρόν) before καινά; neither of which alternatives is satisfying. For similar inexactness in replies see V 465 E note

οἱ ὑεῖς ‐‐ ἐπετίμησαν . In themselves these words can only mean that Machaon and Podalirius (the two chief army doctors to the Greek host, Il. XI 833) found no fault with the damsel who gave the wounded Eurypylus an inflammatory potion, or with Patroclus, who was curing him, for directing or permitting her to do so. In our Homer, however, the potion is given, not to Eurypylus but to the wounded Machaon, by Hecamede, Nestor's slave ( Il. XI 624); and this is correctly related in Ion 538 B. The inconsistency led Ast to suspect the genuineness both of Εὐρυπύλῳ—see however 408 A—and of οὐδὲ Πατρόκλῳ τῷ ἰωμένῳ; but there can be little doubt that the text is sound. We must suppose either that Plato is confused, or else that in his text of Homer such a potion was administered, not only to the wounded Machaon (as in the Ion l.c.), but also to the wounded Eurupylus, with Patroclus' sanction. The first alternative is possible, and approved by Howes ( Harvard Studies etc. VI p. 198): but as it is clear from the Ion— if the Ion is genuine—that Plato was familiar with the story of Machaon's treatment, I think it more likely that Plato's Homer related a similar incident in connexion with the treatment of Eurypylus also. For the healing of Eurypylus see Il. XI 844 ff., XV 394.

οἶνον Πράμνειον . Athenaeus, alluding to this passage, informs us that Pramneian wine was παχὺς καὶ πολυτρόφος (I 10 B).

φλεγματώδη : ‘inflammatory.’ Cf. [Hippocr.] περὶ νούσων IV c. 35 (VII p. 548 Littré) ἐπήν τις φάγῃ τυρὸν ἢ ὅ τι ἐστὶ δριμύ, ἢ ἄλλο τι φάγῃ ἢ πίῃ ὅ τι ἐστὶ φλεγματῶδες, αὐτίκα οἱ ἐπιθέει ἐπὶ τὸ στόμα καὶ τὰς ῥῖνας (Poschenrieder l.c. p. 49).

τῇ παιδαγωγικῇ ‐‐ ἰατρικῇ . Cf. Tim. 89 C παιδαγωγεῖν δεῖ διαίταις πάντα τὰ τοιαῦτα—ἀλλ’ οὐ φαρμακεύοντα κακὸν δύσκολον ἐρεθιστέον.

Ἡρόδικον . Herodicus, a native of Megara, and afterwards a citizen of Selymbria, is mentioned by Plato again in Prot. 316 E and Phaedr. 227 D . He was one of the earliest to study scientifically the therapeutics of exercise and diet, and particularly recommended long walks, according to Plato ( Phaedr. l.c. τὸν περίπατον Μέγαράδε. Cf. Häser Lehrb. d. Gesch. d. Med. etc. I p. 94). The description of his health given here is confirmed by Aristotle Rhet. I 5. 1361^{b} 4—6 πολλοὶ—ὑγιαίνουσιν ὥσπερ Ἠρόδικος λέγεται, οὓς οὐδεὶς ἂν εὐδαιμονίσειε τῆς ὑγιείας διὰ τὸ πάντων ἀπέχεσθαι τῶν ἀνθρωπίνων ἢ τῶν πλείστων (a passage curiously misunderstood by J. and C., who seem to take λέγεται for λέγει). Plato himself thoroughly appreciates the connexion between γυμναστική and ἰατρική: see for example Gorg. 452 A ff., 464 B ff., Soph. 228 E , Pol. 295 C .

νοσώδης γενόμενος . εἰς φθίσιν ἀνήκεστον πάθος ἐμπεσών, says Plutarch ( de his qui sero etc. 554 C).

μακρὸν ‐‐ τὸν θάνατον κτλ. Cf. Eur. Suppl. 1109 —1113 μισῶ δ’ ὅσοι χρῄζουσιν ἐκτείνειν βίον | βρωτοῖσι καὶ ποτοῖσι καὶ μαγεύμασι | παρεκτρέποντες ὀχετὸν ὥστε μὴ θανεῖν· | οὓς χρῆν, ἐπειδὰν μηδὲν ὠφελῶσι γῆν, | θανόντας ἔρρειν κἀκποδὼν εἶναι νέοις, and Aesch. Fr. 395 , Soph. Fr. 689 .

δυσθανατῶν : not “dum malam obit mortem” (Stallbaum), but ‘dying hard’ like δυσθνῄσκων.

καλόν . Because he was the first to profit by his own invention. The assonance γῆρας—γέρας is quite in Plato's manner: cf. IV 439 C, VI 487 C, VIII 557 C notes

οὐδενὶ σχολὴ κτλ. Steinhart ( Platon's Werke V p. 172) thinks it strange that so idealistic a thinker as Plato should not recognise the power of spiritual strength to rise superior to bodily weakness. This truth was not ignored by Plato (see infra 408 E and VI 496 B), although here, perhaps, he forgets that conspicuous examples of fortitude and resignation have a political as well as a private value: “they also serve who only stand and wait.”

καύσει ἢ τομῇ . The two methods of ancient surgery: see Blümner Privatalt. p. 353 note

μακρὰν has less authority than μικράν (see cr. n.), but is probably right. The contrast with the immediate remedies just described seems to require an allusion to the duration of the regimen: cf. also μακρὸν—τὸν θάνατον in B above. μικράν is not sufficiently defended by a reference to κατὰ σμικρόν in 407 D, nor by the allusion to πιλίδια καὶ τὰ τούτοις ἑπόμενα. Moreover σμικρός, and not μικρός, is the prevailing form throughout the Republic. μικρός appears to occur only in V 453 D and VI 498 D. On the inscriptional usage see Meisterhans^{3} p. 89.

πιλίδια . Felt caps were worn by the sick and delicate (see the references in Blümner l.c. p. 180 note 5); but as artisans and sailors usually wore felt caps too ( Dict. Ant. II p. 427), Plato perhaps alludes to some special coverings for the head prescribed by doctors from time to time in a course of medical treatment. The plural also points to this. If not, he uses the expression quite generally, as an example of the treatment he condemns. Well-to-do Greeks generally went bareheaded.

εἶπεν . The ‘momentary’ aorist well expresses the carpenter's decided businesslike tone. His view of life resembles that of the ‘meditative skipper’ in Gorg. 511 D ff.

ὑγιὴς ‐‐ ἀπηλλάγη . He regains his health on losing his doctor, or if he dies, dies without help. Cf. Plut. Apophth. Lac. 231 A τοῦ δὲ ἰατροῦ εἰπόντος αὐτῷ, Γέρων γέγονας, Διότι, εἶπεν, οὐκ ἐχρησάμην σοι ἰατρῷ. (The anecdote is told of Pausanias the Spartan king.)

ἦν . The carpenter is now dismissed: hence the imperfect ἦν, which should be retained in translating. Stallbaum (followed by J. and C.) explains ἦν as the ‘philosophic’ imperfect= ἐστίν, ὡς ἄρτι ἐλέγομεν (in 406 C). This is much less simple and lively. “Wohl weil er ein Geschäft hatte, bei dessen Unterlassung es ihm nicht erspriesslich war zu leben?” Schneider, rightly. Cf. II 361 C note

ἔργον προκείμενον . The view of work and duty here presented recalls I 352 E—353 E.

ἀκούεις . Phocylides, being dead, yet speaketh. The present ἀκούεις is just as legitimate as φησί, and well expresses the living voice of poetry in oral circulation. Heindorf (on Gorg. 503 C ) misses the point of the idiom when he says that ἀκούεις is for ἀκήκοας; while Stallbaum's explanation ‘probas’ is positively wrong. The line, as restored by Bergk Phoc. Fr. 10, is δίζησθαι βιοτήν, ἀρετὴν δ’ ὅταν ᾖ βίος ἤδη. The Horatian ‘quaerenda pecunia primum, | virtus post nummos’ gives the meaning, if primum and post are understood in a strictly temporal sense. Phocylides' maxim is one of the earliest expressions of the all but universal cry χρήματα χρήματ’ ἀνήρ (first in Alcaeus Fr. 49 B ergk), which Socrates and Plato continually preached against. It will be noticed that Plato for his own purposes represents Phocylides as laying the stress on ἀρετὴν ἀσκεῖν rather than on δίζησθαι βιοτήν, where it really falls.

τοῦτο : viz. τὸ ἀρετὴν ἀσκεῖν, as explained in the margin of A.

τῇ προσέξει τοῦ νοῦ is added as a kind of afterthought or additional specification, precisely like the infinitives in Gorg. 513 E ἐπιχειρητέον ἐστι τῇ πόλει καὶ τοῖς πολίταις θεραπεύειν, infra 407 C, IV 437 B, 443 B, V 450 B, X 598 B, Crito 52 B. The datives τεκτονικῇ etc. depend grammatically on ἐμπόδιον only, and have nothing to do with προσέξει. παρακέλευμα presently is of course the accusative, the subject to ἐμποδίζει being νοσοτροφία, and οὐδέν adverbial. Richter (in Fl. Jahrb. 1867 p. 140) should not have revived the reading of Bekker μελετῶντι ἢ νοσοτροφία· τεκτονικῇ μὲν γὰρ κτλ., which is lacking both in authority and point.

ναὶ μὰ τὸν Δία ‐‐ εἰκός γε, ἔφην (in C). See cr. n. With the MS reading εἰκός γ’ ἔφη, the distribution of the speeches causes difficulty. It will be enough to mention three alternatives, for no one has adopted or is likely to adopt the punctuation of A, where σχεδόν γέ τι —περὶ τοῦ σώματος is assigned to Socrates. We may give either (1) the whole speech ναὶ μὰ—περὶ τοῦ σώματος to Glauco, excising εἰκός γε, ἔφη with II q and some other MSS (so Schneider 1830); or (2) ναὶ μὰ—ἐπιμέλεια τοῦ σώματος to Glauco, and καὶ γὰρ—περὶ τοῦ σώματος to Socrates (Stallbaum); or (3) ναὶ μὰ—δύσκολος to Glauco, and τὸ δὲ δὴ—περὶ τοῦ σώματος to Socrates (Baiter and others, including Schneider 1842). The first view fails to account for the appearance of εἰκός γε ἔφη in A, but is right, I think, in assigning the whole speech to Glauco. Neither at καὶ γὰρ πρὸς οἰκονομίας nor at τὸ δὲ δὴ μέγιστον is it easy and natural to change the speakers. The simple expedient of writing ἔφην for ἔφη appears to me to set matters straight. For the corruption see Introd. § 5. οὐκοῦν ταῦτα etc. is also said by Socrates.

ἥ γε περαιτέρω κτλ. ‘This excessive care of the body, which goes beyond what sound bodily regimen permits.’ The Greek has a rhetorical effect like τὸ δεινόν, τὸ μέγα ἐκεῖνο—θρέμμα IX 590 A. With περαιτέρω and the genitive cf. Gorg. 484 C περαιτέρω τοῦ δέοντος. I once conjectured ἥ γε περαιτέρω γυμναστική, ἧς (‘cuius est’) etc. ( Cl. Rev. X p. 385), but Plato seems to mean that treatment of this kind has no claim to the name γυμναστική at all, and not that it is γυμναστική run mad. The MS reading is defended also by a reviewer of my Text of the Republic in Hermathena XX p. 252.

πρὸς ἑαυτόν : with μελέτας, as in μελετᾶν, φροντίζειν πρὸς ἑαυτόν etc.

διατάσεις , though its MS authority (see cr. n.) is slight, can hardly fail to be what Plato wrote. Similarly in 546 C A^{1} has ἕκαστον twice for ἑκατόν. See Introd. § 5.

ὅπῃ ‐‐ ἐμπόδιος . The reading of Ξ (followed by Stallbaum and the older editors) is certainly right. αὕτη (sc. ἔστιν) is νοσοτροφία; and ἐμπόδιος ἀρετῇ ἀσκεῖσθαι is exactly like ἐπιχειρεῖν τῇ πόλει θεραπεύειν (see 407 B note). The presence of νοσοτροφία makes it impossible for virtue to be practised or tested, as when, for example, to take a pedant's illustration, a boy evades both lectures and examinations by cherishing a nervous headache. Recent English editors have followed Baiter, and read ὅπῃ ταύτῃ ἀρετὴ ἀσκεῖται καὶ δοκιμάζεται, taking ταύτῃ as ἐν φιλοσοφίᾳ, but this gives a much less satisfactory meaning. After αὕτη had been changed to ταύτῃ, the rest of the corruption was easy; but a trace of the original reading may survive in the ἀρετῇ (not ἀρετή) of A.

ἀποκεκριμένον : an isolated, local malady; “morbum separatum, non totum corpus afficientem” (Ast). Unnecessary difficulty has been raised. The word is in no sense technical, and ἀποκρίνω in the sense of ‘separate’ is common enough. The corruption ἀποκεκρυμμένον might have been foretold.

τούτοις μέν κτλ. The words τοὺς μὲν ὑγιεινῶς ἔχοντας led us to expect ἰᾶσθαι, but the construction changes in order to introduce the invention of medicine, and the ‘healing’ reappears in a different form in φαρμάκοις τε—δίαιταν. The sentence is bad grammar, but good conversational style of the looser kind. It is not easy to say whether τε after φαρμάκοις connects the clauses, or only φαρμάκοις with τομαῖς. The former use is comparatively rare in Plato (Hoefer, de part. Plat. p. 7). Partly on this ground, and partly because the union of the aorist καταδεῖξαι with προστάττειν is a little awkward, I prefer the second alternative. The asyndeton, which is of the usual explanatory or ampliative kind, is in keeping with the loose structure of the whole sentence, and seems to me to add a certain didactic impressiveness here: cf. 409 B. τὰ δ’ εἴσω—σώματα depends not so much on ἀπαντλοῦντα directly as on the composite notion ἀπαντλοῦντα καὶ ἐπιχέοντα, which expresses a certain mode of treatment, and is as it were a species of the general idiom ποιεῖν τινά τι. φυτεύειν must depend on ποιεῖν. Plato's sentences are seldom so disjointed as this: cf. however VI 488 B ff., VIII 558 A.

μὴ οἴεσθαι : for the negative (which is the more natural here, as it belongs logically to δεῖν, though grammatically to οἴεσθαι) see I 346 E note οἴεσθαι, like ἐπιχειρεῖν, depends on φῶμεν .

λυσιτελῆ is taken by Schneider as the accusative neuter in apposition to the idea in θεραπεύειν. If so, αὐτῷ for αὑτῷ must be written (with A). It is however so natural to take λυσιτελῆ as masculine that Plato would surely have expressed the other meaning in a less ambiguous way. The usual view yields a satisfactory sense, and should be preferred.

δῆλον κτλ. See cr. n. The awkwardness of taking ὅτι as ‘because’ was early felt and led to the insertion of δεικνύοιεν ἄν in several MSS ( καὶ· οἱ παῖδες αὐτοῦ δεικνύοιεν ἂν ὅτι τοιοῦτος ἦν)—a reading adopted by the older editors. Few will now dispute that δεικνύοιεν ἄν is a gloss. Besides Schneider's suggestion, which I adopt, two other proposals merit consideration: (1) δῆλοι, ἦν δ’ ἐγώ, καὶ οἱ παῖδες αὐτοῦ ὅτι τοιοῦτος ἦν (Sauppe, comparing Crito 44 D), (2) δῆλον, ἦν δ’ ἐγώ, καὶ οἱ παῖδες αὐτοῦ ὅτι τοιοῦτοι. ἢ οὐχ ὁρᾷς κτλ. (Madvig). The first, though regarded as possible by Schneider ( Addit. p. 25), involves what is, to say the least, a very exceptional use of δῆλος, with which “subiectum sententiae verbo ὅτι incipientis idem esse solet quod sententiae primariae” (Hartman). Sauppe's parallel from the Crito is a doubtful exception to Hartman's rule. Moreover οὐχ ὁρᾷς κτλ. is too lively: we should expect ἢ (so Π ^{2} Ξ^{2} and other MSS) οὐχ ὁρᾷς κτλ. Madvig's correction already involves two changes ( τοιοῦτοι and ἤ), but would be improved by making a third, viz. δῆλοι for δῆλον. The minimum of dislocation which yields a satisfactory sense is the reading which suggested itself to Schneider, although he did not himself adopt it. Some may be inclined to pronounce ὅτι τοιοῦτος ἦν a marginal gloss on δῆλον, as once occurred to Hartman.

ὡς ἐγὼ λέγω . ὡς is emphatic, ‘in the way I describe.’

αἷμ) ‐‐ ἔπασσον . Il. IV 218 αἷμ’ ἐκμυζήσας ἐπ’ ἄρ’ ἤπια φάρμακα εἰδὼς | πάσσε, said of Machaon only. Plato ingeniously accommodates the line to his own purposes. ἐκμυζήσαντ) is of course the aorist indicative ἐκμυζήσαντο, not the dual participle as J. and C. hold. This was pointed out by Schneider. Verbs denoting any kind of organic action are apt to be middle in Attic (Rutherford New Phrynichus pp. 138 ff.). It would be easy to write ἐκμύζησάν τ) (as I once did) and retain Homer's active, but it is not worth while.

κἂν εἰ has come to mean no more than καὶ εἰ: cf. infra V 477 A, IX 579 D, X 612 C and Jebb on Soph. El. pp. 224 f. The change from the plural πιόντες to the singular νοσώδη has been doubted by Herwerden; but see I 347 A note In illustration of κυκεῶνα πιόντες Schneider ( Addit. p. 25) refers to Hippocr. περὶ διαίτης ὀξέων II p. 304 f. Littré οἱ γὰρ ἀρχόμενοι τῶν ὀξέων νουσημάτων ἔστιν ὅτε οἱ μὲν σιτία ἔφαγον—οἱ δὲ καὶ κυκεῶνα ἐρρόφεον· ἅπαντα δὲ ταῦτα κακίω μέν ἐστιν ἢ εἰ ἑτεροίως τις διαιτηθείη κτλ.

Μίδου πλουσιώτεροι : with reference (as Stallbaum observes) to Tyrt. 12. 6 (Bergk): cf. Laws 660 E.

τραγῳδοποιοί τε καὶ Πίνδαρος . Aesch. Ag. 1022 f., Eur. Alc. 3 , Pind. Pyth. 3. 55—58 ( αἴθων δὲ κεραυνὸς ἐνέσκιμψεν μόρον).

πειθόμεθα was much more likely to be corrupted to πεισόμεθα (so q, with Stallbaum and others) than vice versâ, on account of φήσομεν. The present is more pointed and expressive; our rule has been laid down (391 D), and we abide by it now and always.

ὅσοι κτλ. Glauco's conception of the medical art resembles that of the later ἐμπειρικοί: see Celsus de med. Proem. pp. 5—9 ed. Daremberg, and infra 408 D note

καὶ μάλα ‐‐ λέγω . Socrates replies to ἆρ’ οὐκ—ἰατρούς; ignoring, or nearly ignoring, εἶεν δ’ ἂν—ὡμιληκότες: cf. V 465 E note καὶ μάλα is simply ‘certainly,’ and ἀγαθοὺς λέγω lays stress on ἀγαθούς: ‘that is, if they are really good,’ ‘ good ones, I mean.’ There is perhaps a hint that the good physician and the good judge must also be good men: cf. 409 C. To substitute with Hartman μάλιστα for μάλα (as in many MSS) is to mistake the force of λέγω.

ἰατροὶ κτλ. The combination of scientific knowledge ( πρὸς τῷ μανθάνειν τὴν τέχνην) and medical experience which Plato desiderates reminds us of the standpoint of the μεθοδικοί, whose principles were in some respects a compromise between those of the δογματικοί or Theorists, and those of the Empirics: see Celsus l.c. pp. 9—13 and Häser Lehrb. d. Gesch. d. Med. etc. pp. 245 ff., 268 ff.

ψυχῇ ψυχῆς . Cf. Gorg. 523 C —E.

αὐτήν : ipsam, not eam, as Jowett apparently translates it.

διὸ δὴ καὶ εὐήθεις . “For unstained thoughts do seldom dream on evil: Birds never limed no secret bushes fear” ( Rape of Lucrece). Cf. infra VII 517 D ff., Theaet. 174 C ff. The use of παραδείγματα recalls Theaet. 176 E , though the idea is somewhat different here. The word means ‘models,’ ‘standards,’ not ‘samples of experience’ (J. and C.), and τοῖς πονηροῖς is equivalent to τοῖς τῶν πονηρῶν παραδείγμασι. Cf. infra C, D, where παράδειγμα τοῦ τοιούτου is ‘a model’ (not ‘a sample’) ‘of such a character.’ So also Schneider, who translates by ‘Vorbild.’

ὀψιμαθῆ κτλ. The common taunt ὀψιμαθής is in such a case an epithet of praise.

οὐκ οἰκείαν κτλ. For the asyndeton see 407 D note δεῖ αἰσθάνεσθαι for διαισθάνεσθαι (Stob. Flor. 45. 96) is ingenious, but weak. As Steinhart remarks ( Einleitung p. 173), the scientific knowledge of virtue, according to Socrates and Plato, implies a knowledge of its opposite, viz. vice: see on I 334 A, and cf. infra 409 D.

οἰόμενος . Cf. Theaet. 173 B δεινοί τε καὶ σοφοὶ γεγονότες, ὡς οἴονται, and I 336 A note

καὶ πρεσβυτέροις κτλ. The touching allusion to Socrates' condemnation will not escape the sympathetic reader. Plato seldom talks in this vein without thinking of his master: cf. Theact. 174 C and the still more affecting words in VII 517 A. It is from incidental references such as these that we can best appreciate the profound influence which the death of Socrates exercised upon Plato. See also VIII 560 D note

πονηρία μὲν γὰρ ‐‐ λήψεται . See on 409 B and the suggestive remarks of Stewart on Aristotle's Eth. Nic. V 1. 1129^{a} 17. Strictly speaking, Vice cannot have scientific knowledge ( ἐπιστήμη) even of herselt, since Vice is ignorance (and scientific knowledge of Vice would imply a scientific knowledge of Virtue); but she recognises herself by ἐμπειρία οἰκεία: cf. 409 B.

ἀρετὴ δὲ κτλ. : ‘whereas Virtue will in course of time, if natural endowments are improved by education, attain to scientific knowledge at once of herself and Vice.’ The contrast between πονηρία μέν and ἀρετὴ δέ is much impaired if we connect ἀρετή with φύσεως (in the sense of ‘a virtuous nature’): and for this reason I now agree with Schneider in thinking φύσεως παιδευομένης a genitive absolute. I formerly accepted Richards' emendation παιδευομένη, which is decidedly more logical, if φύσεως depends on ἀρετή: but Schneider's view is better. χρόνῳ belongs to λήψεται, and not to παιδευομένης (as if ‘educated by time,’ Jowett): mere lapse of time will never give ἐπιστήμη. Cf. ὀψιμαθῆ and ἐν πολλῷ χρόνῳ διαισθάνεσθαι in 409 B.

ἀποθνῄσκειν ἐάσουσιν . Cf. Plut. Apophth. Lac. 231 A κράτιστον δὲ ἔλεγε (sc. Παυσανίας ) τοῦτον ἰατρὸν εἶναι τὸν μὴ κατασήποντα τοὺς ἀρρωστοῦντας, ἀλλὰ τάχιστα θάπτοντα. In laying down this law, Plato speaks from the standpoint of the Regal or Political Art, prescribing for the subordinate arts of Medicine and Justice the conditions under which it is good to live and good to die. See Grote Plato I p. 362.

αὐτοί = ipsi is said in opposition to the mere ‘permission to die’ which bodily disease requires. αὐταί (suggested by Richards) is unnecessary: see II 377 C note

ἆρ’ οὖν κτλ. This epilogue describes concisely the aim and underlying principle of Plato's earlier scheme of education. Its object is to produce citizens who shall combine gentleness and strength—sensibility and courage—intellectual activity and moral stedfastness. It is an ideal in which the distinctive virtues of Athens and Sparta—of Greece and Rome—are united and transfigured. See II 375 C and the passages referred to there. The ideal of Pericles ( φιλοσοφεῖν ἄνευ μαλακίας) in many ways resembles Plato's (Thuc. II 40). Cf. also Nettleship Hell. pp. 88—90 and Bosanquet Companion pp. 115—117. It is noteworthy that the doctrine of this section is best explained by a comparison with one of the dialogues often held to be late ( Pol. 306 C —311 C): see also Laws 773 C, D. This is not pointed out by Krohn in his otherwise acute analysis ( Pl. St. pp. 24—28).

ὁ μουσικὸς ‐‐ αἱρήσει . ὁ μουσικός is ὁ τῇ ἁπλῇ μουσικῇ χρώμενος, as defined in the last sentence. ἴχνη διώκων and αἱρήσει are metaphors from the chase: see II 375 A.

αὐτὰ μὴν ‐‐ ἰσχύν . The theory of gymnastic propounded here was apparently new in Plato's time (see on II 376 E), although the practice of athletics as an educative discipline, especially at Sparta, conformed to it in no small measure (see Plut. Lyc. 17 ff., Xen. Rep. Lac. 2 ff.).

μεταχειρίζονται . See cr. n. I have followed Hermann in adopting Galen's text. With οὐχ ( μὴ ) ὡς or ὥσπερ the verb should have for its subject the nominative contained in the ὡς clause: cf. VII 539 D, X 610 D. Symp. 179 E is in reality no exception to this rule.

τινες . It has been supposed that τινες refers to Isocrates, who in his Antidosis (180—185) expounds at length the usual Greek view of gymnastic. This is possible only if the present section was added within the last four years or so of Plato's life, which is most improbable. See Hirmer Entst. u. Komp. d. pl. Pol. p. 663, and Introd. § 4. In other passages the Antidosis has been held to presuppose the Republic: see Dümmler Chronologische Beitr. etc. pp. 12, 13.

καθιστᾶσιν . Cf. Dem. 24. 145 οὗτος γὰρ (sc. ὁ νόμος)— οὐκ ἐπὶ τοῖς κεκριμένοις—κεῖται, ἀλλ’ ἐπὶ τοῖς ἀκρίτοις, ἵνα μὴ—ἀναγκάζοιντο ἀγωνίζεσθαι, and Phil. 34 C (where however it is easy to write λάβωμεν). In the first of these cases the reference is, as here, to the establishment of laws or ordinances. καθιστᾶσιν is used somewhat like φησί 407 A. Madvig's emendation καθίστασαν commends itself to Weber ( Entwick. d. Absichtssätze in Schanz's Beiträge II 2 p. 58) and others, but has not yet been proved to be necessary, and καθιστάναι below tells rather against it. For other examples of the idiom see Kühner Gr. Gr. II pp. 897, 898. Cases like Soph. O. C. 11 and El. 57, 760 are different, and have been justly emended. As regards the sentiment, it is characteristic of Plato to invent a historical sanction for his theories (cf. 414 B ff.); but he doubtless sincerely believed that the spirit of Greek gymnastics had degenerated.

ἦν δ’ ἐγώ . There seems to be no other case in which ἦν δ’ ἐγώ is so long deferred. Stallbaum and Bekker insert the words after σκληρότητος without any MS authority. If change is needed, ἦν δ’ ἐγώ had better be omitted (so q, whose reading is very different here). But it is better to note than to obliterate such peculiarities.

ἢ ὡς is not ‘pro simplici ἤ vel ὡς positum’ (Stallbaum), but=‘quam quomodo.’

αὐτοῦ . Does the pronoun mean τοῦ ἡμέρου or τοῦ φιλοσόφου? Four qualities are first distinguished: viz. the wild, the hard, the soft and the tame. The source of wildness is the spirited element, which if rightly cultivated becomes brave, if unduly strained, hard. So far, all is clear; but difficulties now begin. We should expect Plato to continue: τὸ ἥμερον is an attribute of τὸ φιλόσοφον, and τὸ φιλόσοφον —not τὸ ἥμερον—when relaxed becomes too soft, when rightly educated becomes κόσμιον (the virtue which contrasts with τὸ ἀνδρεῖον). At first sight, then, it looks as if αὐτοῦ meant ‘the philosophic temperament’ (so Stallbaum and J. and C.); but this is grammatically impossible, unless we make τὸ ἥμερον the subject to μαλακώτερον εἴη and therefore to ἥμερόν τε καὶ κόσμιον, which is hardly tolerable. We must therefore acquiesce in taking αὐτοῦ as τοῦ ἡμέρου, unless there is corruption somewhere. If Plato had written καὶ μᾶλλον μὲν ἀνεθὲν μαλακώτερον εἴη τοῦ δέοντος, καλῶς δὲ τραφὲν σῶφρόν τε καὶ κόσμιον, everything would be clear, but I do not venture to change the text. σῶφρον for ἥμερον is suggested also by Krohn ( Pl. St. p. 26). Apelt proposes ἔμμετρον ( Berl. Philol. Wochenschr. 1895 p. 969).

ἀμφοτέρα ‐‐ φύσει : viz. τὸ θυμοειδές and τὸ φιλόσοφον.

σώφρων τε καὶ ἀνδρεία . σωφροσύνη is the virtue of τὸ φιλόσοφον, ἀνδρεία of τὸ θυμοειδές: cf. 399 C and Pol. 307 C . The meaning would be caught more easily if Plato had written—as perhaps he did— σῶφρόν τε καὶ κόσμιον for ἥμερόν τε καὶ κόσμιον above, just as he wrote ἀνδρεῖον (410 D). ἄγροικος (implying, like ἀνελεύθερος, ἀνδραποδώδης, with which it is coupled in Laws 880 A, lack of power to control the feelings) is properly opposed to σώφρων here.

καταυλεῖν ‐‐ ἁρμονίας . καταυλεῖν (as Ast observes) does not govern ἁρμονίας, but is used absolutely: cf. Laws 790 E (of mothers singing and rocking their children to sleep) ἀτεχνῶς οἷον καταυλοῦσι τῶν παιδίων, καθάπερ αἱ τῶν ἐκφρόνων βακχειῶν ἰάσεις, ταύτῃ τῇ τῆς κινήσεως ἅμα χορείᾳ καὶ μούσῃ χρώμεναι. So expressive a word could ill be spared, although van Heusde's καταντλεῖν is ingenious enough. Cobet would read καταντλεῖν and cut out καὶ καταχεῖν, while Hartman inclines to eject καταυλεῖν καί, but the text is sound. ἁρμονίας depends on καταχεῖν. With χώνης cf. (with Hiller Fl. Jahrb. 1874 p. 174) Ar. Thesm. 18 δίκην δὲ χοάνης ὦτα: see Blaydes ad loc. The context in Aristophanes lends some colour to Hiller's notion that the comparison was taken from some earlier philosopher: cf. Theophr. de sensu § 9.

σίδηρον ἐμάλαξε κτλ. See on 387 C. Apparently then the first effect even of the μαλακαὶ ἁρμονίαι is good. This apparent inconsistency with 398 E ff. is emphasized by Krohn ( Pl. St. p. 25), but Krohn fails to observe that Plato is here describing the facts of common experience, whereas before he was making laws of his own. It is quite possible to admit that the relaxing modes are beneficial in moderation, and yet forbid them, because moderation in them is difficult to maintain.

ὅταν ‐‐ τήκει . The object of κηλῇ, τήκει and λείβει is τὸ θυμοειδές: that of ποιήσῃ is τὴν ψυχήν. So much is, I think, certain; but ἐπέχων is less easy. The word has been interpreted as (1) ‘listening to’ (Schneider, comparing 399 B, where, however, ὑπέχοντα should probably be read), (2) ‘pressing on,’ ‘persevering,’ ‘continuing’: cf. Theaet. 165 D ἐπέχων καὶ οὐκ ἀνιείς (J. and C.). The sense which Schneider gives to ἐπέχων is ill-supported: and we must accept the second alternative. Morgenstern's emendation ἐπιχέων (accepted by Herwerden and Hartman) is attractive but not quite convincing (“when he ceases not to pour the music in” etc.). ἐπιχέων would preserve the metaphor, which is clearly intended (in καταχεῖν, χώνης, and σίδηρον ἐμάλαξε) to suggest the process of smelting, and of which an echo still survives in τήκει, λείβει and ἐκτήξῃ. See Blümner Technologie etc. IV pp. 108 ff. notes The θυμοειδές is the iron which music softens and may even dissolve: farther than this the comparison is not to be pressed.

κηλῇ : as one might charm or fascinate a snake: Euthyd. 290 A , Phaedr. 259 A .

καὶ λείβει ‐‐ αἰχμητήν . For λείβει thus used cf. Ar. Knights 327 . μαλθακὸς αἰχμητής is said of Menelaus in Il. XVII 588.

ἐὰν ‐‐ λάβῃ : ‘if he has received,’ not ‘if he act upon’ (J. and C.). Plato means that if the individual in question received at the beginning a soul— ψυχήν is understood—naturally spiritless, he soon makes it a ‘feeble warrior.’ “Wenn er gleich eine von Natur zornlose Seele bekommen hat” (Schneider). The subject throughout is the τις with which the sentence began. For the usual Greek idiom, by which the person concerned is represented as acting on himself ( ἐκτήξῃ τὸν θυμόν etc.) instead of being acted on, cf. Eur. I. A. 187 φοινίσσουσα παρῇδ’ ἐμὰν | αἰσχύνᾳ νεοθαλεῖ with Headlam's note: also V 462 C, D notes and IX 572 A note

ἐρεθιζόμενον . ῥιπιζόμενον, suggested by Herwerden, is picturesque enough: but ‘provoked and extinguished’ is even more natural in Greek than in English, for ἐρεθίζω could readily be used of fanning a fire: see the lexica s.v.

ἀντὶ θυμοειδοῦς ‐‐ ἔμπλεοι . θυμοειδοῦς is of course masculine and not neuter (as J. and C. suggest). Even if we allow that the dative is neuter in cases like Symp. 195 C νέος—ἐστι, πρὸς δὲ τῷ νέῳ ἁπαλός, and Theaet. 185 E , the presence of the article makes all the difference. Ast (with Ξ) reads θυμοειδῶν. So harsh a change from plural to singular ( ὀργίλοι but θυμοειδοῦς) is remarkable, but hardly more so than ἀποθανουμένους ὅς in IV 426 C. Cf. also I 347 A note Krohn points out that ἀνδρεῖον is here represented as a μεσότης between σκληρόν and ὀργίλον ( Pl. St. p. 27).

εὐωχῆται : should be understood literally, of good living.

ὥσπερ θήριον ‐‐ διαπράττεται . If the MSS are right, πάντα is masculine. But although διαπράττεσθαι by itself can be used without an expressed object ( Prot. 319 C al.), it is strange to find διαπράττεσθαι πρός τινα so used: see Crat. 395 B , Alc. II 143 C. On this account διαπράττεται has been by some ejected (Hermann), by others emended into διαράττεται (Morgenstern), διατάττεται (Madvig and one Florentine MS); while others read θηρίον τὰ πάντα διαπράττεται (Lambrechts), or προσδιαπράττεται πάντα (Chandler), or expunge πρός (Bywater). Perhaps we should read ὥσπερ θηρίον πρὸς <θηρίον> κτλ. (‘attains all his ends by violence and ferocity, like one wild beast with another’). Cf. Shakespeare Rape of Lucrece “The rough beast that knows no gentle right.”

σκαιότητι . ‘Ineptitude.’ Cf. Soph. Ant. 1028 with Jebb's note.

ἀχαριστίας is ‘ungraciousness.’

εἰ μὴ εἴη πάρεργον occurs also in Phaed. 91 A (according to the Bodleian MS). Phrases of this kind seldom admit of variation; for which reason we should hesitate to admit the εἰ μὴ εἰ πάρεργον of Π^{1}.

ὅπως ἄν κτλ. The soul has, so to speak, two strings, the φιλόσοφον and the θυμοειδές, which make a kind of ἁρμονία when they are tuned to the proper pitch by Music and Gymnastic. The θυμοειδές is slackened ( ἀνίεται) by μουσική, tightened or braced ( ἐπιτείνεται) by γυμναστική (410 D, 411 A—E); conversely, we must suppose that the φιλόσοφον is slackened by γυμναστική, and tightened by μουσική. Music and Gymnastic are therefore both of them necessary for each of the two strings (cf. IV 441 E note), although the slackening of the θυμοειδές of itself also tightens the φιλόσοφον, which is likewise slackened when the tension of the other is increased. Cf. Tim. 88 B , C. The effect of all this musical imagery is to suggest that Character is the Music of the Soul: cf. Lach. 188 D .

τοῦ ‐‐ ἐπιστάτου . Some MSS (including Ξ and q) omit τοῦ, and no precise parallel has yet been adduced for ὁ τοιοῦτός τις used in this way. In IX 581 E, cited by Schneider ( Addit. p. 27), Paris A has ἐν τοιούτῳ τινί, not ἐν τῷ τοιούτῳ τινί. The article may perhaps be justified by the description of the ἐπιστάτης in the last sentence, and τινὸς taken closely with τοιούτου (‘some such superintendent as we have described’); but there is certainly some ground for suspecting interpolation (with Bekker and others). The ἐπιστάτης, as Jowett observes, is a sort of minister of education, such as we find in Laws 765 D ff. The same function is in Pol. 308 D ff. assigned to the Regal or Political Art.

ὡς οἷόν τέ γε μάλιστα . I have placed a comma before ὡς; cf. Phaed. 74 B φῶμεν μέντοι νὴ Δἴ, ἔφη ὁ Σιμμίας, θαυμαστῶς γε (Hoefer Part. Plat. p. 33).

χορείας ‐‐ ἱππικούς . See Laws 814 D ff., 822 D ff., 830 C ff., 832 D ff.

οὐκέτι : ‘not now,’ sc. when we have trained our Guardians. On such idiomatic uses of οὐκέτι and its opposite ἤδη see Cope's Rhetoric of Aristotle, Vol. I p. 13.

πρεσβυτέρους κτλ. The different principles on which rulers may be appointed are fully discussed in Laws 690 A ff.

φρονίμους κτλ. Intellectual ability and accomplishments, authority, and pa triotic sentiment are the three requisites of the Rulers as laid down here. In VI and VII it is the first which is emphasized, here it is the last. This is in harmony with the whole spirit of I—IV, in which, as Krohn remarks ( Pl. St. p. 29), “the intellect is subordinated to the moral powers, and with the education of the character in richly-endowed natures the fruits of insight ripen of themselves.”

καὶ ἐκείνου . See cr. n. Stobaeus ( Flor. 43. 152) reads καὶ ὅτι (or ὅ τι) μάλιστα ἐκείνου κτλ., which is good enough Greek, and would mean ‘whatever policy he thinks by bringing prosperity to the other brings prosperity also to himself,’ ὅ τι being an accusative of respect belonging to εὖ πράττοντος. If the principle of this interpretation is right, I should read ὅ τι ἄν for ὅταν, taking ἄν with ξυμβαίνειν. ἄν loves the shelter of a relative, particularly ὅ τι, and the corruption is the easier because ὅταν in A and other MSS is written ὅτ’ ἄν. But φιλεῖν cannot well be said of one's attitude to a policy or course of action; and Hermann's proposal gives a more satisfactory sense. The occurrence of ( τοῦτ) ό γ’ ἂν μάλιστα just before may be responsible for the slip. It is to be understood—though Plato has not expressly said so—that the guardians believe their own interests to be best consulted by promoting those of their country. μὴ δέ is μὴ δὲ εὖ πράττοντος, and τοὐναντίον is ξυμβαίνειν καὶ ἑαυτῷ κακῶς πράττειν.

ἐπιλανθανόμενοι helps out the idea in ἐκβάλλουσιν and forms a natural antithesis to φυλακικοί which, while playing upon φύλακες, also implies the notion of remembering: cf. φύλαξ μουσικῆς 413 E. The word has been undeservedly attacked, chiefly because in 413 B it receives a more special and precise signification. But each of the three temptations to be presently enumerated, κλοπή, βία and γοητεία, may be correctly described as varieties of forgetting; nor is it in Plato's manner to introduce a classification prematurely, as he would have done by writing μήτε γοητευόμενοι μήτε βιαζόμενοι — < μήτε > ἐπιλανθανόμενοι (with Heller), or adding μήτε κλεπτόμενοι before ἐκβάλλουσιν (with Hartman). Cobet, more suo, expunges the word.

ἑκουσίως ἢ ἀκουσίως : with reference to the usual Socratic theory that Knowledge or Virtue is voluntary, Ignorance or Vice involuntary: see on II 382 A.

τῶν μὲν ἀγαθῶν ‐‐ ἑκουσίως . See IV 438 A note

ἢ οὐ ‐‐ εἶναι . It is necessary expressly to equate ἀληθεύειν with ἀληθὴς δόξα, because ordinarily it means to speak rather than to think what is true. Cf. II 382 A. Hartman approves of Ast for bracketing the words “quod argumentationem turbant,” but the contrary is true. Men unwillingly relinquish what is good. ἀληθεύειν is good; and ἀληθὴς δόξα is ἀληθεύειν; therefore we unwillingly relinquish ἀληθὴς δόξα—which is just what we wished to prove.

τραγικῶς : i.e. ὑψηλολογούμενος, in lofty high-flown metaphorical language such as may well become obscure: cf. VIII 545 E. κλέπτειν thus used is tragic: cf. (with J. and C.) Soph. Ant. 681 εἰ μὴ τῷ χρόνῳ κεκλέμμεθα .

τοίνυν =‘praeterea’ here, not ‘igitur’: I 339 D note

τι δείσαντες =‘having some fear’ (J. and C.).

τοῦτο ὡς ποιητέον κτλ. : ‘that it is their duty to do that which on each occasion they think it is best for them to do in the interests of the State.’ I have provisionally retained the reading of the best MSS, although it is open to suspicion on several grounds. The position of τοῦτο is unusual, and αὑτοῖς ποιεῖν is, to say the least, superfluous. Gaisford (with whom Cobet agrees) wished to expunge the entire clause as a gloss on δόγματος. This solution, though drastic, may be right: for an explanation of δόγματος is hardly needed after 412 D, E, and τοῦτο looks like the commencement of an explanatory note ‘this, viz. that’ etc. A simpler alternative, adopted by most editors, is to cancel αὑτοῖς ποιεῖν, but it is difficult to see why a scribe should have introduced the words. The sentence, if genuine, seems to want the finishing touch. Cf. 407 D note

προθεμένοις ἔργα . It is clear that Plato is referring to specific tests, and not (as Bosanquet seems to think) to the duties of war and the public service generally. So also Susemihl ( Gen. Entw. II p. 143), and Steinhart ( Einleitung p. 173), the latter of whom compares, not very aptly, the tests of the Pythagorean brotherhood and the appalling spectacles displayed in the mysteries. Three kinds of tests are required: (1) κλοπή, (2) βία, (3) γοητεία. Examples of the second kind are furnished by the severer discipline of gymnastic, the chase etc.: cf. Laws 633 B ff., where the probationary value of these and similar exercises is appropriately insisted on by the Spartan stranger. It was fully recognised in the Spartan ἀγωγή (Plut. Lyc. 17. 4 ff.). The third order of tests may be illustrated from Laws 634 A, B, 635 C, 647 D ff., 649 A, 673 E ff. ἡ ἐν οἴνῳ βάσανος (649 D) consists in giving wine to test men's selfcontrol ( τοῦ σωφρονεῖν ἕνεκα μελέτης 673 E). Plato gives no account of the first variety; but a good illustration of one species of it (cf. τοὺς μεταπεισθέντας 413 B) is provided by the speeches of self-seeking statesmen and unpatriotic sophists and poets. It is a curious fact that Plato's κλοπή still leaves a loophole by which vicious poetry may creep in again. On the general question, Plato does well to insist on the educational value of temptation; the theory and practice of modern times recognises it in connexion with βία, but experience too often shews that κλοπή and γοητεία mean playing with fire. Cf. Grote Plato III p. 328.

τοῦ τῆς ‐‐ θεατέον . Two εἴδη of tests have been described, κλοπή and βία: the third is γοητεία. I incline to think that Stallbaum is right in restoring τοῦ τῆς: see cr. n. and Introd. § 5. τούτοις ‘misere languet,’ and if a dative were needed, it should rather be αὐτοῖς. Herwerden expunges θεατέον; but asyndeton before ὥσπερ is frequent in sentences of this kind.

δυσγοήτευτος . For the change from plural to singular cf. I 347 A note

λαγχάνοντα . The accusative recurs to φύλακα, and is all the easier because τιμὰς δοτέον is little more than τιμητέον. Plato's usage is lax in such matters, and it is better not to emend: cf. (with Schneider) Laws 760 E, 877 A and Engelhardt Anac. Pl. Spec. III p. 45.

ὡς ἐν τύπῳ ‐‐ εἰρῆσθαι . Cf. VI 502 D note

ἐπικούρους . Plato henceforward uses this expression when he wishes specifically to allude to the second class of his citizens. φύλακες remains the general term including both ἄρχοντες and ἐπίκουροι. See on II 374 D.

ὧν νῦν δή . See cr. n. Although νῦν occasionally refers to the immediate past (e.g. I 341 C, IX 592 A, X 611 B: see also Jebb on Soph. Ant. 151 ), neither here nor in οὓς νῦν δή just before can δὴ νῦν be retained: for δή “neque per se intelligi neque ad ὧν referri potest” (Schneider). The reference is to II 382 D, III 389 B.

μάλιστα μέν . See on 415 D.

μηδὲν καινὸν κτλ. We want no novelty, but something with which the Greeks are already familiar, for our city is a Greek city (V 470 E).

Φοινικικόν τι : because the story of the Σπαρτοί was Phoenician, Cadmus the Phoenician having sown the dragon's teeth from which they sprang (Apollod. III 4. 1). Cf. Laws 663 E. Steinhart ( Einleit. p. 177) and Susemihl ( Gen. Entw. II p. 144) find in Φοινικικόν a further hint that the institution of caste was something foreign and non-Hellenic: but the words cannot be thus interpreted. The Egyptian system of caste (see Hdt. II 164 ff.) differed from Plato's in essential points, and there is no real evidence to shew that he was influenced by it in any way: nor is ‘Phoenician’ (‘Sidonian’ in Laws l.c.) equivalent to ‘Egyptian.’ Cf. Hermann Gesch. u. Syst. p. 55 and notes Ψεῦσμα Φοινικικόν afterwards became a proverb, perhaps owing to this passage.

πολλαχοῦ γεγονός means simply ‘which has happened in many places.’ γεγονός and γενόμενον in themselves refer to the actual occurrences, which ὥς φασιν—πεπείκασιν reduces again to legend and matter of faith. πολλαχοῦ is plentifully illustrated in Preller Gr. Myth. pp. 79 ff. Presently οὐδ’ οἶδα εἰ γενόμενον ἄν (for which Herwerden neatly but needlessly suggests οὐδ’ οἶδ’ ἂν εἰ γενόμενον) hints that the age of miracles is past.

ὁποίᾳ ‐‐ ἢ ποίοις . Cf. 400 A note It is very exceptional to find the indirect interrogative preceding the direct: cf. Soph. O. T. 71 with Jebb's note.

ἐρῶ . I have removed the colon after ἐρῶ on Richards' suggestion.

ὥσπερ ὀνείρατα ‐‐ αὐτούς : lit. ‘all these things which they fancied themselves suffering and happening to them were so to speak dreams.’ ἐδόκουν is ‘imagined’ as in Aesch. Pers. 188 (also of a dream) and elsewhere. The object of πάσχειν, viz. ταῦτα πάντα, becomes the subject of γίγνεσθαι: cf. (for the change of subject) Ap. 40 A , Symp. 200 D and supra I 333 C, II 359 D, E, 360 A. It must be allowed that the effect of this idiom is here unusually harsh. I once conjectured ὑπάρχειν for πάσχειν, taking ἐδόκουν still as ‘fancied’: but the text is probably sound.

ὑπὸ γῆς κτλ. Herwerden bids us bracket either ὑπό or ἐντός: but Plato rarely if ever lets the preposition ἐντός follow its noun. ὑπό is ‘under,’ not ‘by’ (it is ὁ θεός, not ἡ γῆ, who πλάττει, infra 415 A), and ἐντός is adverbial; “drinnen unter der Erde” (Schneider). Mortal creatures are similarly moulded within the earth in Protagoras' prehistoric myth ( τυποῦσιν αὐτὰ θεοὶ γῆς ἔνδον 320 D): cf. also Symp. 191 C , Pol. 272 A , Tim. 42 D . The myth of the Politicus (269 A ff.) connects the autochthonous origin of man with the golden age, in agreement with a wide-spread tradition, which gave rise to a considerable literature (Dümmler Proleg. zu Platons Staat p. 46). It is in the spirit of this tradition that Plato here represents the first generation of his ideal city as autochthonous.

καί ‐‐ καί . The double καί marks “the correspondence of the two clauses” (J. and C.). As the Earth proved herself their mother, so they must shew themselves her sons. If the text is sound, it must be explained in this way; but exact parallels are rare. Thuc. IV 8. 9 (cited by Schneider Addit. p. 27) is different: see Classen ad loc. and on VIII 27. 5. More to the point is Soph. Ant. 1192 f. καὶ παρὼν ἐρῶ, | κοὐδὲν παρήσω τῆς ἀληθείας ἔπος: see Jebb ad loc. Ast expunges καί before ἡ γῆ, while Hermann alters it to ὡς (carrying on the ὡς of ὡς ἄρα). Neither change can be called satisfactory. I formerly suggested δημιουργουμένη ἔτι. ἤδη δὲ κτλ. ( Cl. Rev. X p. 385): cf. Symp. 220 C ἤδη ἦν μεσημβρία, καὶ ἄνθρωποι ᾐσθάνοντο. The change is slight, but ἔτι ‘languet,’ and it is better to retain the MS reading.

ὡς περὶ μητρὸς ‐‐ διανοεῖσθαι . Cf. (with J. and C.) Aesch. Sept. 10 —20, 412—416, and infra V 470 D. For the omission of the preposition before τῆς χώρας cf. VIII 553 B note ὑπέρ with τῶν ἄλλων is scarcely more than περί: see II 367 A note

ὡς φήσομεν . The sense (as Schneider observes) is ὡς ὁ μῦθος λέγει, ὃν πρὸς αὐτοὺς ἐροῦμεν. Hartman cancels ὡς, but it was more likely to have been wrongly omitted here than inserted.

χρυσὸν κτλ. The metals are borrowed from Hesiod ( O. D. 109—201), as Plato indicates in VIII 546 E. Hesiod enumerates five ages of men (interposing the age of heroes between those of copper and iron), but the older legend probably recognised four only: see Rohde Psyche^{2} I p. 87. Plato makes the golden and the other classes coexist—a truer and profounder view than Hesiod's. In other respects, the myth (as Jackson has pointed out in Susemihl and Hicks Politics of Aristotle p. 244) is not to be pressed: for “it does not recognise the promotion of ἐπίκουροι” to be ἄρχοντες. We should expect the φύλακες to contain admixtures, both of gold and silver, such as are to be Rulers receiving more gold than silver, and conversely; but the Greek does not favour this idea. Iron again seems to be exclusively (though less emphatically) reserved for the farmers, and copper for the artisans: cf. infra B, C, VIII 547 A, B, and Arist. Pol. B 5. 1264^{b} 14. It makes the ψεῦδος all the more γενναῖον and effective to tell the citizens that the classes are even more distinct than they really are.

ἅτε οὖν ξυγγενεῖς ὄντες is said with reference to the δέ clause, on which the stress falls. The fundamental kinship of the different classes will occasionally reassert itself in their offspring. So J. and C., rightly.

ἐξ ἀργυροῦ : sc. ἐκγόνου, which should also be supplied with χρυσοῦ. Plato sees in fancy the onward march of generations καθάπερ λαμπάδα τὸν βίον παραδιδόντες: cf. IV 424 A. Ast's proposal ἀργύρου should not have received the approval of Hartman; and D. and V. miss a characteristic touch by translating ἀργυροῦ “a silver parent.”

ἐάν τε κτλ. This provision is the corner-stone of Plato's State, and as soon as it gives way, the edifice is doomed (VIII 546 E—547 A). It is only by the elevation of the worthy and the degradation of the unfit that class-distinctions can be made to coincide with those of Nature (cf. IV 423 D); and unless they do, the foundation of the city, which is τὸ ἑαυτοῦ πράττειν, is sapped. Hence the emphasis with which Plato introduces this subject. His theory, it should be noted, conforms at least as much to the interest of the individual as to that of the State; for it provides congenial work for all according to their natural capacities, and uncongenial labour, whether above or below one's powers, is a fertile source of misery and crime. Aristotle ( Pol. B 4. 1262^{b} 27) seems to doubt if Plato's scheme was feasible. Granted rulers who are φρόνιμοι εἰς τοῦτο, δυνατοί, and κηδεμόνες τῆς πόλεως (412 C), in a small city—a thousand warriors, says Plato, will suffice (IV 423 A, cf. Grote Plato III p. 206 note) —it could probably be worked without much difficulty. See also IV 423 E ff. We are not of course to suppose that the child was once for all assigned to his class at birth; he would be watched and tested again and again, before being finally disposed of, so that the likelihood of mistakes on the part of the Rulers is greatly lessened. Cf. Tim. 19 A .

τιμήσαντες : not “having estimated their values” (J. and C.): but simply ‘they will do him honour and’ etc. The suggestions ἀντιτιμήσαντες or τιμήσαντες κατ’ ἀξίαν will hardly command assent. τιμήν in τιμὴν ἀποδόντες above may also be translated ‘honour’ if τὴν τῇ φύσει προσήκουσαν is taken in its full force: the honour appropriate to his nature and no more.

ὅπως μέντ’ ἂν κτλ. Cf. Laws 663 E—664 A. Grote justly observes that “Plato has fair reason for his confident assertion that if such legends could once be imprinted on the minds of his citizens, as portions of an established creed, they would maintain themselves for a long time in unimpaired force and credit” (l. c. III p. 188). The first generation of citizens would remain incredulous, but the γενναῖον φεῦδος would be impressed upon their children, and soon be universally believed. It would require but little effort for a Greek city like Plato's (V 470 E) to entertain in course of time a view which has so many points of contact with Greek tradition. Here Plato seems to hint that even his Rulers (for οἱ τούτων ὑεῖς must include these also) will in time believe; the Rulers of VI—VII might teach the legend as an ἐν δέοντι ψεῦδος, but would themselves refuse their assent.

σχεδὸν ‐‐ λέγεις : viz. that the story is intended to form part of the city's permanent religious creed, and so encourage patriotism and fraternity.

τοῦτο ‐‐ ἀγάγῃ : ‘this will be as the vox populi shall determine’: i.e. it will depend upon φήμη whether our fable is believed or not. φήμη is not of course an oracle (as Ficinus supposed), but the half-personified voice of popular belief. Cf. Laws 838 C, D.

τούς τε ἔνδον κτλ. Henkel ( Studien zur Gesch. d. Gr. Lehre vom Staat p. 52 note 13) remarks that the prevention of faction inside the city is characteristically put in the foreground. The greatest danger to a Greek city was from internal dissension: cf. V 470 C ff. notes

στρατοπεδευσάμενοι . The Spartan government was compared to that of a στρατόπεδον (Isocr. 6. 81: cf. Gilbert Gr. Const. Ant. E. T. pp. 61 ff.). Plato's city is literally a camp. His proposals would probably strike the average Athenian as a dangerous and tyrannical exaggeration of Spartan usages. See Jowett Introd. p. 176.

διαφέρειν ἐκείνου is rejected by Herwerden; but Schneider's explanation hits the mark: “ αὖ alterum hoc de discrimine insolentius dictum notat: prius fuerat quod domos εὐνάς dixerat.”

κακουργεῖν . See 407 B note The idiom is abundantly attested, both in Plato and in other Greek authors, although Madvig and Cobet have done their best to expel it from Plato's text here and wherever else it occurs.

μὴ τοιοῦτον ‐‐ ποιήσωσι . For τοιοῦτον cf. 388 D note Richter conjectured μὴ τοιοῦτοι—ποιηθῶσι, “parum venuste,” as Hartman mercifully says.

ἀντὶ ξυμμάχων ‐‐ ἀφομοιωθῶσιν . For the usual ampliative or explanatory asyndeton cf. 409 B. Aristotle objects that Plato's regulations would virtually divide his city into two hostile camps ( Pol. B 5. 1264^{a} 24), and Grote does not see “what reply the Platonic Republic furnishes to this objection” (l.c. III p. 213). In reply to Aristotle, Plato might have pointed to his regulations about the interchange of classes (415 B ff.), which would have the effect of binding them together more securely. Moreover, where each individual has the work to do for which he is best qualified, one fruitful cause of discontent and sedition is removed. The wives and families of the lower class would also tend to keep them quiet. Nor does Aristotle's objection allow sufficient weight to the training by which Plato tries to protect his guardians from such ‘spiritual pride’ as would alienate their subjects.

καὶ ἐγὼ εἶπον . See cr. n. καὶ ἔγωγ’ εἶπον, though generally retained, is surely wrong: it could only mean ‘I too, said I.’ No editor cites any other instance of ἔγωγε in this formula.

τοῦτο μὲν κτλ. prepares us for the second scheme of education in Book VII: cf. 412 B, 414 A notes

ἥτις ‐‐ ἐπαροῖ . αὐτούς is emphatic: “ipsos per se” (Schneider). The contrast is between the guardians in themselves, and in their dealings with the others. It is difficult to decide between παύσει—ἐπαρεῖ (Bekker and others) and παύσοι—ἐπαροῖ. The latter is exquisitius, and better supported on the whole. For the confusion between ‐ῃ (subjunctive) and ‐οι (optative) in A see Introd. § 5. Cobet calls for τοῦ instead of τούς before φύλακας, but φύλακας requires the article. παύειν with the infinitive is rare, and means ‘prevent,’ not ‘make to cease’: cf. Hdt. V 67 (with Stein's note) and Ar. Ach. 634 , where Reiske's conjecture πείσας should not be accepted.

πρῶτον μὲν κτλ. A certain measure of communism in property seems to have existed among the Pythagoreans (RP.^{7} p. 43); but there is no reason to suppose that Plato is deliberately borrowing from them here: cf. Steinhart Einleitung p. 179. The main object of Plato is of course to prevent the formation of private interests likely to compete with the claims of public duty. We remark that there has been no hint so far of common wives and children, although Blaschke ( der Zusammenhang d. Fam. u. Gütergemeinschaft d. pl. St. m. d. pol. u. phil. Syst. Platos p. 7) thinks he finds one in 415 A. Cf. 415 D note

ἂν μὴ πᾶσα ἀνάγκη . For ᾖ omitted see II 371 A note The conjecture ἥν for ἄν (Herwerden) is elegant, but superfluous.

τὰ δ) ἐπιτήδεια ‐‐ μισθόν . It is fair that the lower classes should provide the others with the means of leisure, for it is they who ‘reap all the benefit of the laborious training bestowed on the guardians.’ They are the ‘ultimate and capital objects’ of Plato's solicitude. Grote justly adds that “this is a larger and more generous view of the purpose of political institutions than we find either in Aristotle or in Xenophon” (l.c. III p. 213).

ταξαμένους is strangely represented in Schneider's translation by “zu bestimmten Zeiten.” It refers to the fixing of fees or payments in return for services rendered. Cf. Men. 91 B .

ξυσσίτια . A Spartan feature: see Gilbert Gk. Const. Ant. E. T. p. 65. Cf. Laws 762 B ff.

χρυσίον κτλ. So also in Sparta, according to Xen. Rep. Lac. 7. 6; with which cf. Plut. Lys. 19. 6, where the ephors are said to have put to death a friend of Lysander λαβόντες ἀργύριον ἰδίᾳ κεκτημένον. Plato is keenly conscious of the corrupting influence of wealth: see Gorg. 525 D ff., and cf. II 373 E, IV 421 D notes His guardians are φύσει πλουσίω τὰς ψυχάς (VIII 547 B) and need no other riches.

ὑπὸ ‐‐ ἰέναι : as though Wealth communicated a taint, like a murderer sub isdem trabibus ( ὁμωρόφιος). The Greek is much more expressive and picturesque than Apelt's conjecture ὑπὸ τὸν αὑτῶν ὄροφον προσιέναι ( Observ. Cr. p. 11).

σῴζοιντό τ’ ἂν καὶ σῴζοιεν . Cf. 621 B μῦθος ἐσώθη—καὶ ἡμᾶς ἂν σώσειεν. σῴζεσθαι of moral salvation is common in Plato: cf. e.g. VI 492 E, 502 B.

ὁπότε δ) αὐτοὶ ‐‐ ἔσονται . From this sentence it seems clear that the community of goods does not extend to the lower classes, although Aristotle complains that Plato has not said anything precise upon the matter ( Pol. B 5. 1264^{a} 15). Aristotle seems, however, to have understood that they were not to have common wives, for he cynically observes that it would have been better if they had, as then they would have been more divided and less likely to combine against the guardians (ib. 4. 1262^{a} 40 ff.).

καὶ ὁ Ἀδείμαντος κτλ. Adimantus' objection is the dying echo of the view already advocated by Thrasymachus, that a ruler should rule for his own profit: cf. I 343 A, 344 B notes Socrates declines to discuss the question now, because it is irrelevant. In the further account of the communism of the ruling class, the difficulty solves itself. A higher happiness—so we are told— comes from self-victory than from indulgence (V 465 D ff.: cf. IX 583 C note). Compare the conversation of Socrates with Aristippus in Mem. II 1. 17 ff.

μή . On μή with the infinitive after verbs of saying see I 346 E note

δἰ ἑαυτούς : i.e. they have themselves to thank for not being εὐδαίμονες. Cf. V 465 E οὐκ οἶδα ὅτου λόγος ἡμῖν ἐπέπληξεν ὅτι τοὺς φύλακας οὐκ εὐδαίμονας ποιοῖμεν, οἶς ἐξὸν πάντα ἔχειν τὰ τῶν πολιτῶν οὐδὲν ἔχοιεν; and Solon 33 1 f. οὐκ ἔφυ Σόλων βαθύφρων οὐδὲ βουλήεις ἀνήρ· | ἐσθλὰ γὰρ θεοῦ διδόντος αὐτὸς οὐκ ἐδέξατο (he of his own initiative refused). διά is used exactly as in I 354 B. This view, which is Ast's, gives an excellent meaning, and Schneider, who at first proposed a subtler explanation, adopts it in his translation (“durch ihre eigene Schuld”). The various conjectures δή, αὐτοῖς ὧν (Stephanus), δὴ αὐτοὺς ὧν (Buttmann), αὐτοὺς δἰ ὧν (Herwerden) need no refutation.

ἄλλοι : not οἱ ἄλλοι (Bekker, Stallbaum, etc.), which might be taken as referring to the lower classes in Plato's State. Plato would not be likely to permit these to have οἰκίαι καλαὶ καὶ μεγάλαι. ἄλλοι means ‘other rulers,’ i.e. rulers in other cities; and κεκτημένοι belongs to οἱ δέ: ‘possessing, like other rulers, lands,’ etc. So Schneider, rightly. For the idiomatic position of οἷον ἄλλοι cf. VII 515 A, 528 B, IX 589 B al.

μισθωτοί is not otiose as Badham supposes. We should translate ‘just like paid auxiliaries.’ The emphasis on μισθωτοί prepares us for Socrates' correction when he says they do not, strictly speaking, even get μισθός.

ἐπισίτιοι . ἐπίσιτοι (which Cobet and Hartman call for) would be more in accordance with the analogy of παράσιτος etc.; but the longer form is established by fragments of comic poets (ap. Ath. VI 246 F—247 A, where ἐπισιτίων in the fragment of Timocles defies emendation).

οἱ ἄλλοι : sc. ἐπίκουροι or mercenaries.

ἀποδημῆσαι . Regulations about ἀποδημία are laid down in Laws 949 E ff.

οἷα δὴ κτλ. For οἷα Hermann once conjectured οἷ: neatly, but οἷ is too precise. οἷα δή= οἷα δὴ ἀναλώματα. With οἱ εὐδαίμονες δοκοῦντες εἶναι cf. III 406 C. It is εὐδαιμονία in the popular sense of ‘having a good time’ which Adimantus complains is denied to the guardians.

οἶμον . A poetic word. Plato is perhaps thinking of some such phrase as Pindar's ἐπέων ο<*>μος ( Ol. IX 47). The ‘way’ is simply that each class must do its own appointed work, if the city is to be a happy and harmonious whole: cf. 423 D.

εἰ καὶ οὗτοι κτλ. καί means ‘as well as the rest of the city.’ Aristotle misrepresents Plato when, in spite of this sentence and V 465 D ff., he says that the guardians are deprived of εὐδαιμονία ( Pol. B 5. 1264^{b} 15 ff., with Susemihl's note). They are happy not only because they triumph over self (465 D), but—like the others—because they do the work to which Nature has called them: cf. I 352 D —354 A.

ὅπως ἕν τι κτλ. Cf. Laws 715 B and Thuc. II 60. 2, where Pericles says ἐγὼ γὰρ ἡγοῦμαι πόλιν πλείω ξύμπασαν ὀρθουμένην ὠφελεῖν τοὺς ἰδιώτας ἢ καθ’ ἕκαστον τῶν πολιτῶν εὐπραγοῦσαν, ἀθρόαν δὲ σφαλλομένην.

ᾠήθημεν ‐‐ σκεψόμεθα . See on II 369 A.

οὐκ ἀπολαβόντες ‐‐ τιθέντες . ἀπολαβόντες is absolute, almost adverbial (cf. Gorg. 495 E ); and ὀλίγους goes with τιθέντες. So Schneider and others rightly explain the construction.

αὐτίκα δὲ τὴν ἐναντίαν κτλ. Here we have the first express promise of Books VIII and IX, although the promise is afterwards fulfilled in an ampler manner than is indicated here. See also 427 D.

ὥσπερ οὖν ἂν ‐‐ μέλανι . Cf. (with J. and C.) Hipp. Maior 290 B. ἀνδριάντας γράφοντας means ‘painting statues of men.’ Cf. Euripides Fr. 764. 2 γραπτοὺς ἐν ἀετοῖσι προσβλέπων τύπους. The question whether statues were ever painted in the best period is an old controversy, the echoes of which have hardly yet died away. Schubart ( Fl. Jahr. 1874, pp. 20 ff.) and others prefer to take ἀνδριάντας merely as ‘likenesses of men,’ but the word was regularly, if not indeed always, used of statues. That the surface of archaic statues was regularly painted is now no longer doubtful: see Gardner Handbook of Greek Sculpture pp. 28 ff. During the best period, in the case of marble or other polished surfaces, the painting was regularly confined to the eyes, eyelids, eyebrows, hair and the like. See on the whole subject Sittl's Arch. der Kunst (in Iwan Müller's Handbuch) pp. 413, 414. μέλανι does not necessarily mean jet black, but only some dark and quiet colour. In point of fact, the eyes of the early marble statues on the Acropolis “are painted with a dark pigment, almost black” (Gardner l.c. p. 30). The use—regular in Greek—of γράφειν for painting is an interesting survival of the time when decorative art was little beyond carving in relief (Sittl l.c. p. 416). The present passage is strangely ignored by Sertorius in his interesting article “Plato und die Malerei” in Arch. f. Gesch. d. Phil. IX pp. 123—148.

ξυστίδας . The name ξυστίς was given to various kinds of purple robes or mantles—among them those worn by kings upon the stage, and by riders in festal processions. The authorities are cited in Müller Gr. Bühnenalt. p. 234 note I. If the Scholiasts on Ar. Clouds 70 and Theocr. II 74 are to be trusted, we should write ξύστιδας, not ξυστίδας.

ἐπὶ δεξιά . Whether we read ἐπιδέξια or ἐπὶ δεξιά the word should be understood as ‘from left to right.’ At a Greek banquet, the guests were always placed ἐπὶ δεξιά, i.e. so that the guest on your right hand occupied a lower place ( ὑποκατακεκλιμένος) than you, and the wine circulated from left to right of the banqueters. See Blümner Privatalt. p. 237 note 7 and Darbishire Relliq. Philol. p. 78. The word suggests a banquet with all the formalities, and heightens the incongruity of the situation, like the purple robes and golden crowns of the farmers. Schneider's exhaustive discussion seems to me conclusive in favour of writing ἐπὶ δεξιά as two words. Casaubon has been followed by most of the editors (except Schneider) in taking ἐπιδέξια as an adverb=‘commode’ (Ast), ‘commode et eleganter’ (Stallbaum etc.), or ‘dexterously,’ ‘cleverly’ (J. and C.); but it may well be doubted if the word could mean ‘commode,’ and ‘dexterously’ is inappropriate. Cf. Darbishire l.c. p. 78 note I. ἐπὶ δεξιά goes with κατακλίναντες and πρὸς τὸ πῦρ (cf. Blaydes on Ar. Ach. 751 ) with διαπίνοντας. The fire is that by which the potters bake their pottery; their workshop has for the nonce become a hall of banqueting.

ἐξ ὧν : i.e. τούτων τῶν σχημάτων ἐξ ὧν. Cf. II 373 E note

καὶ αὖ : rursusque (Ficinus), i.e. sicuti et contra, as Ast observes.

εἰ μὲν οὖν ‐‐ λέγοι . This difficult passage has suffered severely at the hands of critics, but the text is probably nearly, if not quite, sound. If we take the words as they stand in A, they mean, broadly speaking, that if we are making true guardians, and he ( ὁ ἐκεῖνο λέγων means the τις in 419 A) is making something different, he cannot, like ourselves, be speaking of a πόλις, but of something else. This is logical and gives an excellent sense: cf. 422 E εὐδαίμων εἶ— ὅτι οἴει ἄξιον εἶναι ἄλλην τινὰ προσειπει—ν πόλιν ἢ τὴν τοιαύτην οἵαν ἡμεῖς κατεσκευάζομεν. Now we are making guardians in the true sense of the term, such as are least likely to harm the city; whereas the author of the other proposals is making (not guardians, but since he gives them ἀγροί 419 A) a sort of farmers (cf. III 417 B οἰκονόμοι μὲν καὶ γεωργοὶ ἀντὶ φυλάκων ἔσονται) and men who do harm their city, because they “for their bellies' sake, Creep, and intrude and climb into the fold.” The advocates of such a theory must mean something different from a city—something like the “shearers' feast” in Lycidas: cf. I 343 A note γεωργούς is possibly corrupt; if so, I think we should read λεωργούς to contrast with ἥκιστα κακουργούς. The word occurs in the Memorabilia, if not in Plato. See Cl. Rev. X p. 385. Other emendations are enumerated in App. I.

ἐκεῖνο κτλ. : i.e. to pursue the other policy, which we enjoin. ἐκεῖνο does more than merely anticipate ὅπως— ἔσονται.

ἐατέον . The infinitive, which would naturally follow ἐατέον, is ‘drawn into construction’ with ὅπως —ἀποδίδωσι. I once thought of ἑκτέον (i.q. δεῖ ἔχειν, cf. V 468 A), taking the genitive as in πῶς ἔχεις τοῦ μεταλαμβάνειν εὐδαιμονίας; But the MS reading is satisfactory enough.

ὥστε ‐‐ γίγνεσθαι : ‘so that they also become bad.’ These words, though expunged by Hartman, are welcome, if not necessary, in view of κακίων χυτρεὺς γίγνεται and χείρους δὲ αὐτοί in D and E. καὶ indicates that κακοὺς γίγνεσθαι is more than διαφθείρει; and so it is represented in the sequel. The reading of A (see cr. n.) perhaps points to a variant διαφθερεῖ.

πλουτήσας ‐‐ τέχνης . Ar. Plut. 510 —534 (cited by Ast) furnishes an excellent commentary on this text. See also on III 416 E.

παρέχεσθαι is ‘to provide out of his own resources’ (de suo praebere): cf. VIII 554 A. Cobet cancels the word; Herwerden and Hartman prefer πορίζεσθαι, for which there is no MS authority. πορίζεσθαι would imply that the χυτρεύς buys his ὄργανα ready-made from others, whereas παρέχεσθαι expresses no opinion on this point.

διδάξεται . Thompson, Cobet, and others peremptorily call for διδάξει. See however Riddell Digest of Idioms § 87. Riddell conclusively shews (1) that in Men. 93 D ἐδιδάξατο as well as ἐπαιδεύσατο is said of a parent teaching his own son (a passage misunderstood—I think—by Jebb on Soph. Ant. 356 : cf. Men. 93 C ), (2) that ἐδίδαξε is used of a parent getting his sons taught by others in Men. 94 B and 94 D (bis). Another example of the second usage is Prot. 324 D . The fact is that “the Active Voice is quite as susceptible as the Middle of the meaning ‘to get a thing done by another’; neither Voice, however, by any proper inherent force, but in virtue solely of the common principle, that qui facit per alium facit per se,” Riddell. Jebb (l.c.) observes that “once or twice ἐδιδαξάμην is merely ἐδίδαξα with the idea of the teacher's interest superadded”: it may be doubted if “once or twice” is strong enough, but at all events this is the usage here, and in V 467 E. The active διδάσκῃ is appropriately used of teaching others ( ἄλλους κτλ.); in διδάξεται the personal interest reappears, for it is the sons who are the prominent pupils (whence ἢ ἄλλους and not καὶ ἄλλους). Richter's view ( Fl. Jahrb. 1867 p. 147) that διδάξεται denotes the result of the action rather than the action itself is partly true, but it is not the middle which gives it this force. In Ar. Clouds 783 , as Socrates is not Strepsiades' father, we may accept Elmsley's emendation διδάξαιμ’ ἄν for διδαξαίμην without prejudice to the present case.

αὐτοί : viz. οἱ τεχνῖται: see II 377 C note We need not change τεχνῶν to τεχνιτῶν.

ποιοῦντος = ‘producing’ gives a satisfactory sense. Wealth and Poverty are not to be allowed παραδῦναι εἰς τὴν πόλιν, because—we have here the statement of a general law—they are the authors of luxury etc. ἐμποιοῦντος (in Ξ and other MSS) is an obvious ‘emendation,’ though adopted by Stallbaum and others: cf. 444 D.

κακοεργίαν . If the form is right, Plato must intend to draw attention to the etymology of the word. κακουργίαν appears in two or three inferior MSS, and (as ε seems to be written over an erasure) was perhaps the original reading in Paris A.

πολλάκις : not ‘perhaps’ (one of J. and C.'s alternative suggestions) but ‘frequently,’ ‘repeatedly.’ πολλάκις does not mean ‘perhaps,’ except after εἰ, ἐάν, ἵνα, μή and the like: see Ast's lex. Plat. III p. 144 and Heindorf on Phaed. 60 E .

τί δέ; κτλ. ἂν πρεσβείαν κτλ. has for its apodosis οἴει τινὰς κτλ. I have placed a mark of interrogation after τί δέ. The alternatives are to place it after τῶν ἑτέρων, or else to suppose with Ast that the construction is suddenly changed at οἴει. Neither solution is so simple as to write τί δ); Cf. 425 C, 426 A, and (for the elision before a pause) 428 C.

κυσί . In the game of πόλεις, the counters were called ‘Dogs’ (Pollux IX 98). The comparison of our auxiliaries to dogs prepares the way for the allusion in 422 E: where see note. This has been pointed out by Ridgeway ( Journal of Hell. Studies XVI p. 288), who gives illustrations of three ‘dogs’ of this description found in Egypt and now in the British Museum.

εὐδαίμων εἶ κτλ. : ‘you are fortunate to be able to think etc.’: cf. V 450 C. εὐδαίμων is less common in this ironical sense than μακάριος.

ἑκάστη γὰρ κτλ. : ‘for each of them is, as the saying goes, no city, but a-many cities.’ The phrase τὸ τῶν παιζόντων in Plato seems always to mean ‘as they say in the proverb’ or ‘proverbial saying’: see IX 573 C, Laws 780 C, and cf. ib. 723 D. Now it is probable from the position of τὸ τῶν παιζόντων that ἀλλ’ οὐ πόλις forms part of the proverb: so that the whole saying may have run πόλεις μέν εἰσι παμπόλεις, ἀλλ’ οὐ πόλις. (Herwerden, more suo, cancels ἀλλ’ οὐ πόλις, but we have of course no right to take this step.) The form παμπόλεις for πάμπολλαι may be allowed in a pun on πόλεις, especially as the Epic plural of πολύς is sometimes found with feminine nouns. It should be remarked also that the first hand in Paris A wrote πάμπολαι (see cr. n.), though this may be merely accidental. What the ordinary application of the proverb was, we cannot say: presumably it was generally employed, as here by Plato, in speaking of a city divided against itself. The origin of the saying is to be sought in the variety of πεττεία known as πόλεις παίζειν, an expression which, according to the Scholiast on this passage of Plato, as well as Suidas s.v. πόλις, and Hesychius s.v. πόλεις παίζειν, had itself also a proverbial signification. In this game the abacus was divided into 60 spaces, each of which was called πόλις in ancient times (Photius s.v. πόλεις παίζειν ed. Porson. Porson's alteration of ξ/ i.e. 60 into ζ/ is a gratuitous change, as Schneider hints. See also Eustathius on Od. I p. 29. 13 ff., ed. Lips., quoted by Schneider). The name πόλις was moreover sometimes applied to the game itself (Cratinus Δραπετίδες Fr. 3 ed. Meineke καὶ κύνα καὶ πόλιν ἣν παίζουσιν), as well as to the πλινθίον or abacus on which it was played (Pollux IX 98). There is also, I think, some reason for believing that each of the players' sides was called collectively his πόλις. In Susemihl and Hicks Politics of Aristotle p. 148 note, Dr Jackson remarks that the words πάμπολλαι πόλεις, ἀλλ’ οὐ πόλις make it likely “that a compact body of pieces was called πόλις.” If we may go further, and suppose that the whole of a player's side was called his πόλις, the words of Plato δύο μὲν—πολεμία ἀλλἡλαις, ἡ μὲν πενήτων, ἡ δὲ πλουσίων· τούτων δ’ ἐν ἑκατέρᾳ πάνυ πολλαί receive additional point by becoming an exact counterpart of the game. A defeated player, gazing ruefully at his depopulated squares, each of which, as well as the whole of his side, is a ‘city,’ might therefore well exclaim, ‘Cities upon cities, but no city!’ for there can be no city without men ( ἔρημος ἀνδρῶν μὴ ξυνοικούντων ἔσω Soph. O. T. 57 ). I have thought of other possibilities, but this hypothesis as to the origin of the proverb suits the words of Plato better than any other which I can devise. For a different view see Hoffmann in Fl. Jahrb. 1863 pp. 240 ff. Cf. also Meineke Fr. Com. Gr. II pp. 44 f. It should be mentioned that Stewart ( Cl. Rev. VII p. 359) thinks there need be no allusion to the game of πόλεις in this passage, but only a jest about making one into many (cf. Men. 77 A παῦσαι πολλὰ ποιῶν ἐκ τοῦ ἑνός, ὅπερ φασὶ τοὺς συντρίβοντάς τι ἑκάστοτε οἱ σκώπτοντες), while Schneider finds only a “lusus in verbis atque in consociatione singularis et pluralis.” Neither of these suggestions meets the situation.

δύο ‐‐ κἂν ὁτιοῦν ᾖ : ‘two, in any case,’ lit. ‘if there be even anything at all,’ i.e. ‘whatever there be.’ So also Schneider. The subject to ὁτιοῦν ᾖ is impersonal, and not the city, as Jowett seems to suppose.

πολεμία . On this—comparatively rare —termination of the dual feminine in Plato see Roeper de dual. usu Pl. pp. 3 ff. Cf. IX 587 B note

εὐδοκιμεῖν . Stallbaum and others read δοκεῖν with one inferior MS. But εὐδοκιμεῖν is at least equally good: ‘great, I do not say in fame, but great in the true sense of the word “great.”’ σωφροσύνη is a city's truest greatness, not aggression, and “the applauding thunder at its heels. Which men call Fame.”

χιλίων . Aristotle takes this seriously as fixing the number of Plato's ἐπίκουροι ( Pol. B 6. 1265^{a} 9), but it is only the minimum: see 423 B note We hear of constitutions of a thousand very frequently throughout Greek political history, especially in the Greek colonies of Italy; and Plato may have had some of these precedents in his mind, both here and in Pol. 292 E ἐν χιλιάνδρῳ πόλει. See Whibley Gk. Oligarchies pp. 134 ff. By Aristotle's time the ruling Spartans, it is interesting to notice, numbered under 1000 ( Pol. B 9. 1270^{a} 29—31). See Grote Plato III p. 206 note

καὶ πολλαπλασίας is the predicate to δοκούσας, and καί means ‘even.’ So J. and C. rightly.

μέχρι ‐‐ πέρα δὲ μή . The extent to which the city may safely increase beyond 1000 προπολεμοῦντες (and the necessary farmers etc.) is therefore left to the judgment of the guardians. Like every natural organism, it should grow to the limits prescribed for it by nature (cf. 424 A note); but Plato probably conceived of it even in its maturity as relatively small. The regulations about marriage and the interchange between the different classes would be easier to work if the State was not too large. See also on χιλίων in 423 A, and on the general subject Newman's Politics of Aristotle I pp. 313— 315.

μεγάλη δοκοῦσα : ‘seeminggreat’: see 422 E.

ἴσως points the irony, which is continued in φαυλότερον.

πρόσθεν . III 415 B, C notes

ἕκαστον : with ἕνα, not of course with ἔργον, as Hartman seems to suppose. With what follows cf. Laws 847 B ἀναγκαζόντων ἕνα μόνον ἀλλὰ μὴ πολλοὺς εἶναι and infra 443 E.

μία ‐‐ ἀλλὰ μὴ πολλαί . Aristotle's criticism ( Pol. B 2. 1261^{a} 17—^{b} 15) is interesting, but captious. Plato would entirely agree with him that τὸ ἴσον τὸ ἀντιπεπονθὸς σῴζει τὰς πολιτείας. ‘The reciprocity of services and functions’ between the three classes is the very foundation of Plato's city, which is far from being an undifferentiated unity. It is rather a ἓν ἐκ πολλῶν, the πολλά being the three divisions of the State. See Susemihl and Hicks l.c. I p. 215. φύηται should be noted; unity of this kind is κατὰ φύσιν.

σμικρότερον is still ironical. In what follows Plato speaks his real mind: cf. Laws 813 D.

ἓν μέγα . ἓν balances πολλά: we need but one regulation, ‘the proverbial one great thing, or rather not great, but adequate.’ J. and C. err in translating ἱκανόν “to a sufficient extent”; and Stallbaum in making λεγόμενον “quod dicebamus.” ἓν μέγα is illustrated by J. and C. from Pol. 297 A .

εὖ παιδευόμενοι . Does this refer to the scheme of education already given, or is it a promise of the philosopher's training in Books VI and VII? Krohn takes the former view ( Pl. St. p. 127), and (if we have regard only to the preceding discussion) it must be allowed that this is the natural interpretation of Plato's words. At the same time, it is not easy to see how the musical education of II and III would enable the guardians to grasp such a conception as the community of wives and children. And in the later books Plato expressly declares that the training necessary for the Rulers was inadequately discussed before: see VI 497 C ff., 502 D. For these reasons we must, I think, suppose that Plato when he wrote these words was thinking of the education still to be provided. Cf. also III 414 A.

τήν τε τῶν γυναικῶν κτλ. is the first mention of communism in wives and children. According to an ingenious chorizontic theory, it was this sentence which inspired the Ecclesiazusae of Aristophanes, to whose caricature Plato replies in Book V (Stein de Ar. Eccles. arg. etc. and Brandt Zur Entwick. d. Pl. Lehre v. d. Seelentheilen, p. 6). See on the whole subject App. I to Book V. In γάμων and παιδοποιίας there is a kind of zeugma: for κτῆσιν suits only γυναικῶν. Plato marks the difference by placing τε after τήν and not after τῶν. γάμους (conjectured by Richards) would depend on διόψονται; but διόψονται γάμους καὶ παιδοποιίας is surely an impossible expression.

κοινὰ τὰ φίλων . “Locus brevitatem loquendi paullo insolentiorem habet, quam sic explico: δεῖ πάντα ταῦτα ὅ τι μάλιστα ποιεῖσθαι κοινά, <*>στε κατὰ τὴν παροιμίαν κοινὰ τὰ φίλων εἶναι” (Schneider). Hartman's proposal to omit τὰ φίλων has much in its favour. It is more elegant to suggest than quote so familiar a proverb; and the note τὰ φίλων might well have been added by a scribe upon the margin. In V 449 C on the other hand the addition of τὰ φίλων is appropriate and right.

ὀρθότατα κτλ. Adimantus accepts the principle, both here and in V 449 C. The doubts which he expresses later concern not the principle, but the τρόπος τῆς κοινωνίας (ib.). It is obvious that the principle κοινὰ τὰ φίλων might be applied to marriage etc. in a sentimental kind of way, without involving such a kind of community as is afterwards described. As Rettig points out ( Proleg. p. 95 note), Adimantus takes ὅ τι μάλιστα as “in quantum fieri posset maxime.”

ἔρχεται κτλ. : ‘goes on growing like a circle.’ So Schneider, rightly. Others take κύκλος (1) as a hoop or wheel—“goes on with accumulating force like a wheel” (J. and C.), or (2) as an ever-widening circle in ruffled water (Krohn, Herwerden etc.). As to (2), κύκλος cannot mean a circle in water, unless we insert ἐν ὕδατι, which Herwerden has the audacity to do. If we adopt the first solution, we make κύκλος a specific kind of circle: but nothing in the context warrants this. It is also very doubtful if αὐξανομένη can= ‘with accumulating force’: certainly κύκλος αὐξάνεται could not bear this meaning; and to exclude αὐξανομένη from the comparison (as J. and C. also suggest) renders ὥσπερ κύκλος practically otiose. The fact is that the growth of a natural ( κατὰ φύσιν) city is just like the drawing of a circle in Plato's way of thinking. Like a circle it grows and expands, like a circle too, when its zenith is passed, it narrows to the inevitable end. Here it is only the growth which is dwelt upon; but ὥσπερ κύκλος seems to warn us of impending decay and foreshadow Books VIII—IX. For more on this point see my Number of Plato pp. 58—62. αὐξανομένη is ‘growing’ in the widest sense i.e. reaching its full maturity of size and strength and beauty; but in what follows Plato characteristically confines himself to what he conceived to be a city's truest growth, the improvement of the citizens.

τροφὴ γὰρ κτλ. Plato seems therefore to hold that acquired characters can be transmitted to posterity. The general sentiment may be illustrated by the quaint catches sung by choirs of old men, men in their prime, and boys at Sparta: (1) Ἁμὲς πόκ· ἦμες ἄλκιμοι νεανίαι, (2) Ἁμὲς δέ γ’ εἰμές· εἰ δὲ λῇς, πεῖραν λάβε, (3) Ἁμὲς δέ γ’ ἐσσόμεσθα πολλῷ κάρρονες (ap. Plut. Lyc. 21. 3). Cf. V 461 A.

εἰς τὸ γεννᾶν ‐‐ ζῴοις . Cf. V 459 A ff.

τούτου is not intended to anticipate the ὅπως clause, but means— like αὐτό below—our system of education. This is clear from διαφθαρέν, which is the antithesis to σῳζομένη above, and like it, is said of the παιδεία. τὸ μὴ νεωτερίζειν is in loose apposition to αὐτό.

τὴν ‐‐ ἀμφιπέληται . Od. I 351 f. τὴν γὰρ ἀοιδὴν μᾶλλον ἐπικλείουσ’ ἄνθρωποι κτλ. Plato's variant probably points to a different recension; for ἐπιφρονεύουσιν (sic) ἐπακούουσιν in Hesychius seems to refer to the same passage (Schneider). For the sentiment cf. Pind. Ol. 9. 48 αἴνει δὲ παλαιὸν μὲν οἶνον, ἄνθεα δ’ ὕμνων | νεωτέρων, Xen. Cyr. I 6. 38 and many other illustrations in Smyth Greek Melic Poets p. 174.

πολλάκις . 422 C note

τρόπον ᾠδῆς νέον . Pind. Ol. 3. 4 Μοῖσα δ’ οὕτω μοι παρεστάκοι νεοσίγαλον εὑρόντι τρόπον. Pindar would incur Plato's censure for these words.

ὑπολαμβάνειν : i.e. understand such to be the poet's meaning.

μουσικῆς τρόποι . In later musical theory τρόποι was technically used to denote the three varieties of musical composition— νομικός, διθυραμβικός, τραγικός. They were called τρόποι (according to Aristid. Quint. p. 30 Meib.) because they expressed different psychical characters ( διὰ τὸ συνεμφαίνειν πως τὸ ἦθος κατὰ τὰ μέλη τῆς διανοίας), because, in short, they were μιμήματα τρόπων. Plato's μουσικῆς τρόποι need not however be confined to Aristides' three varieties. On the connexion between musical and political changes see Laws 700 A—701 D. The connexion was recognised universally throughout Greece, and particularly at Sparta, where—as Pausanias (III 12. 10) tells us—Timotheus had his lyre confiscated for adding to it four new strings: cf. also Cic. de Leg. II 39. Wherever in the ancient Greek πόλις the conception of the individual is hardly separated from that of the citizen, moral and political changes are believed to go hand in hand; and the effect of music on morality is explained in III 400 D—401 A: cf. Laws 673 A τὰ μὲν τοίνυν τῆς φωνῆς μέχρι τῆς ψυχῆς πρὸς ἀρετῆς παιδείαν οὐκ οἶδ’ ὅντινα τρόπον ὠνομάσαμεν μουσικήν. Bosanquet raises the question whether musical innovations are the cause or only the symptoms of political. Plato, I think, regarded them primarily as the cause ( Laws ll.cc.). We can better understand their effect if we remember that they were accompanied by changes not only in rhythm, but also in the quality, ethical and otherwise, of the words sung; and if we also bear in mind the enormous influence of the theatre in Greek life. The latter point is emphasized in this connexion by Plato (ll.cc.) and Aristoxenus (ap. Ath. XIV 31). See on the whole subject Newman's Politics of Aristotle I pp. 359—369 and Nettleship Hellenica pp. 123—130.

φησί ‐‐ Δάμων . III 400 B note

τοίνυν =‘also’: see I 339 D note

φυλακτήριον ‐‐ μουσικῇ . μουσική is at once the vital and the most vulnerable —see next note—part of our State; hence the guard-house must be built in Music. ἐν is quasi-local, as ἐνταῦθά που shews; we shall confuse the metaphor if we suppose (as some have done) that Music is itself the guard-house.

ἡ γοῦν ‐‐ αὕτη . αὕτη is ἡ ἐν μουσικῇ. Madvig's suggestion ταύτῃ should not be accepted; it would make παρανομία ‘lawlessness’ in general, whereas Socrates' reply and Adimantus' next remark shew that only ἡ ἄμουσος παρανομία ( Laws 700 D) is meant. παρανομία is aptly used of heterodoxy in music, thanks to the musical sense of νόμος. Cf. infra 424 E and Shorey in Chicago Studies in Cl. Phil. I p. 222 note 4. The position of αὕτη increases its emphasis.

ἐν παιδιᾶς γε μέρει . Plato is animadverting on the common view that music should be cultivated πρὸς παιδιάν rather than πρὸς παιδείαν. Aristotle allows a threefold use of music—for pastime ( παιδιά), education, and the rational employment of leisure: Pol. Θ 5. 1339^{a} 16 and^{b} 14 ff.

ὑπορρεῖ κτλ. : as a gentle river may become a destructive torrent before its course is ended. The sentence eloquently describes the decay of Athenian music, character, and politics from the simplicity of earlier times, as appears from Laws 700 A—701 D. See also on οἱ πρότερον 425 A. For πολιτείας Hartman would read the singular; but the plural is more forcible. Laws and constitutions are overthrown by the devouring flood. σύν in Plato (as in good Attic generally) is rare; one of its recognised uses is in modal phrases of this kind, especially where (as here and in VI 492 B, VIII 564 C, X 619 B) the style seeks elevation: cf. Lina De praep. usu. Plat. pp. 32 —34 and Mommsen Beiträge z. d. Lehre v. d. Gr. Praep. pp. 376 ff.

ὃ ‐‐ ἐλέγομεν : ‘as we were trying to say at the outset,’ i.e. of this discussion 424 A. No specific reference to an earlier part of the dialogue is intended: at all events II 377 B is not in point. According to Plato παιδιά should—(to borrow a saying of Aristotle's) — παιδεύειν πρὸς τὴν πολιτείαν—educate children in the spirit of their commonwealth: Laws 798 B ff. Conversely, Aristotle reminds us, education is itself the older boys' rattle ( Pol. Θ 6. 1340^{b} 30). It should be noted that παιδιᾶς (cf. παίξειν in 425 A) refers like παιδιᾶς in D above to music; if music is to be a pastime, it must be one which is ἔννομος. In ἐννομωτέρου and παρανόμου there may also be a play on the musical sense of νόμος: cf. 424 D note

τοιούτων : viz. παρανόμων.

)κείνοις : those whom Adimantus in effect described in 424 D. See also next note.

οἱ πρότερον : ‘their predecessors’ (Jowett), i.e. the predecessors of our citizens. The expression betrays the fact that Plato is now censuring the decay of Athenian manners, as of Athenian music and character in 424 D. In ἐξευρίσκουσιν —πάντα Plato speaks as if his regulations were a programme for the reform of his native city. Cf. Krohn Pl. St. pp. 32, 33.

σιγάς τε ‐‐ τοιαῦτα . Cf. Ar. Clouds 961 —1023. Aristophanes mentions the σιγαὶ τῶν νεωτέρων (963), the ὑπαναστάσεις (993), the γονέων θεραπεῖαι (994, 998), and various details of τοῦ σώματος σχηματισμός (973, 983).

κατακλίσεις means literally ‘settings down,’ i.e. causing or permitting others to sit down, as when the Spartans, for example, in the well-known story, made way for the aged stranger at the Panathenaea (Plut. Apophth. Lac. 52. 235 D). Cf. κατακλίναντες in 420 E and II 363 C. The word—which has been curiously misunderstood—is coupled with ὑπανάστασις also in Arist. Eth. Nic. IX 2. 1165^{a} 28. See also Xen. Mem. II 3. 16. After πρέπει supply σιγᾶν out of σιγάς. The older editors read ὡς for ἅς with several deterioris notae MSS.

καὶ ‐‐ γέ with κουράς marks the transition to a new class of particulars: cf. Crit. 47 B , Gorg. 450 D al. Hartman should not have suggested καὶ—τε. It was the Spartans who laid greatest stress upon the points enumerated here: cf. Xen. Rep. Lac. 3. 5, Plut. Cleom. 9. 1 ( κείρεσθαι τὸν μύστακα καὶ προσέχειν τοῖς νόμοις). See also Xen. Cyr. VIII 7. 10.

οὔτε γάρ ‐‐ νομοθετηθέντα . Plato means that specific enactments are powerless either to produce or to maintain civilities and proprieties of this kind. The flowers of civilisation must bloom naturally, or not at all. With the general sentiment of this passage cf. Isocr. Areop. 41 δεῖν δὲ τοὺς ὀρθῶς πολιτευομένους οὐ τὰς στοὰς ἐμπιμπλάναι γραμμάτων ἀλλ’ ἐν ταῖς ψυχαῖς ἔχειν τὸ δίκαιον· οὐ γὰρ τοῖς ψηφίσμασιν ἀλλὰ τοῖς ἤθεσι καλῶς οἰκεῖσθαι τὰς πόλεις.

ὅποι ‐‐ εἶναι : “the bent given by education will determine all that follows” (D. and V.): “wohin einer die Richtung durch die Erziehung bekommen hat, dem auch das folgende entspricht” (Schneider). The sense is satisfactory, nor is the apparent correlation of ὅποι and τοιαῦτα a sufficient reason for impugning the text, as (in common with Dobree and others) I formerly did. ὅπῃ (so Ast with q) would convey the idea of direction more precisely than ὅποι, but as the route is determined by the goal, we may be satisfied. Of the various emendations— ὁποἶ (Heller), ὁποῖος (Stallbaum, who afterwards recanted), ὁποίας (Dobree)— that of Dobree deserves high praise for elegance and point. The meaning would be ‘as is the education from which one starts, so is the sequel’; and for ὁποίας = ἐξ ὁποίας we might compare III 402 A, VII 520 D. I once thought of ὁποία ἄν τις ὁρμὴ ᾖ, but am now content with the text as it stands.

τελευτῶν ‐‐ ἀγαθόν . Cf. 424 A ἐάνπερ ἅπαξ ὁρμήσῃ εὖ, ἔρχεται ὥσπερ κύκλος αὐξανομένη.

οὐκ ἂν ἔτι . On ἔτι see III 412 B note

τί δέ; κτλ. ‘Once more: in heaven's name, said I, these markettroubles about contracts which the different classes of citizens make with one another in the market-place etc.—shall we condescend to make laws about any of them?’ I have placed a mark of interrogation after τί δέ (quid vero?): cf. 422 D note and 426 A. This increases the emphasis on ὦ πρὸς θεῶν: cf. I 332 C ὦ πρὸς Διός, ἦν δ’ ἐγώ, εἰ οὖν κτλ. Herwerden puts the pause after τάδε, where it is less suitable; others wrongly omit the word. ταδε (see cr. n.) cannot well be dispensed with: it means ‘these familiar’: cf. III 403 E, and for the omission in A Introd. § 5. Herwerden also cuts out ἀγοραῖα on account of κατ’ ἀγοράν, but the reduplication is quite in Plato's way. The postponement of ἅ throws emphasis on κατ’ ἀγοράν, and thereby helps to contrast ἀγοραῖα ξυμβόλαια with χειροτεχνικά etc.: cf. III 390 B. It is natural to see in this sentence a reference to the judicial and mercantile arrangements of Athens and her empire: see 424 D note

χειροτεχνικῶν κτλ. χειροτεχνικὰ ξυμβόλαια are contracts with builders and the like ( Laws 920 D).

δικῶν λήξεως means simply ‘the bringing of lawsuits’: originally ‘obtaining (by lot) one's rights,’ hence ‘obtaining leave to claim one's rights’ (Meier and Schömann Att. Process pp. 790—794). The reading λήξεις (see cr. n.) cannot be defended.

θέσεις : not ‘the imposition of taxes’ (L. and S.), but ‘the payments,’ as πράξεις is ‘the exactions.’

τὸ παράπαν means ‘in general,’ ‘generally.’ τὸ πάμπαν (see cr. n.) is never (I believe) so used, not even in Tim. 64 E cited by Baiter. Regulations on nearly all the points here specified are laid down in the Laws: on ξυμβόλαια 913 A ff., 920 D ff., on λοιδορία 934 E ff., on αἰκεία (unprovoked assault) 879 B ff., on δικῶν λῆξις 949 C, on δικαστῶν κατάστασις 767 A ff., 956 B ff., on ἀστυνόμοι and ἀγορανόμοι 763 C ff. There is no taxation in the city of the Laws (847 B).

καλοῖς κἀγαθοῖς . Cf. VI 489 E note

ὅσα δεῖ νομοθετήσασθαι shews that Plato does not wish to leave all these matters undefined by legislation; but the legislation is to come from the guardians he has educated. One reason is that laws on matters of this kind can never be final: cf. Laws 769 D. If the guardians are true to the spirit of Plato's commonwealth, they will easily frame such minor regulations, and re-adjust them—should it prove necessary—from time to time. The effort to obtain finality ( οἰόμενοι ἐπιλήψεσθαι τοῦ βελτίστου) in such matters is foredoomed to failure (cf. 426 E), and no one makes it, until he has forgotten the real foundation of a nation's greatness, and lost his sense of the proportion of things. This is Plato's meaning.

πλήν γε κτλ. If the text is sound we must take πλήν γε as πλήν γε ὅτι (which H. Wolf was wishful to restore) and καί before ἀεὶ ἐλπίζοντες as= idque (with Stallbaum), unless we supply διάγουσι or the like by a sort of zeugma after ἐλπίζοντες. As regards καὶ ἀεὶ ἐλπίζοντες, J. and C. hold that the participle is resumed from ἰατρευόμενοι; but the effect of this interpretation is very harsh, because ἰατρευόμενοι goes so closely with οὐδὲν περαίνουσι as almost to form a single expression. It is not ‘they make no advance, submitting to a cure and always hoping,’ but ‘they make no advance under treatment.’ The troublesome καί before ἐλπίζοντες is omitted by some inferior MSS, is dotted in q, and apparently erased in Ξ. I once conjectured ποιοῦντες, comparing Critias 109 B πλὴν οὐ—βιαζόμενοι, but it is perhaps safer to acquiesce in the MS reading. Dümmler ( Chron. Beitr. pp. 9—11) believes that Isocrates Antid. 62 expressly alludes to this passage. Isocrates at all events censures τοὺς ἐπιπλήττοντας τοῖς νῦν ἁμαρτανομένοις in words that might easily refer to Plato. See also on 426 C.

αὐτῶν ‐‐ μεθύων . On the plural passing into the singular see I 347 A note

τοιοῦτον . Cf. III 388 D note

προαγορεύουσι κτλ. Athens is plainly in Plato's mind. The Athenians carefully guarded their constitution by means of the γραφὴ παρανόμων and the εἰσαγγελία (see Gilbert's Gk. Const. Ant. E.T. pp. 299, 304 ff.); but nowhere were ψηφίσματα so common, and in these the demagogue found a wide field for exercising the arts of flattery and insinuation. Cf. Gilbert Beiträge zur innern Gesch. Athens pp. 73—93. With ἀποθανουμένους ὅς cf. III 411 C note, VIII 566 D ( πάντας ᾧ ἂν περιτυγχάνῃ).

ὃς δ’ ἂν σφᾶς κτλ. Dümmler (l.c.) takes this to be Isocrates, who is also—so he thinks—satirised in the similar passage VI 493 A ff., and elsewhere. If so, σοφὸς τὰ μεγάλα, οἴονται τῇ ἀληθείᾳ πολιτικοὶ εἶναι, and ἢ οἴει—περὶ αὑτοῦ (D, E) are sufficiently true and scathing. We must however observe that Plato is describing a type, and the type is that of the demagogue rather than the merely academic and sophistical rhetorician, as appears from δεινὸς ᾖ ἀποπληροῦν and 426 E. These two types are cast in similar moulds; and Dümmler may be right in supposing that Plato thought of Isocrates as he wrote this satire, and pointed his shafts accordingly. If so, they hit the mark, and rankled, as it was natural they should. Isocrates apparently attempts a reply in his Antidosis (Dümmler l.c. p. 9).

οὗτος ἄρα ‐‐ ἔσται . To insert ὡς after οὗτος (as Richards proposes) would spoil the effect, and be grammatically awkward. Plato wishes to suggest the language of a proclamation ‘he shall be a good man and true,’ etc. ἄρα is enough (as Hartman notes) to mark the indirect: cf. II 358 C note

ἀνδρείας ‐‐ εὐχερείας : ‘courage and complaisance.’ εὐχέρεια is not ‘dexterity (L. and S., with the English translators), a meaning which the word never bears in Plato; but ‘facilitas,’ ‘humanitas,’ kind, obliging behaviour. “Herzhaftigkeit und Gutmüthigkeit,” Schneider, rightly.

τετράπηχυς : ‘a six-footer.’ Dümmler (l.c.) questions this word, without saying why. It is more appropriate than a word expressing greater height; especially if any personal allusion is intended. Isocrates was not an intellectual giant, nor would even his applauding contemporaries (I think) have called him so.

οὐκ αὖ ‐‐ τοῦτό γε : sc. οἴομαι. The point of αὖ is that Adimantus returned an affirmative answer last time (426 D). οὐκ ἄν, which is generally read, has not sufficient authority, and is difficult to justify. For οὐκ αὖ cf. III 393 D and infra 442 A.

πάντων χαριέστατοι . To this perhaps Isocrates replies in Antid. 62 χαριέντως μὲν εἰρῆσθαι ταῦτα φήσουσι, τὸ γὰρ εὖ φθονήσουσιν εἰπεῖν (Dümmler l.c.).

νομοθετοῦντες κτλ. It improves the rhetorical effect to treat all the participles as coordinate, instead of making the first two dependent on the third, or the third subordinate to them. For this reason I have placed a comma after ἐπανορθοῦντες.

οὔτ) ‐‐ ᾤμην ἄν : ‘I should not have thought so’ were it not for these great authorities. Jowett misses the irony by neglecting the tense (‘I conceive that the true legislator will not trouble himself,’ etc.). τὸν ἀληθινὸν νομοθέτην and κἂν ὁστισοῦν εὕροι would strike home, if Isocrates is meant.

ἀνωφελῆ ‐‐ ἐπιτηδευμάτων . For ἀνωφελῆ Ξ has ἀνωφελές, an obvious ‘correction.’ The plural, as Schneider observes, is supported by τὰ μὲν αὐτῶν (where αὐτῶν is also neuter). ὅτι after τὰ δέ has been called in question by Stallbaum and Hartman. Taken strictly, it must depend on a verbal notion supplied out of πραγματεύεσθαι (Stallbaum) or κἂν ὁστισοῦν εὕροι; but in a halfadverbial phrase like τὰ δέ, we should not pry too closely into the grammatical construction. The effect is exactly like the English ‘because some of them, etc., in other cases, because,’ etc.

τί οὖν κτλ. With this section of the Republic we should compare V 461 E, 469 A, VII 540 C, and Laws 738 B ff. Plato would fain be no iconoclast: his object is to purify, rather than to abolish, the old religion. He tries, in short, to put new wine into old bottles. In particular, when he makes Apollo preside at the foundation of his city ( οἰκίζοντές τε πόλιν οὐδενὶ ἄλλῳ πεισόμεθα), he is acting in accordance with the universal custom of the Greeks, who consulted the oracle at Delphi before planting colonies, and revered him as the universal ἀρχηγέτης and οἰκιστής (Preller Gr. Myth. p. 269). It is equally in harmony with Hellenic, and especially Athenian, usage to refer all matters of public worship to Apollo: see on 427 C. Delphi was the abiding centre of Greek religious and political unity; and it is therefore right that a Greek city (V 470 E), one of whose objects is to promote unity and comity among Greeks (ib. 469 B ff.), should attach itself to Apollo.

τῷ μέντοι Ἀπόλλωνι κτλ. Cf. Mem. I 3. 1 (of Socrates) φανερὸς ἦν καὶ ποιῶν καὶ λέγων, ᾗπερ ἡ Πυθία ἀποκρίνεται τοῖς ἐρωτῶσι, πῶς δεῖ ποιεῖν ἢ περὶ θυσίας ἢ περὶ προγόνων θεραπείας ἢ περὶ ἄλλου τινὸς τῶν τοιούτων. The answer of the priestess was ‘Serve the gods νόμῳ πόλεως’ (l.c. and IV 3. 16). The spirit in which we worship matters, rather than whom or how we worship. So large and tolerant a sentiment is worthy of the Delphic priesthood and of Plato.

τελευτησάντων τε . See cr. n. Asyndeton is indefensible here. We must either with all the editors (except J. and C.) read τε, or add καί after θεραπεῖαι.

ἐξηγητῇ ‐‐ πατρίῳ . πατρῴῳ instead of πατρίῳ is called for by Ast on slight MS authority. Ἀπόλλων was ancestor of the Ionians, being father of Ion ( Euthyd. 302 D ), and was worshipped by them as Ἀπόλλων πατρῷος (Preller Gr. Myth. p. 272). But (as Schneider observes) “Socrates hic non magis quam alibi in his libris tanquam Atheniensis loquitur, sed tanquam Graecus. Graecis autem omnibus πάτριος, hoc est, a maioribus traditus harum rerum arbiter et interpres erat Delphicus Apollo.” An allusion to the special connexion of Ionians with Apollo would be out of place, particularly as πᾶσιν ἀνθρώποις follows. In Athens the ἐξηγηταί formed a college of three members, charged with religious duties. According to Schöll (in Hermes VI pp. 36 ff.) the members were partly chosen by Apollo in his capacity of πάτριος ἐξηγητής; apparently the Athenians chose nine, out of whom three were selected—one from each triad —by the representatives of the god: whence their designation πυθόχρηστοι. It is on this model that Plato perhaps frames his regulations in Laws 759 D.

πᾶσιν ἀνθρώποις . Delphi is then a religious centre, not for Greeks only, but for all mankind. It was certainly the nearest approach to such a centre that antiquity provided, for it commanded the homage of barbarians as well as Greeks. See Middleton Journ. of Hell. Studies IX p. 308. Middleton cites Livy XXXVIII 48. 2 “commune humani generis oraculum,” Cicero pro Font. 30 “oraculum orbis terrae,” and gives examples of the offerings paid by foreigners at Apollo's shrine. Even now, perhaps, Plato would deny that the oracle is dumb, though—true to its own principle of worshipping νόμῳ πόλεως—it speaks through other voices, and of other gods. See also on V 470 C.

ἐν μέσῳ ‐‐ ἐξηγεῖται . Cf. Eur. Ion 5 , 6 ὀμφαλὸν | μέσον καθίζων Φοῖβος ὑμνῳδεῖ βροτοῖς. The ὀμφαλός was “a conical mass of ‘white marble or stone’” (Paus. X 16) in the sanctuary of Apollo at Delphi, “said to mark the centre of the earth.” Two gold eagles stood at its sides, representing the eagles which, according to the legend, met there, having been despatched simultaneously by Zeus from the extreme East and West of the world (Strabo IX 3. 6). The ὀμφαλός is frequently represented as the seat of Apollo ( ἐπὶ τοῦ ὀμφαλοῦ καθήμενος), “especially upon coins, when he is represented in the character of the giver of oracles”: see for example ImhoofBlumner and P. Gardner in J. H. S. VIII p. 18, and Plate LXXIV vii. Middleton, on whose article “The Temple of Apollo at Delphi” (cited above) this note is chiefly based, thinks “the word ὀμφαλός was probably derived from ὀμφή, a voice, because the divine voice was heard there.” If this is true, the legends associating the shrine with the ‘navel’ or centre of the earth may be due to popular etymology. ὀμφαλός, ‘navel,’ is an Indo-Germanic word (Brugmann Grundriss II p. 187). Herwerden's excision of the words ἐν μέσῳ betrays ignorance of what the ὀμφαλός really was. See also Frazer on Paus. l.c.

αὐτός τε καὶ ‐‐ παρακάλει . For the idiom cf. (with Schneider) Phaedr. 253 B μιμούμενοι αὐτοί τε καὶ τὰ παιδικὰ πείθοντες.

ποῦ ἡ ἀδικία . If our city is τελέως ἀγαθή (427 E), it is useless to look for ἀδικία in it. On this difficulty see II 369 A note

πότερον . Herwerden's ποτέραν is quite unnecessary, as Hartman shews; cf. 428 A, 433 D, 434 C, 445 B, V 449 D.

ἐάν τε λανθάνῃ κτλ. recalls II 367 E.

ὡς οὐχ ὅσιον ‐‐ τρόρῳ : II 368 B, C.

σοφή ‐‐ δικαία . This is apparently the earliest passage in Greek literature where the doctrine of four cardinal virtues (if by cardinal virtues we mean those which make up the sum of perfect goodness) is expressly enunciated. The doctrine may of course be Pythagorean, but evidence is wanting, and it is doubtful whether Pindar's τέσσαρες ἀρεταί Nem. III 74 are to be interpreted as the cardinal virtues: see Bury ad loc. The nearest approach to the doctrine before Plato is in Xen. Mem. III 9. 1 —5 (as Krohn has pointed out Pl. St. p. 372), with which compare IV 6. 1—12, where Justice, Wisdom, and Courage are named, as well as other virtues, including εὐσέβεια. Cf. also Aesch. Sept. 610 σώφρων δίκαιος ἀγαθὸς εὐσεβὴς ἀνήρ. From other passages in Plato, none of which are so precise and technical as this, it would seem that ὁσιότης made a good fight for a fifth place: Prot. 329 C , Lach. 199 D , Men. 78 D , Gorg. 507 B . In Phaed. 69 C and Laws 631 C σωφροσύνη, δικαιοσύνη, ἀνδρεία and φρόνησις (not σοφία) are named together, without ὁσιότης, which in the Euthyphro (12 D ff.) is a subdivision of δικαιοσύνη. From Adimantus' ready assent (cf. V 476 A note), we may reasonably infer that the doctrine of four cardinal virtues was already a familiar tenet of the Platonic school. Schleiermacher thinks it may have been taken over “aus dem allgemeinen Gebrauch” ( Einleitung p. 26). There is however no evidence to shew that these four virtues and no others were regarded as the essential elements of a perfect character before Plato. If the theory was originated by Plato himself, it is possible enough that in restricting the number to four, Plato was not uninfluenced by the sacred character of the number four in Pythagoreanism, just as Aristotle has been supposed to have limited his categories to ten on similar grounds. An interesting conjecture is suggested by the remarks of Schleiermacher (l.c. p. 21). Our city is ex hypothesi perfectly virtuous. Its constituent elements are Rulers, Auxiliaries, Farmers and Artisans. Now the virtues which are exhibited in the lives and mutual relationship of these classes are, as Plato holds, Wisdom, Courage, Temperance, and Justice. Consequently these virtues are the component factors of moral perfection; in other words they are the cardinal virtues. We may admit that there is no petitio principii in such a method of investigation, which is, in fact, akin to the perfectly legitimate method described in Men. 86 E : cf. also V 458 A. If this suggestion is correct, the doctrine of four cardinal virtues will be directly descended from the arrangements of Plato's ideal city. But it is clear from what Plato himself says, both here and in 429 A, 430 D, 432 B, 433 B f., that the doctrine is already an accepted part of his ethical system, and not merely a provisional hypothesis which is intended to be confirmed by what follows. For the relative value and importance of the four cardinal virtues in Plato's way of thinking see Laws 630 D ff.

οὐκοῦν ‐‐ ηὑρημένον . Essentially the same method is used by Aristotle to reach his conclusion that virtue is a ἕξις ( Eth. Nic. II 4). Cf. also (with J. and C.) Lys. 216 D , E. Jowett observes that the true function of “this half-logical, halfmathematical method of residues” is in dealing with “abstract quantity” and “the laws of Nature.” It is undeniable that this method is much more likely to lead us astray in ethics than in mathematics or the natural sciences, owing to the nature of the subject; but it is valid if our analysis of the phenomena is exhaustive and exact. A similar method was frequently employed in the Eleatic school: see II 380 D note Plato not unfrequently extends the methods of mathematical reasoning beyond what we should consider their proper sphere: the whole of the preliminary studies, for example, in Book VII are to be pursued according to the methods of pure mathematics. See on VII 528 E ff. and the Appendix to Book VII “On the propaedeutic studies of the Republic.”

ὥσπερ τοίνυν ‐‐ αὐτῶν . For the logically superfluous (though welcome) αὐτῶν cf. II 375 E, infra 439 B, VIII 558 A, and Heindorf on Gorg. 482 D . Theaet. 155 E is a much harsher example, and has often been emended. The apodosis to the ὥσπερ clause is contained in οὐκοῦν— ζητητέον.

οὐκ ἄλλο ἔτι ἦν . On ἔτι (i.e. “after the other three were found” J. and C.) see III 412 B note

αὐτῷ . A corrector in q wrote αὐτῇ, which Schleiermacher preferred. Hartman suggests αὐτοῖς. αὐτῷ is, however, not the city, but simply ‘the matter,’ ‘the subject under discussion’; an idiomatic usage for which cf. I 339 E note For the neuter κατάδηλον cf. 427 D note Hartman's κατάδηλος is unnecessary.

σοφία as here described means φρόνησις—so it is called in 433 B, C— in its application to politics, not metaphysical knowledge of the Idea of Good. It deliberates for the good of the whole city (428 D), but the good is not yet elevated to the rank of an Idea. This point has been rightly emphasized by Krohn ( Pl. St. pp. 40, 362), who points out the essentially Socratic character of this virtue, comparing Xen. Mem. I 2. 64 and IV 1. 2 (a sentiment of which Books II—IV of the Republic are an amplification and exposition in detail). See also Prot. 352 B and Laws III 689 B. Commentators before Krohn (Steinhart for example Einleit. p. 185, and Susemihl Gen. Entw. II p. 153) did not sufficiently grasp the almost exclusively political character of σοφία here, although it is expressly dwelt upon by Plato throughout, and particularly in 429 A. I say ‘almost,’ because here, as elsewhere, Plato, as his manner is, contrives to drop some hints preparing us for a still higher conception of the virtue of the guardians. See on 429 C and 442 C.

ἄτοπον : because it is its smallest section which makes the whole city wise (428 E).

εὔβουλος . εὐβουλία was primarily a political virtue: see on I 348 D.

βουλευομένην . Heindorf's emendation (see cr. n.), which is accepted by Ast, Stallbaum, Baiter and Hartman, appears to me certain for these reasons. First, in τὴν ὑπὲρ τῶν ἐκ τοῦ χαλκοῦ ἤ τινα ἄλλην τῶν τοιούτων below we must understand ἐπιστήμην βουλευομένην, so that βουλευομένην and not βουλευομένη must have been written before. Secondly, if we read βουλευομένη, we must write (with Hermann etc. and a few inferior MSS) ᾗ for ἥ before οὐχ ὑπὲρ τῶν below. Schneider retains βουλευομένη, but understands βουλευομένην before ἐπιστήμην— an indefensible construction, which Laws 807 C (to which he appeals in Addit. p. 31) in no way justifies.

τὴν ὑπὲρ ‐‐ τοιούτων . For the carrying on of the preposition (here διά) cf. (with Schneider) Phaed. 64 D ἐσπουδακέναι περὶ τὰς ἡδονὰς καλουμένας τὰς τοιάσδε, οἷον σιτίων κτλ. Ἥκιστά γε κτλ. Τί δέ; τὰς τῶν ἀφροδισίων;

ὅντιν) ἂν ‐‐ ὁμιλοῖ. ἄν cannot, I think, be dispensed with here. It is better to insert it after ὅντινα than (with Baiter) after ἄριστα, for (as Schneider shews by many examples) ἄν likes to attach itself to the relative in sentences of this kind. The political wisdom here described is akin to the βασιλικὴ τέχνη of Euthyd. 291 C ff. and elsewhere, as well as to Aristotle's view of πολιτική as the architectonic art ( Eth. Nic. I 1. 1094^{b} 27 with Stewart's note). It knows what is good and evil, and legislates for the other arts, but the good which it knows is a political and moral conception, not (as yet) the metaphysical Idea of Book VI.

νῦν δή . III 414 B ( φύλακας παντελεῖς).

τί τὴν πόλιν προσαγορεύεις . It should be noted that ‘wise’ (to confine ourselves for the present to the virtue of wisdom) is used (1) of the rulers in the State and the λογιστικόν in man, (2) of the city and the individual as wholes: cf. 441 D ff. Which of these two meanings is intended to be original and primary? This subject is admirably discussed by Hirzel Hermes VIII pp. 379 ff., who shews that the wisdom of the rulers and the λογιστικόν is the fundamental one: cf. Arist. Top. V 8. 138^{b} 1 ff., where τὸ πρῶτον φρόνιμον is said to be ἴδιον λογιστικοῦ. The same, mutatis mutandis, holds good of Courage; and also, though with a difference, of Temperance and Justice. In calling the whole city wise because the rulers are wise, Plato is influenced by its analogy with the individual man, whom we readily and easily call wise, although strictly speaking he is wise only by reason of the λογιστικόν within him. Comparing 443 C ff., we observe that the city is wise because its rulers are wise, and its rulers are wise because their λογιστικόν is wise. In other words the wisdom of the λογιστικόν is the unit out of which the wisdom of the whole city is constructed. See on 443 B ff.

πότερον οὖν . See cr. n. We have still to explain τι ἄτοπον in 428 B, for Adimantus' τί has not yet been answered. For this reason οὖν after πότερον is welcome, if not (as Schneider thinks) indispensable.

πολὺ ‐‐ χαλκέας . Cf. II 379 C note

ὅλη σοφὴ κτλ. The subject is πόλις κατὰ φύσιν οἰκισθεῖσα, ‘a city founded in accordance with Nature.’ On κατὰ φύσιν see II 370 A note

ἣν μόνην ‐‐ σοφίαν καλεῖσθαι . Pfleiderer ( Zur Lösung d. Pl. Frage pp. 46 ff.) compares Symp. 209 A ff. πολὺ δὲ μεγίστη—καὶ καλλίστη τῆς φρονήσεως ἡ περὶ τὰς τῶν πόλεών τε καὶ οἰκήσεων διακοσμήσεις, ᾗ δὴ ὄνομά ἐστι σωφροσύνη τε καὶ δικαιοσύνη. The difference in phraseology does not obscure the essential kinship of the two passages.

ὁ νομοθέτης ‐‐ παιδείᾳ . The δόξα is then prescribed by the legislator (i.e. in Plato's city, by Plato, cf. ὑπὸ νόμου below, νομίμου in 430 B, and σὺ ὁ νομοθέτης in VI 497 D), not by the rulers from time to time. It is important to notice this point, because it shews that the rulers are not here, as in a certain sense they are in VI—VII, in the position of the original legislator: see VI 497 C note Cf. however III 414 A note and infra 442 C note

ποίαν δὴ σωτηρίαν ; ποίαν expresses incredulity and wonder, which δή saves from falling into contempt. See I 330 A note On the definition of courage given here see 430 C note

γεγονυίας . Cf. γεγονυῖαν in 430 B.

αὐτῆς σωτηρίαν . See cr. n. αὐτήν of the MSS must mean either (1) the σωτηρία or (2) ἀνδρεία (so Hartman). In either case the αὐτήν which follows has a different antecedent viz. τὴν περὶ τῶν δεινῶν δόξαν, so that the sentence becomes both awkward and obscure. Moreover, in whichever way we understand αὐτήν, the MSS leave us with three accusatives ( αὐτήν, σωτηρίαν and the clause introduced by τό), the precise relationship of which is far from clear. Various suggestions have been made to escape these difficulties. Instead of αὐτήν Jackson suggests αὖ τήν ( J. Ph. IV p. 148); while Stallbaum and others read τῷ (eo quod) for τό, before which Hartman for his part wishes to insert διά. Hermann and Baiter cut the knot by expunging both αὐτήν and σωτηρίαν. Jackson's remedy is the simplest, but αὖ creates a difficulty. The new point in the explanation which he supposes it to mark is, I think, emphasized too much by αὖ; nor indeed is it quite easy to separate αὖ from ἔλεγον. I believe Plato wrote αὐτῆς. The words διὰ παντὸς αὐτῆς σωτηρίαν recall and correspond exactly to ἣ διὰ παντὸς σώσει τὴν περὶ τῶν δεινῶν δόξαν, and to 430 B σωτηρίαν διὰ παντὸς δόξης κτλ., and the meaning is ‘by preserving it perpetually I meant preserving it throughout when one is in pains and in pleasures’ etc. Grammatically, the infinitives are the direct object of ἔλεγον (‘I called’), and διὰ παντὸς αὐτῆς σωτηρίαν is its secondary object. The presence of αὐτῆς σωτηρίαν is necessary to correspond to διασῴζεσθαι αὐτήν, but διὰ παντὸς takes the first place, because it is the phrase requiring elucidation. The corruption of αὐτῆς to αὐτήν is of a piece with that of γεγονυίας to γεγονυῖαν (see cr. n.) and its all but inevitable consequence. The correction printed above is accepted by a reviewer of my Text of the Republic in Hermathena XXIV p. 252.

λύπαις ‐‐ φόβοις . III 412 E ff.

ἁλουργά =‘purple’: see Tim. 68 B , with Archer-Hind's note. Herwerden cuts out ὥστ’ εἶναι, but without these words the wool which we are dyeing would be purple, whereas it is white, and we are making it purple. See on ἐάν τε καὶ ταῦτα in E.

πρῶτον μὲν κτλ. As far as concerns the language and grammatical construction of this passage it is clear that the object of ἐκλέγονται should be the same as that of προπαρασκευάζουσιν, θεραπεύσαντες, and βάπτουσι, and identical with the subject of δέζεται. Now the object of βάπτουσι is the wool selected to be dyed; it is therefore the wool which is subjected to προπαρασκευή, and consequently white substances of wool are meant by μίαν φύσιν τὴν τῶν λευκῶν (so also Blümner Technologie etc. I pp. 221 ff.). That this interpretation is right, appears also from the application of the simile. The guardians are the white woollen substances specially selected (note ἐξελεγόμεθα 429 E), their education is the προπαρασκευή; and the δόξα περὶ δεινῶν κτλ. is the dye. This is expressly pointed out in 429 E—430 A. τοσούτων is strictly in point, for woollen substances may be of any colour, since they may have been already dyed. Plato informs us that dyers selected white woollen substances when they wished to impart a lasting purple hue. Cf. Tim. 50 D , E. The προπαρασκευή included the process called στύψις, i.e. steeping the wool in an astringent solution ( πρόστυμμα) to make it take the dye better (Arist. de Col. 4. 794^{a} 29 and Probl. XXII 11. 931^{a} 13 ff. προβρέχουσιν ἐν τοῖς στρυφνοῖς τῷ διεργασθὲν μᾶλλον δέχεσθαι τὴν βαφήν: cf. also Theoph. de Odor. 17 ὑποστύφουσι γὰρ πᾶν εἰς τὸ δέξασθαι μᾶλλον τὴν ὀσμὴν ὥσπερ τὰ ἔρια εἰς τὴν βαφήν). Aristotle uses a metaphor from dyeing in a similar way in Eth. Nic. II 2. 1105^{a} 3. Cf. also Cicero Hortens. Fr. 62 ed. Nobbe “ut ei qui combibi purpuream volunt, sufficiunt prius lanam medicamentis quibusdam, sic litteris talibusque doctrinis ante excoli animos et ad sapientiam concipiendam imbui et praeparari decet,” and see on the whole subject Blümner l. c. I pp. 221 ff., 238 ff.

θεραπεύσαντες . If the text is sound, we must suppose either that two processes of preparation are alluded to, viz. θεραπεία and προπαρασκευή; or else that θεραπεύσαντες is used for θεραπεύοντες. The first alternative is inadmissible: for προθεραπεύσας in E shews that the θεραπεία and προπαρασκευή are identical. As for the second, Schneider remarks “aoristum ipsum pro praesenti positum vix credo.” There are some instances in which “an aorist participle denoting that in which the action of a verb of past time consists may express time coincident with that of the verb, when the actions of the verb and the participle are practically one” (Goodwin MT. p. 52: cf. Kühner Gr. Gr. II pp. 161 ff.), but as προπαρασκευάζουσιν is a verb of present or universal time, Goodwin's rule is inapplicable here. Hartman ejects the participle, and Schneider is anxious to read θεραπεύοντες. In my edition of the Text, I had recourse to transposition, and placed θεραπεύσαντες before οὕτω δή (‘and they do not dip the wool till they have finished dressing it’). It is, however, safer to adhere to the MSS and regard θεραπεύσαντες as one of those ‘timeless aorists,’ of which many examples are quoted by F. Carter in Cl. Rev. V pp. 4 ff. The MS reading is supported not only by Stobaeus ( Flor. 43. 97), but also (as Jackson has pointed out to me) by Theo Smyrnaeus de utilit. math. p. 13 ed. Hiller.

τὸ ἄνθος : the colour, as appears from Arist. de Col. l. c. 794^{a} 34 et al. Though it is used of purple here, it was not confined to purple: see on VIII 557 C.

δευσοποιὸν ‐‐ βαφέν . δευσοποιόν· ἔμμονον καὶ δυσαπόπλυτον (Timaeus Lex. s. v. δευσοποιόν, where Ruhnken illustrates the word very fully). The point of course is that such προπαρασκευή rendered the colour proof against washing. δευσοποιός, ἀνέκπλυτος, and μόνιμος were constantly used in connexion with dyeing: see Blümner l. c. I p. 221 notes The words τὸ βαφέν are bracketed by Herwerden; but ὃ ἄν is not ‘quod,’ but ‘si quid’ (Schneider).

ῥυμμάτων . ῥύμματα is the generic word for detergents of any kind (Blümner Privatalt. p. 214 note 1): cf. παντὸς ἄλλου ῥύμματος 430 B.

ἐάν τε καὶ ταῦτα . ταῦτα is τὰ λευκά i.e. white substances: cf. τῶν λευκῶν in D above. Even white wool, unless specially prepared, will not retain the dye when it is dipped: much less other colours. This is the force of καί in καὶ ταῦτα. The words ἄλλα χρώματα refer to the colour of the wool which is dipped, not to the colour of the dye, as Herwerden supposes when he calls for τοῦτο: cf. note on ὥστ’ εἶναι in D above.

ἔκπλυτα καὶ γελοῖα : a sort of hendiadys: cf. VIII 558 A θεσπεσία καὶ ἡδεῖα. Stallbaum's suggestion ἀγελαῖα for γελοῖα is itself γελοιότερον. For τοιοῦτον cf. III 388 D note

ἐκπλύναι . Not ἐκπλύνοι (with Herwerden); for the action of ἐκπλύναι is more rapid than that of γίγνοιτο.

χαλεστραίου κτλ. χαλεστραῖον λίτρον (or νίτρον, but λίτρον is the Attic form) came from Χαλέστρα, a lake and city in Macedonia. λίτρον is supposed to be ‘native carbonate of soda’: see Blaydes on Ar. Frogs 712 . The spelling χαλεστραῖον is established (as against χαλαστραῖον in Tim. Lex. s. v. and the Scholiast) by Hdt. VII 123 ( χαλέστρα) and other authorities quoted by Schneider. κονία as appears from ψευδολίτρου κονίας in Ar. l. c. was a preparation of λίτρον, whence Plato couples them here. See on the subject generally Dict. Ant. I p. 881.

παντὸς ἄλλου ῥύμματος is cancelled by Badham and others. It is difficult however not to feel that something is wanted to balance χαλεστραίου and κονίας, especially as these are two specific detergents of the same class. Further, without παντὸς ἄλλου ῥύμματος Plato would probably have written καὶ λύπη κτλ. The sentence as it stands rings Platonic; nor was παντὸς ἄλλου ῥύμματος at all likely to be added by a scribe. The words were also in the text used by Stobaeus and Theo Smyrnaeus: see Flor. 43. 97 and de utilit. math. p. 14. I suggest the following interpretation. The action of pleasure differs from that of pain, fear, and desire, in being more gentle, and less violent ( βίαιος). Pleasure in short relaxes ( χαλᾷ) while pain (of which fear and desire as such are both varieties) contracts: cf. III 411 A on the effect of γλυκεῖαι ἁρμονίαι, Tim. 66 C and Stallbaum on Phil. 46 D . Now χαλεστραίου suggests χαλᾶν, and it is probably for this reason that Plato compares pleasure to it. Such a play on words is quite in Plato's manner: cf. Prot. 361 D . If we suppose that other ῥύμματα were harder, and less agreeable in their action, the point of comparing pain etc. with ‘every other detergent’ will appear.

τὴν ὀρθὴν δόξαν has been questioned, on the ground that beasts cannot have ὀρθὴ δόξα. It was no doubt a feeling of this kind which gave birth to the reading αὐτήν for ὀρθήν in some inferior MSS. Herwerden employs his favourite remedy of excision; and other equally unsatisfactory remedies will be found in Hartman. The text is quite sound. True opinion is in Plato the basis of action done in ignorance of what is right but in obedience to an authority which knows. A dog and a slave act from true opinion as often as they obey a master who orders them to do what is right. So also (among others) Rettig ( Proleg. p. 109) and Krohn ( Pl. St. p. 42) rightly understand the passage. Cf. note on πολιτικήν in C below.

οὔτε ‐‐ τε =‘not only not—but also’ lays stress on the second clause: cf. 427 C, VIII 566 D, E, IX 587 A al.

μόνιμον . See cr. n. The reading of some of Stobaeus' MSS ( Flor. 43. 97) (which Dobree and others approved) appears to me almost certainly right, although it has been adopted by no recent editor. νόμιμον, as Rettig shews ( Proleg. p. 110), must be used in precisely the same sense as in δόξης ὀρθῆς τε καὶ νομίμου just before. If so, Plato flatly (except for the οὔτε πάνυ) contradicts himself. For the only reason why a δόξα is ὀρθή is that it is νόμιμος ‘in accordance with the law’: nor is it possible for even a dog to possess an ὀρθὴ δόξα which is not νόμιμος. In obeying a just command, the δόξα of a dog is therefore not οὐ πάνυ νόμιμος, but wholly νόμιμος. On the other hand μόνιμον is not only appropriate but necessary in what is practically a résumé of Socrates' whole account of courage ( δοκεῖς γάρ μοι —καλεῖν). The only difference between the ὀρθὴ δόξα of a guardian and a dog lies in this, that the former has received παιδεία, while the latter has not. And it is precisely this difference which makes the guardian's δόξα lasting, as the whole of the simile from dyeing was intended to shew ( ἵνα δευσοποιὸς κτλ. 430 A). Finally, the soldier's ὀρθὴ δόξα has just been defined (in 430 B) as σωτηρίαν διὰ παντός κτλ. To διὰ παντός the words οὐ πάνυ μόνιμον are the necessary contrast: the δόξα is in both cases ὀρθή τε καὶ νόμιμος, only you can depend on the guardian always, ἔν τε λύπαις καὶ ἐν ἡδοναῖς καὶ ἐν ἐπιθυμίαις καὶ ἐν φόβοις (429 D), but not always on your dog and slave. Cf. Men. 97 E f.

ἄλλο τε ‐‐ ἀνδρείαν . With the sentiment cf. Lach. 197 A ff., where however it is because they are destitute of knowledge that courage is denied to the lower animals. Isocrates Antid. 211 speaks of dogs etc. as brave.

πολιτικήν γε ‐‐ δίιμεν . In this passage πολιτικὴν ἀνδρείαν means, I think, primarily the virtue of a πόλις as opposed to that of an ἰδιώτης: cf. 442 D πόλεώς τε καὶ ἰδιώτου. Our πόλις is brave because her soldiers are brave (429 B); so that in describing the courage of the soldiers we have really and truly been describing that of our city. But the ἀνδρεία with which we are now concerned is πολιτική in another, and more important sense, being based on ‘correct opinion’ (cf. Phaed. 82 A , B), i.e. in this instance on opinion which is in conformity with the law of the πόλις (cf. Aristotle's πολιτικὴ ἀνδρεία Eth. Nic. III 11. 1116^{a} 16 ff.), and not on ‘knowledge,’ like the scientific or philosophic virtue to which we are introduced in Books VI and VII. In this Platonic connotation of the term, δημοτική or πολιτικὴ ἀνδρεία is inferior both to the courage which rests upon knowledge in the Socratic sense ( Lach. 195 A , 196 E ff., Prot. 349 D ff.) and to that which rests on knowledge of the Idea of the Good (cf. VI 506 A), although it is nevertheless on a much higher plane than the so-called courage of slaves and brute beasts, because it is μετὰ παιδείας γεγονυῖα. In αὖθις—δίιμεν Siebeck ( Zur Chron. d. Pl. Dial. pp. 126 ff.) finds a promise of the Laches. To this view it seems to me a serious objection that the Laches has nothing to say of the characteristically Platonic distinction between ἐπιστήμη and ὀρθὴ δόξα: for that very reason it is probably earlier than this passage. Courage in the Laches is little more than Socratic courage (cf. Mem. IV 6. 10 ff.), for the knowledge of the good into which it is finally resolved is not knowledge of the Idea. Others have found in αὖθις a reference to the account of Courage in the individual (442 B), or to V 467 A ff., or to VI 486 B. None of these references are in point; and it is simplest to take Plato at his word. He drops the subject because further discussion of it would be irrelevant; he will resume it on another occasion if Adimantus wishes, but Adimantus is content. Cf. VII 532 D note and see also on 1 347 E. The whole of this section of the dialogue is important because it emphatically reaffirms the principle that courage as well as the other virtues enumerated here rests on ὀρθὴ δόξα and not on ἐπιστήμη. We have already seen that Plato's earlier scheme of education aims at implanting only ὀρθὴ δόξα. Cf. II 376 E note

νῦν γὰρ ‐‐ ἐζητοῦμεν . νῦν=‘as it is’: so that Cobet's ζητοῦμεν (found also in one or two MSS) is unnecessary.

πρότερον is omitted by Richards as illogical. So slight a flaw is easy to forgive; and ἔτι in μηκέτι suggests that πρότερον is genuine. Nor could Adimantus well have said that in any event he did not wish Justice— οὗ δὴ ἕνεκα πάντα ζητοῦμεν—to be discovered.

εἰ μὴ ἀδικῶ . Cf. X 608 D, 612 D, Charm. 156 A , Menex. 236 B . The translation “as I am an honest man” (D. and V.) is inaccurate; but Schneider's “ich thäte ja sonst nichts recht” hits the mark. In English we require an independent clause, ‘I have no right to refuse.’

ὥς γε ἐντεῦθεν ἰδεῖν : ‘seen from where we stand,’ i.e. on a first view: cf. ὡς ἐνθένδε ἰδεῖν Pol. 289 D , infra 432 B, X 595 B, and see Grünenwald in Schanz's Beiträge etc. II 3 pp. 1—37.

ξυμφωνίᾳ ‐‐ ἁρμονίᾳ . On ἁρμονία see III 398 E note In its musical application συμφωνία is used both of consonance as in the octave or double octave and also of other musical intervals: cf. VII 531 A and von Jan's Mus. Script. Gr. p. 102 and passim. The ξυμφωνία in which σωφροσύνη consists is apparently of the former kind: cf. 432 A note

ἡδονῶν ‐‐ ἐγκράτεια . It is chiefly this which is insisted on in the popular view of σωφροσύνη taken in III 389 D ff. Cf. Xen. Cyr. VIII 1. 32 , Isocr. 3. 44, and other passages cited by Nägelsbach Nachhom. Theol. II p. 233. Here the essential mark of σωφροσύνη is ξυμφωνία as to who shall be rulers, and who subjects; a point which is not mentioned in III. In other fundamental respects, also, the two descriptions differ; and Hirzel rightly insists that the σωφροσύνη of Book IV must be examined independently and by itself (l.c. p. 409).

κρείττω ‐‐ αὑτοῦ : a common formula in the popular acceptation of σωφροσύνη: see Nägelsbach l.c.

λέγοντες . See cr. n. λέγοντες is found also in Flor. A, in some MSS of Stobaeus ( Flor. 43, 97) and in Cesenas M. λέγοντες should (with Stallbaum) be taken as agreeing with the nominative of φασι, ‘as men say, calling one lord of oneself in some mysterious way.’ δή, ‘forsooth,’ helps out οὐκ οἶδ’ ὅντινα τρόπον. For other views on this passage see App. II.

κρείττω αὑτοῦ . Stallbaum reads κρείττων αὑτοῦ, and wishes to do so also in 431 A below. The accusative is more natural in both places, partly because it suggests τὸ κρείττω αὑτοῦ εἶναι (cf. ὁ γὰρ ἑαυτοῦ κρείττων καὶ ἥττων—ἂν αὑτοῦ εἴη), partly because of κρείττω αὑτοῦ just before.

τι ‐‐ τὸ μὲν ‐‐ τὸ δέ . For the subdivision of τις cf. V 463 B, VIII 560 A, Gorg. 499 C . Other examples of ‘partitive apposition’ are V 461 D, 477 C, VIII 552 C: cf. also VIII 556 B, IX 592 A, X 618 E.

τὸ κρείττω αὑτοῦ : sc. φαίνεταί μοι τοῦτο λέγειν. The expression κρείττω αὑτοῦ is the subject, and τοῦτο the object. τόν for τό (see cr. n.) is indefensible. See also on τοῦτο δέ below.

πλήθους τοῦ χείρονος . II 379 C note

τοῦτο δὲ ‐‐ διακείμενον . τοῦτο is the object of ψέγειν, whose subject is still strictly speaking τὸ κρείττω αὑτοῦ or (which is the same thing) οὗτος ὁ λόγος. In ψέγειν and καλεῖν the λόγος is halfpersonified: ‘this the phrase censures as something disgraceful, and calls the man who is in this condition a slave to himself and intemperate.’ For the recapitulatory τοῦτο δέ cf. Ap. 28 E with my note ad loc. Hartman's τοῦτο δή is an unhappy suggestion.

κρείττω ‐‐ αὑτῆς . Cf. Laws 626 E ff., where κρείττων αὑτῆς is similarly applied to a city and explained in the same way.

οὗ : not the adverb, as Stallbaum supposed, but a partitive genitive: ‘that whereof the better part rules the worse’ etc.

παισί . See cr. n. The corruption—an easy one in minuscule MSS—recurs in VI 494 B. See Introd. § 5 and Bast Comm. Pal. p. 705. The object of this part of the argument is to shew that our city is σώφρων not only as being κρείττων αὑτῆς but as being κρείττων ἡδονῶν τε καὶ ἐπιθυμιῶν—a kindred, but not quite identical, notion: cf. 431 D. In adding γυναιξί Plato speaks from the ordinary Greek standpoint; in permitting some women to be guardians, he tacitly allows that in some cases their desires (unlike those of οἰκέται etc.) are μετὰ νοῦ. Cf. Laws 780 E ff.

λεγομένων is emphatic. No one is free who is a slave to his desires. Cf. 1 336 A note

τὰς δέ γε κτλ. I have returned to the MS reading. The accusative with τυγχάνω and its congeners is—except with neuter pronouns (Jebb on Soph. O. T. 1298 )—almost unexampled ( ἐπιτόσσαις with accusative in Pind. Pyth. 10. 33), and Herwerden reads the dative, an easy correction; but it is perhaps safer to take the accusative as a sort of anacoluthon “occasioned by the parallel of the previous sentence” τάς γε πολλὰς—εὕροι (J. and C.). Baiter brackets the verb ἐπιτεύξει.

καὶ μὴν ‐‐ σφόδρα gives a third feature of the σωφροσύνη of a city. We have shewn our city to be (1) κρείττων αὑτῆς. (2) κρείττων ἡδονῶν τε καὶ ἐπιθυμιῶν. It is also (3) ὁμονοητικὴ περὶ τοῦ οὕστινας δεῖ ἄρχειν. (3) corresponds to κόσμος, (2) to ἐγκράτεια, (1) to κρείττω αὑτοῦ in 430 E. Thus the discussion in this chapter follows a chiastic order.

ἐν ἀμφοτέροις . Cf. 442 C, D. Aristotle and others seem to have supposed that σωφροσύνη was the special virtue of the lowest class in the State and the lowest element in the soul: see Top. V 6. 136^{b} 10 ff. and 8. 138^{b} 1 ff. and [Arist.] περἰ ἀρετῶν καὶ κακιῶν 1. 1249^{a} 30 ff. ἐν ἀμφοτέροις proves this view erroneous. The error arose partly perhaps from a desire to make the theory superficially symmetrical, partly perhaps from a notion that Plato's rulers would not be likely to dispute their own right to rule. But σωφροσύνη in Plato's sense is necessary for his Rulers as well as for their subjects; without it, they might nolle episcopari: cf. I 346 D note

δἰ ὅλης ‐‐ διὰ πασῶν . δἰ ὅλης sc. τῆς πόλεως, not λύρας, as J. and C. strangely suppose. διὰ πασῶν sc. τῶν χορδῶν should be taken with ξυνᾴδοντας (so also Schneider). ἡ διὰ πασῶν συμφωνία is the octave (Arist. Probl. XIX 35. 920^{a} 27 ff.), the καλλίστη συμφωνία, according to the Greeks (Arist. l.c.), readily sounding to the ear as absolute unison; hence the point of ταὐτόν, which is an accusative depending directly on ξυνᾴδοντας. See Arist. l.c. 14. 918^{b} 7 ff. διὰ τί λανθάνει τὸ διὰ πασῶν καὶ δοκεῖ ὁμόφωνον εἶναι, οἷον ἐν τῷ φοινικίῳ καὶ ἐν τῷ ἀνθρώπῳ; The whole expression διὰ πασῶν ξυνᾴδοντας ταὐτόν therefore means that the concord of the citizens on the matter in question is absolute and complete. Further than this I do not think the comparison is to be pressed. If we seek to find analogies between ἀσθενεστάτους, ἰσχυροτάτους, μέσους and the ὑπάτη, νήτη and μέση of the scale, we are met by the difficulty that the μέση cannot be said to produce the same ( ταὐτόν) note as the ὑπάτη and νήτη, and we are not at liberty to suppose that Plato is thinking of ἡ δὶς διὰ πασῶν in the face of his own words, which refer only to a single octave ( διὰ πασῶν παρεχομ νη κτλ.). In talking of σωφροσύνη Plato usually distinguishes only between two classes—rulers and ruled: 431 D, E and infra χείρονός τε καὶ ἀμείνονος. See also on 443 D.

φρονήσει ‐‐ ἰσχύϊ ‐‐ πλήθει define ἀσθενεστάτους, ἰσχυροτάτους, μέσους. The equipoise and measured cadence of this stately sentence may well suggest a chorus of voices singing in unison. Cf. III 401 C. Cobet's excision of the second βούλει is sadly out of tune.

ταύτην τὴν ὁμόνοιαν prepares us for the definition about to follow. There are various ὁμόνοιαι: this one is agreement ὁπότερον δεῖ ἄρχειν etc.

χείρονος κτλ. : ‘concord between the naturally better and the naturally worse, on the question which should rule, whether in a city or in an individual.’ ἐν ἑνὶ ἑκάστῳ anticipates 442 C f.; but is justified here by 431 A, B.

ὥς γε ‐‐ δόξαι . This phrase is apparently quite unique in Plato: see Grünenwald cited on 430 E.

ὥσπερ κυνηγέτας . The image is a favourite one with Plato: cf. Laws 654 E, Parm. 128 C , Lys. 218 C . Other examples may be found in Stallbaum's note on this passage. The particular kind of hunting from which Plato takes his illustration is clearly described in Xen. de Ven. 8. 4—8. A net was drawn round the bush where the hare was, and the hunters stood round, ready μεταθεῖν κατὰ τὰ ἴχνη, ἐὰν ἐκκυλισθῇ ἐκ τῶν δικτύων.

φράσῃς : ‘point out.’ There is no occasion to read (with Ast and q) καί μοι φράσεις.

μετρίῳ . See cr. n. μετρίως χρῆσθαι could only mean ‘to treat fairly,’ but this is not to the point. The only relevant meaning is ‘you will find me very tolerable,’ and μετρίῳ μοι χρήσει conveys this sense exactly. Cf. ἑπομένῳ χρῇ—καὶ δυναμένῳ and Xen. Cyr. III 2. 4 ὀλίγοις τε καὶ ἀσθενέσι χρησαίμεθ’ ἂν πολεμίοις, Symp. 2 9, 10. On the error see Introd. § 5.

εὐξάμενος : like a pious huntsman: cf. Xen. de Ven. 6. 13 εὐξάμενον τῷ Ἀπόλλωνι καὶ τῇ Ἀρτέμιδι τῇ Ἀγροτέρᾳ μεταδοῦναι τῆς θήρας. Cf. also (with Stallbaum) Phil. 25 B εὔχου δὴ καὶ σκόπει and Tim. 27 C . ἕπου οὖν (suggested by Richards) seems to me much less spirited and picturesque than ἕπου; and the cacophony is also unpleasing. For the asyndeton cf. II 373 E note

ἔστι γοῦν ‐‐ δυσδιερεύνητος has been objected to as adding little or nothing to δύσβατος—ἐπίσκιος. But δυσδιερεύνητος, ‘difficult to beat,’ said of beating or scouring the brake to rouse ( κινεῖν Xen. de Ven. 8. 7) the game and drive it out into the net, could ill be spared; so apt a word is much too good for a copyist. Cf. Menex. 240 B , where διερευνᾶσθαι is used of the famous ‘beating’ of Euboea by Datis' soldiers: and see also Laws 698 D.

ἰοῦ, ἰοῦ : ‘Joy! Joy!’ ἰού dolentis, ἰοῦ gaudentis, according to the Scholiast on Ar. Peace 318 : cf. Suidas s.v. Ancient authorities differed on the point (see Blaydes' critical note l.c.), but modern scholars for the most part agree with Suidas.

ἐκφευξεῖσθαι ‐‐ εὖ ἀγγέλλεις . The contracted form of the future of φεύγω is established by the authority both of the Paris MS, and also of Aristophanes and Euripides, as Schanz has proved (Vol. XII p. xvi). Schanz may be right in supposing that it is borrowed “ex ore populi.” For εὖ ἀγγέλλεις Phrynichus (S. V. εὐαγγελίζομαί σε) apparently read εὐαγγελεῖς, on which see Lobeck Phryn. p. 632 and Cobet N. L. p. 163. εὐαγγελῶ does not seem to be used in Attic prose. In Theaet. 144 B εὖ ἀγγέλλεις is read by B, εὖ ἀγγελεῖς by T.

ἤτοι . See 1 344 E note

τι εἶδος , like τρόπον τινά in 432 E and 433 B, hints, I think, that Civic Justice is not, after all, the true and original form of Justice. Hence, in 434 D, Plato is careful to warn us that the subject of Justice is not exhausted till individual Justice has been discussed. See on τοιοῦτο in 443 C.

ἐπιτηδειοτάτη . ἐπιτηδειότατα (Herwerden) is not good: cf. II 374 E and supra 430 A. A few MSS omit πεφυκυῖα, not unnaturally; but the reduplication in φύσις—πεφυκυῖα adds to the emphasis. Plato never tires of emphasizing the ‘natural’ features of his city in Books II—IV.

δικαιοσύνη has been questioned by Richards, on the ground that “the inference announced in τοῦτο τοίνυν κτλ. is already stated in καὶ μὴν ὅτι κτλ., which from its form ( καὶ μήν) is yet evidently only a step in the reasoning.” Richards suggests δίκαιον, and Hartman δικαιοσύνης, neatly but needlessly. τοίνυν in B does not express an inference, but is simply ‘well,’ as in II 369 B, III 413 C, IV 436 B and a host of other passages collected by Kugler ( de part. τοι etc. p. 35). Plato first states a popular view, and then proceeds to shew that it is mainly right on grounds presently to be stated (whence οἷσθ’ ὅθεν τεκμαίρομαι;). No stress should be laid on the fact that δικαιοσύνη is in one case the predicate, and in the other the subject: complete identity is predicated in both cases, as the abstract δικαιοσύνη shews. It might be different if we read δίκαιον, but for this there is no occasion. There is still however a difficulty in δικαιοσύνη: see next note.

εἰρήκαμεν γάρ . This has not been said in the Republic, nor (so far as I know) in any of Plato's earlier dialogues (if we except Alc. I 127 C), so that εἰρήκαμεν refers to ordinary conversation. Such a view has affinities with the legal view of Justice as the virtue which respects the rights of others (cf. 433 E and I 331 A ff.), and is natural enough, especially with the loose connotation which δικαιοσύνη had in popular language. It is however curious that in Charm. 161 B ff. precisely the same account is given of Temperance: ἄρτι γὰρ ἀνεμνήσθην ὂ ἤδη του ἤκουσα λέγοντος, ὅτι σωφροσύνη ἂν εἴη τὸ τὰ ἑαυτοῦ πράττειν: cf. Tim. 72 A εὖ καὶ πάλαι λέγεται τὸ πράττειν καὶ γνῶναι τά τε αὑτοῦ καὶ ἑαυτὸν σώφρονι μόνῳ προσήκειν. In its popular connotation, σωφροσύνη was not always distinguished from δικαιοσύνη, and even the philosophers (as Strabo VII 3. 4 observes) sometimes used the words in nearly an identical sense. See Nägelsbach Nachhom. Theol. p. 238. Steinhart and others find in the difference between this passage and the Charmides l.c. an indication of the Socratic and Platonic doctrine of the unity of Virtue. No doubt there is a certain sense in which virtue is one (see below on 434 C), but we must insist that the specific virtues are represented by Plato in the Republic as distinct; on any other hypothesis, the perfect City falls to pieces. Perhaps δικαιοσύνη after πολυπραγμονεῖν is an error for σωφροσύνη, and Plato is here deliberately correcting the popular view. If so, καὶ μὴν—γε means ‘and yet,’ i.e. in spite of what we now say that Justice is εἷς ἓν κατὰ φύσιν, ‘we and others have also said that Temperance is τὰ αὑτοῦ πράττειν.’ Adimantus assents. ‘Well,’ continues Socrates, ‘it is apparently (not Temperance, but) Justice which is τὰ αὑτοῦ πράττειν.’ This view gives a much better sense to καί in καὶ τοῦτο, and ἡ δικαιοσύνη receives the proper emphasis.

δοκεῖ ‐‐ εὕροιμεν . Things which are equal to the same thing are equal to one another. Now (1) the Virtue which enables the others to take root, and (2) Justice, each= τὸ ὑπόλοιπον. Therefore Justice enables the other Virtues to take root. τὰ αὑτοῦ πράττειν. Consequently Justice is τὰ αὑτοῦ πράττειν.> Plato seldom leaves so much to be mentally supplied in his reasoning.

παρέχει . See cr. n. Former editors (except Ast) retain παρέχειν and explain it as depending directly on δοκεῖ. If this is right, καί before ἐγγενομένοις joins τοῦτο εἶναι and παρέχειν; but καὶ ἐγγενομένοις γε, following immediately on ἐγγενέσθαι, naturally suggests that παρέχειν and ἐγγενέσθαι are coordinate and both under the government of ὥστε. That this was felt in antiquity is proved by the variant ἐγγενόμενα for ἐγγενομένοις, preserved in Stobaeus ( Flor. 43. 98) and in Ξ. The author of the reading ἐγγενόμενα must have understood Plato to mean ‘which enabled them all to make their appearance in the city, and having done so, to keep it safe, so long as they are there,’ and this, I think, is the natural meaning of Plato's words, if παρέχειν is retained. But the sentiment is compara tively weak; and consequently Ast and Hartman wish to cancel παρέχειν, making σωτηρίαν depend upon παρέσχεν; but a present tense is necessary. παρέχει seems to me what Plato wrote, ‘aye, and after they have appeared it preserves them, so long as it is present in the city.’ A relative clause often passes into an independent sentence (see on II 357 B); and the idiom is appropriate here because it responds to the emphatic καὶ—γέ. For καὶ—γε cf. 425 B note

δούλῳ ‐‐ ἀρχομένῳ . On δούλῳ see V 469 C note Richards would insert καὶ γεωργῷ after δημιουργῷ, pointing out that the other words go in pairs; but the difference between δημιουργῷ and γεωργῷ is insignificant, since both artisan and farmer belong to the same class in the city.

εἷς ὤν . Most of Stobaeus' MSS ( Flor. l.c.) read εἷς ὢν ἕν. ἕν is unnecessary with καὶ οὐκ ἐπολυπραγμόνει following (Schneider).

σκόπει κτλ. This τεκμήριον turns on the judicial sense of δικαιοσύνη: cf. I 331 E ff. The judicial functions of the rulers follow naturally from 428 D, where it is said that σοφία βουλεύεται—ὅντιν’ ἂν τρόπον αὐτή τε (sc. ἡ πόλις) πρὸς αὑτὴν—ἄριστα ὁμιλοῖ κτλ. It is clear that no class except the rulers can be judges in the State, and judges are necessary: see III 408 D ff.

ἑαυτοῦ κτλ. ἑαυτοῦ is a possessive genitive depending on τοῦ. It should be noted that although ἕξις τοῦ οἰκείου is not the same thing as πρᾶξις τοῦ οἰκείου, the latter involves the former. Plato is looking for a point of contact between his own view of Justice and the popular judicial meaning of the word, and finds it in ἕξις τοῦ οἰκείου. Krohn ( Pl. St. p. 49) appears to me to attach too much weight to ἕξις τοῦ οἰκείου when he calls it a new “Begriffselement,” and complains that it is “weder sachlich erläutert, noch logisch streng abgeleitet.” Cf. 442 E. It should be mentioned that the poet Gray (with less than his usual critical acumen) conjectured τοῦ <ποιεῖν> τὸ οἰκεῖόν τε καὶ <τὸ> ἑαυτοῦ, comparing οἰκειοπραγία in 434 C.

πάντα τἄλλα means everything except what Socrates is about to mention, that is everything except the interchange of rulers and ruled. So J. and C., rightly, I think: cf. 421 A, VII 518 D and Laws 798 D. Other editors explain τἄλλα as “reliquorum opificum opera”; and so also q, reading ἢ πάντα τἄλλα τά γε τοιαῦτα; but it is difficult to extract this meaning out of τἄλλα without τά γε τοιαῦτα, and the asyndeton is also very harsh. Madvig's conjecture ταῦτα is improbable, though adopted by Baiter. Adimantus would catch the meaning all the more easily on account of the similar statement in 421 A, and because πάντα would be pronounced with emphasis, as the asyndeton also indicates. I have removed the comma usually printed after μεταλλαττόμενα; for πάντα τἄλλα includes within its scope all the cases mentioned, and is directly the subject of δοκεῖ.

ἀλλ’ ὅταν κτλ. Plato is probably thinking of Athens again: cf. supra 424 D note and Krohn Pl. St. p. 46. φύσει belongs to ὤν. Hartman needlessly expunges ὤν and reads φύς for φύσει. The subject to ὤν is simply the pronoun ‘he,’ used loosely, as often in English.

βουλευτικοῦ ‐‐ ὤν . “Valde miror editt. verba βουλευτικοῦ καὶ φύλακος ἀνάξιος ὤν concoxisse” cries Hartman. The genitives of course depend on τὸ ( εἶδος), and ἀνάξιος is used absolutely, as often.

μάλιστα κακουργία . μάλιστα is omitted in Ξ and one or two other MSS; but cf. VII 532 B ἔτι ἀδυναμία, VIII 564 A εἰς ἄγαν δουλείαν (with Stallbaum ad loc.), and other examples in Kühner Gr. Gr. II p. 526. That μάλιστα should be taken with κακουργία is clear from μεγίστη βλάβη and κακουργίαν τὴν μεγίστην.

πάλιν : not ‘again,’ but ‘conversely,’ “umgekehrt” (Schneider).

ἑκάστου ‐‐ πόλει is cancelled by Herwerden as a marginal note on οἰκειοπραγία. The words add to the weight and impressiveness of the sentence, and have a decidedly Platonic sound.

τοὐναντίον ἐκείνου . ἐκείνου is ‘the other,’ i.e. πολυπραγμοσύνης (rather than ἀδικίας); and τοὐναντίον is probably nominative, and not adverbial accusative. So also Schneider. It is not necessary to add ὄν after τοὐναντίον as I formerly did. The style of argument is the familiar τόπος ἐκ τῶν ἐναντίων (see Arist. Rhet. II 23. 1397^{a} 7 ff.).

παγίως ‐‐ λέγωμεν : cf. V 479 C παγίως νοῆσαι, Theaet. 157 A νοῆσαι —παγίως, Tim. 49 D . οὐκ ἔστι παγίως νοῆσαι was probably a phrase in vogue among Heraclitus' followers: see Wohlrab on Theaet. l.c.

ἰὸν τὸ εἶδος . εἶδος is not yet the Idea (III 402 C) but refers to οἰκειοπραγία. For ἰόν Richards conjectures ἰοῦσιν; but surely εἰς would then be wrong. How can ‘we’ be said to pass into an individual? The εἶδος is half personified (cf. ὅταν—ἐλθὸν ἐρώτημα ἔρηται VII 538 D); it is said to ‘pass into’ the individual merely because we have discovered it first in the State. See also on ἀπαμβλύνεται 442 D. The passage in Phaedr. 249 B is different, whether we accept Badham's conjecture ἰόντ) or not.

ἣν is a loose internal accusative, exactly like ὅ in 443 B below. The reference is to II 368 D.

ἐκεῖ . The reading ἐκεῖνο, found in Ξ and other second-rate MSS, would probably have been discarded sooner, if it had been known that A as well as Π reads ἐκεῖ. Campbell first pointed this out. ἐκεῖνο is not quite suitable because, although it must mean justice, it suggests something more remote. ἐκεῖ on the other hand helps out the antithesis between ἐν μείζονι—ἐχόντων and ἐν ἑνὶ ἀνθρώπῳ, and is in harmony with ἐκεῖ ἐφάνη below. δικαιοσύνην depends on θεάσασθαι, and τῶν ἐχόντων is ‘its possessors’: cf. II 367 B, D, E. In reciting the sentence, the voice pauses after ἐχόντων and pronounces ἐκεῖ with emphasis. ἐκεῖ (with which cf. ἐκείνου in Parm. 133 D ) was rightly retained by Stallbaum, who did not know that it was the reading of A.

τοῦτο : i.e. τὸ μεῖζον τῶν ἐχόντων δικαιοσύνην.

καθ’ ὁδόν . Cf. (with Schneider) infra VII 533 B and Crat. 425 B . μέθοδον for καθ’ ὁδόν (Herwerden) is a sorry piece of criticism.

μεῖζον ‐‐ ἔλαττον : ‘whether greater or smaller.’ The insertion of ὄν after ἔλαττον, suggested by Dobree, is unnecessary.

τὰ αὐτὰ ταῦτα εἴδη . εἴδη used in this sense is slightly confusing after εἶδος has just been applied to δικαιοσύνη; and τῶν αὐτῶν τούτων γενῶν would lead us to expect γένη. The psychological elements are called εἴδη, γένη, or μέρη: εἴδη in 435 B, C, E, 439 E, γένη in 441 C, 443 D, μέρη in 442 B, C and (by implication) 439 B, C, D and passim. Cf. Brandt l.c. p. 17 and Zeller^{4} II 1, p. 845. εἴδη ψυχῆς does not, strictly speaking, mean ‘ varieties of soul’ but rather ‘kinds’ belonging to or present in soul ( εἴδη ἐν ψυχῇ 439 E: see also on III 402 C), and much the same is true of γένη. There is some authority for holding that the Pythagoreans before the time of Plato recognised at least two ‘parts’ of soul— an ἄλογον and a λογικόν (see Diels Dox. Gr. pp. 389 f. and other evidence in Rohde Psyche^{2} II p. 170 note); but Zeller I^{5} pp. 447, 448 may be right in regarding the Pythagorean form of this theory as post-Platonic.

φαῦλον is of course ironical, although Glauco pretends to take it seriously. Cf. (with J. and C.) 423 C—E, 426 A, B.

καὶ εὖ γ) ‐‐ ἐξαρκέσει . The difficulties connected with this passage have led to much discussion: see for example Rettig Proleg. pp. 126 ff., Krohn Pl. St. pp. 128 ff., 144, Pfleiderer Zur Lösung etc. pp. 25, 73, Hirmer Entst. u. Komp. etc. p. 618. τοῦτο in ἀκριβῶς μὲν τοῦτο and in ἡ ἐπὶ τοῦτο ἄγουσα ought, so far as grammar goes, to mean the question whether the soul has τρία εἴδη or not. But the μακροτέρα περίοδος in VI 504 B ff., where Plato expressly refers back to this passage, eschews the psychological problem altogether. The μακροτέρα περίοδος of Books VI—VII is in harmony with the present enquiry in so far as it seeks to determine the nature of Justice and the other virtues (VI 504 D, 506 A), but it is nowhere in the Republic expressly used either to confirm or to overthrow the triple division of soul which is here propounded. (The analysis of mental faculties in VI 509 D—511 E is introductory to the μακροτέρα περίοδος, not a result obtained by it; nor has that analysis, strictly speaking, any bearing on the question whether soul has three εἴδη or not: cf. Pfleiderer Zur Lösung etc. p. 25.) Krohn accordingly holds that the ‘longer ways’ of IV and VI are different and distinct ( Pl. St. p. 128); and Schleiermacher supposes ( Einleitung p. 71) that the πλείων ὁδός of IV is to be found in the psychology of the Timaeus; but that Plato meant the two ways to be identical is certain, for he explicitly says that they are (VI 504 B ff.). The only way out of these difficulties is to suppose that τοῦτο here was not intended by Plato to refer to the psychological, but to the ethical question, to which the psychological enquiry is introductory. τοῦτο must then be taken as δικαιοσύνης τε πέρι καὶ σωφροσύνης καὶ ἀνδρείας καὶ σοφίας δ ἕκαστόν ἐστι (VI 504 A). This view becomes easy if we suppose that the words καὶ εὖ γε—ἐξαρκέσει were not written by Plato immediately after he wrote 435 C, but at a later time, when VI 504 A—D was composed. It is in itself highly probable that the most important passages referring forward or backward to one another throughout the dialogue were either written together, or at all events revised by Plato side by side. Cf. Brandt l. c. p. 13 note 3, where a kindred view is taken. In any case, we must adhere to our explanation of τοῦτο, if we would preserve the artistic unity of the Republic. See also on VI 504 A—D.

ἄλλη . See cr. n. ἄλλη is in itself much better, to say the least, than ἀλλά, and is confirmed by ἄλλη μακροτέρα— περίοδος in VI 504 B. The corruption was easy, owing to the frequency of ἀλλὰ γάρ .

ὅτι γε ‐‐ πόλει . Broadly speaking, what Plato says is true, that the predominant character of a State depends on the predominant character of the individual citizens (cf. Bosanquet Companion pp. 147 f.): but it does not necessarily follow, because a city contains three psychologically different classes of citizens, that each of us ( ἑκάστῳ ἡμῶν) has within his soul the three corresponding psychological elements. In making this assertion, Plato relies upon the fundamental hypothesis of the Republic, viz. that the individual is a commonwealth writ small. See on II 369 A. γε after ὅτι, though omitted in Ξ, is strictly appropriate, and warns us of a further point— τόδε δὲ ἤδη χαλεπόν 436 A—on which agreement is not so easy.

οἳ δὴ ‐‐ αἰτίαν : ‘that is, among peoples who bear this reputation.’ ταύτην is τοῦ θυμοειδεῖς εἷναι. The phrase αἰτίαν ἔχειν is used both in a good and in a bad sense as the passive of αἰτιῶμαι: for the good sense cf. (with Ast) Gorg. 503 B . What follows is (as Teichmüller observes Lit. Fehd. I p. 146) conceived in the vein of Hippocrates' enquiries as to the influence of climate on character: see his treatise de aere aquis locis 12 ff. ed. Kuehlewein, and cf. also Arist. Physiog. 2. 806^{b} 15, Probl. XIV 8, 15, 16, and especially Pol. H 7. 1327^{b} 23—33 with Susemihl's note. Aristotle for his part represents the Greek nature as the mean between the two extremes of oriental διανοητικόν and τεχνικόν and northern θυμός. There is no good reason for supposing (with Steinhart Einleitung p. 191) that Plato was thinking of the wild races of the North when he instituted his second order of citizens, and of Egyptians etc. when he established his third. On the Phoenician and Egyptian characters cf. Laws 747 C ff.

τὸν ἄνω τόπον : ‘the Northern region,’ not ‘the highland country’ (L. and S.): cf. Arist. Meteor. II 5. 362^{a} 33 τὸν ἄνω πόλον and Hdt. I 142 al.

αἰτιάσαιτο . εἶναι should be understood. For the construction cf. X 599 E.

φιλοχρήματον is another name for ἐπιθυμητικόν, ὅτι διὰ χρημάτων μάλιστα ἀποτελοῦνται αἱ τοιαῦται ἐπιθυμίαι (IX 580 E).

τούτων ἕκαστα refers to the actions described in μανθάνομεν μὲν ἑτέρῳ etc. τούτῳ (see cr. n.) can only be defended by referring it (with Schneider) “to the subject of the triple predicate τὸ θυμοειδές, τὸ φιλομαθές, and τὸ φιλοχρήματον.” There is a certain obscurity in this construction, and τούτων ἕκαστα prepares us for μανθάνομεν μὲν ἑτέρῳ, θυμούμεθα δέ etc. better than ἕκαστα alone would do.

ταὐτὸν ‐‐ ἅμα is the earliest explicit statement in Greek literature of the maxim of Contradiction; cf. Theaet. 188 A , Phaed. 102 E , 103 B, Soph. 230 B and infra X 602 E. Plato may have been led to formulate it in opposition to Heracliteanism, which was supposed by some to be the negation of the principle (see Arist. Met. Γ 3. 1005^{b} 24 and Theaet. 152 D ff.), or against the Megarian puzzles (see RP.^{7} § 226), or as a counterblast to both. Many of the sophistries of the Euthydemus turn on the violation of this law. In Aristotle's formula ( Met. l.c. 1005^{b} 19) πρὸς ταὐτόν does not occur; and Hartman would cancel καὶ πρὸς ταὐτόν here and πρὸς τὸ αὐτό in 436 E, on the ground that it means the same as κατὰ ταὐτόν. But assuredly it does not. κατὰ ταὐτόν is ‘in the same part of it’ as the instances presently cited shew; while πρὸς ταὐτόν is ‘relatively to the same thing,’ viz. to something other than the subject of the proposition. πρὸς τὰ αὐτά and κατὰ ταὐτά are also both of them found in the parallel passage Soph. 230 B . πρὸς ταὐτόν covers such cases as are adduced in Theaet. 154 C —155 C: six dice are πλείους πρὸς τέτταρας, ἐλάττους πρὸς δώδεκα, but they are not ἐναντία πρὸς ταὐτόν. Cf. VII 524 A ff., and see also on ἢ καὶ εἴη in 437 A.

ἦν is not precisely ἐστὶν ὥσπερ ᾠόμεθα (Stallbaum); for the reference is actually to the past, and the past tense should be kept in translating it. See II 357 A note and cf. X 609 B. The so-called ‘philosophic imperfect’ gets credit for more than it deserves, because we are apt to suppose that the past excludes the present, which is not always true: cf. VI 497 C note

εἶεν by itself in replies is rare. It occurs (if the MSS are right) in Symp. 206 E , Crat. 410 C , Men. 75 C . In the last two passages, Heindorf (on Crat. l.c.) is inclined to rearrange the speakers; but it is safer, both there and here, to keep the traditional arrangement. See on I 332 D.

χαριεντίζοιτο ‐‐ κομψευόμενος may refer to some Megarian quibbles on this subject. Zeno's argument to shew that ἡ οἰστὸς φερομένη ἕστηκεν proceeded on a different principle: see Arist. Phys. VI 9. 239^{b} 30 ff.

ἢ καὶ ‐‐ δρᾷ . “Repetendum ὡς ex praegressis” (Stallbaum). Schneider connects δρᾷ with ὅταν: in that case we must understand after τοῦτο δρᾷ something like ὡς καὶ τοῦτο ὅλον ἕστηκέ τε ἅμα καὶ κινεῖται. Stallbaum's view is the simpler, and should, I think, be preferred. I have accordingly removed the comma usually printed after κινοῦνται .

ὡς οὐ ‐‐ φερομένων . This clause has proved a source of great perplexity. Schneider suggests that μενόντων is a partitive genitive, ἐστί being omitted; Stallbaum, that τὰ τοιαῦτα is adverbial, like τοιουτοτρόπως; while, according to J. and C., τὰ τοιαῦτα “is to be taken as cognate accusative with the participles.” Rather than accept any of these suggestions, it would, I think, be preferable to expunge τὰ τοιαῦτα altogether (with Ast), or to place it after ἀποδεχοίμεθα (as Gildersleeve suggests, A. J. Ph. VI p. 333 note 2), or even perhaps to read τῶν τοιούτων with Richards, although little short of a miracle could have corrupted τῶν τοιούτων to τὰ τοιαῦτα. The following interpretation, which appears to me right, has not, so far as I know, been hitherto suggested. ταὐτά goes closely with the partitive genitive ἑαυτῶν, and is a predicate to τὰ τοιαῦτα, which is also governed by κατά (cf. the familiar usage with ὥσπερ and a preposition in similes, e.g. Theaet. 170 A ὥσπερ πρὸς θεοὺς ἔχειν τοὺς ἐν ἑκάσταις ἄρχοντας etc.: see on VIII 553 B). μενόντων τε καὶ φερομένων is a genitive absolute. The sentence is in every respect an elegant and idiomatic piece of Greek, and means: ‘because such parts, in respect of which they both stand still and move on these occasions, are different parts of them.’ τὰ τοιαῦτα—the meaning of which is easy to catch after the examples given above —forms a welcome preparation for εὐθύ τε καὶ περιφερές in the following clause.

ἔστιν . I formerly rejected this word (with Galen de Hipp. et Plat. decr. IX Vol. V p. 799 ed. Kühn, Herwerden, and Flor. U). It is certainly more pointed to connect ἑστάναι with φαῖμεν ἄν, and Glauco's καὶ ὀρθῶς γε (sc. φαῖμεν ἄν) is easier without ἔστιν. But there is not sufficient ground for deserting the best MSS. For other examples of replies referring to the earlier part of the previous sentence see V 465 E note

ἢ καὶ εἴη . I agree with Bekker, Schneider, and J. and C. in retaining these words, which Galen l.c. also read, and only a few inferior MSS (with the majority of editors) omit. If the words are spurious, no satisfactory theory has yet been advanced to account for their presence in the text; certainly no scribe is at all likely to have added them. A fuller and more emphatic statement of the maxim is natural enough after the emphasis with which the sentence opens ( οὐδὲν—ἐκπλήξει), and Schneider truly observes: “obiter et quodam modo praeter exspectationem eius” (i.e. τοῦ εἶναι), “mentionem fieri adiectum καὶ indicat, quod semel positum mox sine offensione repetitur, omissis vero verbis ἢ καὶ εἴη ante ποιήσειεν non magis quam supra p. 436 B ante πάσχειν locum habiturum fuisset.” πάθοι and ποιήσειεν have reference to actions, εἴη to a state, and εἴη naturally follows πάθοι because e.g. πλείους γίγνεσθαι (an example of πάσχειν) leads up to πλείους εἶναι. It should also be observed that the meaning of πρὸς τὸ αὐτό, which the discussion has not yet brought out, is best apprehended in examples not of πάσχειν or ποιεῖν, but of εἶναι τἀναντία: see 436 B note

ἀλλ’ ὅμως κτλ. The usual Greek idiom, as shewn for example in ἀλγῶ τὴν κεφαλήν (cf. V 462 C ff.), rests on a psychological theory which is inconsistent with that now proposed by Plato. This may be one reason why Plato is at such pains to establish and emphasize his point.

λαβεῖν has been doubted: but see III 407 B note

ἂν (see cr. n.) is better inserted after ἐναντίων than after θείης (Ast) or τοιαῦτα (Hartman). Stallbaum (who formerly read ἂν θείης) in his last edition acquiesces, like Schneider, in the omission of ἄν; but few will agree with him. I have noted the—certain or probable— omission of ἄν in all or the best MSS in Phaed. 62 C , 109 E, Euthyd. 291 E (?), Rep. V 457 D, VII 516 E, VIII 558 D, where the omission is lipographical; also in Phaed. 72 B , Euthyd. 281 C , Crat. 389 E , 409 A, Alc. I 132 B, 133 E, Soph. 266 A , Phil. 47 B , H. Mai. 295 A. Sometimes (as occasionally after πρίν) the omission is perhaps a poetical touch: see my note in Cl. Rev. IV p. 103.

καὶ αὖ . Krohn ( Pl. St. p. 57) presses αὖ too much when he says that ἐθέλειν and βούλεσθαι are definitely represented as not belonging to the category of ἐπιθυμίαι. Plato expresses no opinion on this point; for αὖ, ‘also,’ merely marks the introduction of two new terms.

ἐθέλει ‐‐ ἐρωτῶντος . The difference between ἐθέλει, ‘is willing,’ and βούλεται, ‘wishes,’ is well brought out by the contrast between the more active process described in προσάγεσθαι and the passive assent which ἐπινεύειν expresses. The point is missed by translating (with J. and C.) ‘beckons this with a nod towards herself’: it is merely ‘nods assent to this in reply to herself.’ One part of the soul asks, and the other answers, the psychological process being compared to a kind of dialectic or question and answer inside the soul: see III 400 D note and cf. Isocr. Antid. 256. For the confusion of ἐρῶντος and ἐρωτῶντος—ἐρῶντος is found in several MSS— cf. [ Erast.] 132 D, and Euthyphr. 14 C . With the analysis of desire in this passage cf. Phil. 34 E ff.

ἀπ) αὐτῆς . ἀφ) αὑτῆς Hartman (with Vind. E only), but ἀπελαύνειν is active, not middle. The actions are described as though by a spectator ab externo.

ἐπιθυμιῶν : a defining genitive. For εἶδος see III 402 C note

ἆρ’ οὖν κτλ. This discussion (down to 438 E) is apparently regarded by Susemihl ( Gen. Entw. II pp. 163 f.) as unnecessary for the immediate purposes of the argument, but it is not so. Plato's object is to remove a difficulty which might be felt in holding that desire is restrained, and that by the λογιστικόν. Why should thirst be restrained? an objector might ask. You yourself, Socrates, hold that (1) desire is always of the good; consequently (2) thirst is always the desire of good drink, and (3) is therefore always good. See 438 A, where the gist of the objection is contained. Socrates would reply: The fallacy lurks in (2), for ‘good’ drink is ambiguous. If ‘good’ drink means drink which desire thinks good, then (2) is true; if it means drink which is in reality good, (2) is not true. Desire cannot know what is good. We must therefore amend (2) by omitting ‘good,’ for in reality it is sometimes good and sometimes bad to drink. To what then is the final appeal? To the λογιστικόν. It is this which decides on each occasion whether it is really good or bad to drink, and gives or refuses its assent accordingly (439 C). Bosanquet takes a somewhat similar view ( Companion p. 154). See also notes on 438 A.

οἷον δίψα ‐‐ ψυχροῦ . ‘Thus thirst is thirst—of hot drink, is it, or of cold?’ For the genitive with δίψα (which Richards doubts) cf. 439 A. The repetition of δίψα is like that of ἐπιστήμη in 438 C, and makes the statement formal and precise.

ψυχροῦ ‐‐ θερμοῦ . Hermann transposes these words and is followed by Stallbaum, Baiter, and others. “Palmaria emendatio,” cries Stallbaum; whereas J. and C. hold that it “makes nonsense of the passage.” It is not at first sight quite easy to decide between these conflicting views. The words ἐὰν μέν τις—προσπαρέχοιτ’ ἄν clearly mean that the desire of cold drink is due to thirst plus heat, i.e. thirst supplies the desire of drink, and the heat present in the thirst supplies in addition ( προσπαρέχοιτ’ ἄν) the desire of cold: see also on τοῦ δὲ—προσγιγνόμενα below. This is in harmony with common sense and also with the theory of Lys. 215 E ἐπιθυμεῖν γὰρ τοῦ τοιούτου (sc. ἐναντίου ) ἕκαστον, ἀλλ’ οὐ τοῦ ὁμοίου. τὸ μὲν γὰρ ξηρὸν ὑγροῦ, τὸ δὲ ψυχρὸν θερμοῦ κτλ.: cf. also Symp. 186 B . But ἐὰν δὲ—πολλὴ ἡ δίψα ᾖ seems to proceed on the opposite or homoeopathic principle. The presence of πλῆθος produces a desire not for its opposite but for itself. The solution of the difficulty is to be found in the different character of the notions θερμότης and πλῆθος. θερμότης is something distinct from δίψος, though superadded to it, for which reason Plato does not use the expression θερμὸν δίψος; whereas πλῆθος is in reality πλῆθος δίψης, and πολλὴ δίψα, as experience shews, desires much drink. The common sense point of view is taken by Plato throughout, and is expressly justified by him in 438 E οὔ τι λέγω ὡς οἵων ἂν ᾖ, τοιαῦτα καὶ ἔστιν. For these reasons I heartily agree with the Oxford editors. Hermann's proposal is a product of the inveterate tendency to suppose that wherever we turn in Plato we rub against the theory of Ideas; but the use of παρουσία here (in spite of Peiper's Ontol. Pl. pp. 602 ff., Zeller^{4} II 1, p. 560 note, and many other critics) is not metaphysical, but logical, and πλῆθος is certainly not an Idea in this passage. See on this point 438 B, 438 C notes

αὐτοῦ πώματος : ‘merely of drink’ )( much drink, cold drink, etc. Cf. VIII 559 A αὐτοῦ σίτου τε καὶ ὄψου. For καὶ αὖ κτλ. Herwerden would expect καὶ αὐτὸ <τὸ> πεινῆν <αὐτοῦ> βρώματος. Further specification than Plato gives is unnecessary, for τὸ πεινῆν as well as αὐτὸ τὸ δίψος is subject to οὐ μή ποτε—οὖπερ πέφυκεν. The voice pauses slightly after πεινῆν.

τοῦ δὲ ‐‐ προσγιγνόμενα : as e.g. θερμότης, where it προσγίγνεται τῷ δίψει (cf. E above), is the desire of ψυχροῦ, ψυχρότης of θερμοῦ. The type of desires illustrated by the desire of ψυχρὸν πῶμα appears to Plato composite and not simple.

μήτοι has been doubted, and is not, apparently, elsewhere so used in Plato (Kugler de part. τοί etc. p. 11), though often in Tragedy. Here too it strikes, I think, a lofty note ‘Wherefore let not any’ etc. θορυβήσῃ is also highly dramatic. All this parade is affected because it is a deduction from one of his own favourite commonplaces which Socrates is about to parry: see next note.

πάντες γὰρ κτλ. γὰρ ἄρα—a rare combination—occurs also in Prot. 315 D , Symp. 205 B (according to Ven. T, but the Bodleian reads γάρ), Laws 698 D. ἄρα indicates that the objector is quoting another man's view (II 358 C note), and the doctrine that all men desire the good was in point of fact a commonplace in the Platonic school. See for example Gorg. 468 A , Men. 77 C ff., Symp. 204 E and Rep. III 413 A, VI 505 D. Here, as always, Socrates would of course concede that all men desire the good; but we need the λογιστικόν in each act of desire to specify what the good really is (437 D note). Moreover, according to our present theory, the desire of good drink is the product of two desires, viz. (1) thirst or the desire of drink, and (2) the desire of good. That (2) is in a certain sense universal, does not alter the fact that the two desires are logically distinct. See on τοῦ δὲ—προσγιγνόμενα 437 E.

αὐτὰ ἕκαστα . αὐτά is ipsa, i.e. by themselves, alone, without qualification: cf. αὐτὰ—μόνα αὐτῶν μόνων in D and αὐτοῦ πώματος etc. 437 E. Plato now proceeds to establish the universality of his rule. It is obvious that the reasons for believing the rule true of ἐπιθυμία are confirmed if we can shew that it is true universally. The phraseology of this passage— πλήθους παρουσία, αὐτὰ ἕκαστα, αὐτὴ ἐπιστήμη— is no doubt interesting for the light which it throws on the origin of the terminology adopted in the Theory of Ideas (cf. VI 507 B note): but we could make no greater mistake than to suppose that Plato is here speaking of hypostasized Ideas. Cf. Pfleiderer Zur Lösung etc. p. 19.

τὸ μεῖζον ‐‐ μεῖζον . Cf. (with Stallbaum) Charm. 168 B ff., where the nature of relative notions is similarly defined: also Gorg. 476 B ff.

τὰ βραδύτερα . Stallbaum and others read βραδύτερα without the article (on slight MS authority), but praestat lectio difficilior. Cf. εἴτε ἐγγείων εἴτε τῶν ζῴων VI 491 D. τά is certainly not wrong, and the variety of expression is pleasing: ‘and heavier also to lighter, and swifter to that which is slower—do they not stand to one another in this relation?’ i.e. such that if βαρύτερα, for example, is qualified, κουφότερα is qualified too.

ἐπιστήμη μὲν αὐτή . ‘Knowledge and nothing more,’ as opposed to knowledge plus some specification, e.g. astronomical knowledge, literary knowledge etc. It is interesting and instructive to study Parm. 134 A ff. side by side with this passage. There αὐτὴ ἐπιστήμη has for its object τῆς ὃ ἔστιν ἀλήθεια, i.e. the Ideas; here we do not soar so high, for μαθήματος αὐτοῦ is only ‘learning and nothing more’ )( e.g. physical learning, classical learning, etc.

ἐπειδὴ ‐‐ κληθῆναι . Plato's theory is very clearly conceived. οἰκοδομικὴ ἐπιστήμη is a combination of αὐτὴ ἐπιστήμη and οἰκοδομία: ἐπιστήμη correlates with μάθημα, οἰκοδομία with οἰκίας ἐργασία, so that ἐπιστήμη οἰκοδομική is ἐπιστήμη οἰκίας ἐργασίας μαθήματος: it is therefore ποίου τινός (i.e. in this case οἰκοδομικοῦ) μαθήματος. Cf. note on τοῦ δὲ—προσγιγνόμενα in 437 E.

τῶν ὑγιεινῶν καὶ νοσωδῶν . If we carry the analysis less far than Plato, we can still make the added determinants the same by saying that ἰατρικὴ ἐπιστήμη is of ἰατρικὸν μάθημα. But this will not suit with κακή, for ‘bad knowledge’ is not ‘knowledge of bad things’; nor does it—in many cases— apply to desires. Cf. 437 E note

αὐτοῦ οὗπερ ‐‐ ἐστίν : i.e. μαθήματος αὐτοῦ. αὐτοῦ is emphatic and contrasted with ποιοῦ τινός.

τὸ δὲ δὴ δίψος κτλ. Here begins the application of the argument on Relativity.

οὐ τούτων κτλ. : i.q. οὐ θήσεις τὸ δίψος εἶναι τοῦτο, ὄπερ ἐστί, τούτων τῶν τινός sitim esse id, quod est, inter ea s. tanquam unum eorum, quae alicuius sunt (Schneider). We must, I think, acquiesce in this interpretation, if the text is sound; but there is grave difficulty in taking εἶναι twice over, as Schneider virtually does (‘is that which it is, and is one of,’ etc.). I am strongly inclined to think that Plato wrote οὐ τούτων θήσεις τῶν τινός, <καὶ τινὸς> εἶναι τοῦτο ὅπερ ἐστὶν κτλ. With this emendation the meaning is: ‘Well now, about thirst, will you not place it in this category of things relative, and hold that it is what it is—that is, of course, thirst—relatively to something? Yes, said he, relatively to drink.’ τὰ τινός i.e. ‘the things relative to something’ for ‘the category of things relative,’ is further explained in καὶ τινὸς —ὅπερ ἐστίν. ἔγωγε answers the first part of Socrates' question, and πώματός γε the second. For other views on this passage see App. III.

δίψος δ) οὖν κτλ. δ) οὖν=‘however,’ as in I 337 C. The reading δ’ αὖ ( q and some other inferior MSS) is unpleasantly cacophonous before αὐτό .

οὐ γὰρ δὴ ‐‐ πράττει . See cr. n. Ast's emendation πράττει is preferable to inserting ἄν or changing δή to ἄν (with Schanz). The particle δή could ill be spared. The infinitive πράττειν is read by Galen ( de Hipp. et Plat. decr. V p. 488 ed. Kühn) and two inferior MSS. Those who retain the MS reading suppose that ἄν is carried on from ἕτερον ἂν εἴη; but the instances cited in support (I 352 E, II 360 C, 382 D, III 398 A) are very much easier than this. περὶ τὸ αὐτό refers of course to the object of the action in question: πῶμα for instance in a case of thirst. Note that Plato betrays a sense of the unity of soul when he uses the expressions αὐτὴν—διψῶσαν, and τό γε αὐτὸ—πράττει. So also in D below ᾧ λογίζεται sc. ἡ ψυχή. See on 435 A ff.

αὐτοῦ . See 428 A note The illustration, as Bosanquet conjectures, may have been suggested by Heraclitus' παλίντροπος ἀρμονίη ὄκωσπερ τόξου καὶ λύρης ( Fr. 45 B yw.).

οὐκ ἐθέλειν : ‘refuse’: cf. infra τὸ κωλῦον—κρατοῦν τοῦ κελεύοντος. So also Bosanquet ‘decline to drink.’

ἐνεῖναι δέ . The repetition of ἐνεῖναι with μέν and δέ has almost the force of a conjunction: cf. Phaed. 83 A ἀπάτης μὲν μεστὴ ἡ διὰ τῶν ὀμμάτων σκέψις, ἀπάτης δὲ ἡ διὰ τῶν ὤτων. It is quite unnecessary to insert καὶ after δὲ (with Ast and Hartman). For the verbal play in κελεῦον —κωλῦον cf. III 406 B note

ὅταν ἐγγίγνηται ‐‐ παραγίγνεται . See cr. n. The present ἐγγίγνηται is, I think, necessary, and the corruption (through ἐγγίνηται) easy enough. ὄταν ἐγγένηται could scarcely mean ἑκάστοτε, which is the meaning here required. The subject to ἐγγίγνηται is τὸ κωλῦον. It is not hinted that ‘all men have not right reason’ (J. and C.), but only that there is not on every occasion a conflict between reason and desire. See 431 C and 437 D note Reason readily acquiesces when it is good to gratify desire. τὰ—ἄγοντα καὶ ἕλκοντα is translated by Jowett ‘that which bids and attracts’: but ἄγοντα is said like ἄγοντος in 439 B and ἕλκοντα is ‘dragging.’ The plural should also be retained in the translation, otherwise τὰ ἄγοντα may be identified with the ἐπιθυμητικόν, which would be a mistake, for the appetitive part of soul is certainly not produced by παθήματα of any kind. τὰ ἄγοντα καὶ ἔλκοντα are in reality ‘impulses leading and dragging’ the soul, impulses engendered by ‘particular conditions and diseases’ (not ‘passive states’ or ‘passion,’ etc. with the English translators), i.e. in other words by abnormal bodily states favourable to desires, as for example fevers etc.: cf. Phil. 45 A , B. These impulses are no doubt special instances of the action of ἐπιθυμητικόν, but should be distinguished from the appetitive principle itself.

λογιστικόν . The φιλόσοφον of II and III shewed itself in moral rather than in intellectual relations: see II 376 B note λογιστικόν, though as yet directed only to moral questions, is intellectual more than moral. Intellect gradually asserts its predominance over will until in Books VI and VII it achieves its final triumph. Cf. 439 E, 441 E notes

τὸ δὲ δὴ τοῦ θυμοῦ κτλ. Hitherto θυμοειδές has been chiefly the source of courage and the natural antithesis of φιλόσοφον (II 375 A ff., III 410 D, 411 C). It now enters on a wider sphere as the ally of λογιστικόν, and becomes, thus far, more intellectual, as Krohn points out: note also the ὀρθὴ δόξα of 430 B. Its ethical connotation is also intensified; for it is not now simply spirit, but the sentiment of moral indignation at everything evil— “ein edler Unwille über alles Schlechte” (Krohn Pl. St. p. 55)—everything which tends to destroy the πολιτεία ἐν ἡμῖν. It becomes in short, as Brandt ( Zur Entwick. d. Pl. Lehr. v. d. Seelentheilen p. 18) says truly enough though ponderously, “leidenschaftlicher Selbsterhaltungs- und Selbstvervollkommnungstrieb.” Cf. Simson der Begriff der Seele bei Plato p. 110, and see also on II 375 A.

ἴσως κτλ. The θνητὸν εἶδος ψυχῆς of the Timaeus includes both the θυμοειδές and the ἐπιθυμητικόν: see 69 C ff. and cf. Pol. 309 C . Similarly in the Phaedrus the two lower faculties are figured as the two horses, and the highest as the charioteer of the soul's chariot (253 D): cf. Simson l.c. p. 109 notes

ποτὲ ‐‐ τούτῳ . The antecedent of τούτῳ is τι: ‘having once heard something I trust to this,’ i.e. ‘I rely on an incident which I once heard.’ πιστεύω means that he relies on it for a proof; and ὡς ἄρα goes with ἀκούσας. So Schneider correctly explains the Greek. The precise force of πιστεύω τούτῳ has, I think, been missed by most of those who have suspected corruption. For τι there have been various conjectures: ἔτι (Madvig), ἄρτι (Liebhold Fl. Jahrb. 1888 p. 110), τινος (Zeller Archiv f. Gesch. d. Phil. II p. 694)—all superfluous, and the first two very weak; while Campbell suggests that οὐ has dropped before πιστεύω, taking τούτῳ to refer to Glauco's suggestion. But in that case τοῦτο would be necessary.

Αεόντιος . “Ad hunc Leontium eiusque insanam cupiditatem spectat depravatissimus Theopompi comici Καπηλίδων locus” (Herwerden Mn. N.S. XI p. 346). The fragment is emended by Kock ( Com. Att. Frag. I p. 739) into Αεωτροφίδης ὁ τρίμνεως ( trium librarum homo, i.e. levissimus) Αεοντίῳ | εὔχρως τε φαίνεται χαρίεις θ’ ὤσπερ vεκρός. Bergk was the first to connect the two passages.

ὑπὸ ‐‐ ἐκτός : ‘close to the outer side of the North wall.’ Cf. (with Stallbaum) Lys. 203 A τὴν ἔξω τείχους ὑπ’ αὐτὸ τὸ τεῖχος. The North wall was the outer of the two walls connecting Athens with the Piraeus; the other, or South wall, was called τὸ διὰ μέσου τεῖχος, because it lay between the βόρειον and the φαληρικόν, which connected Athens and the Phalerum. See Gorg. 455 E and the other authorities cited by Milchhöfer Schriftquellen zur Topographie von Athen pp. CXIII ff., and Curtius u. Kaupert Atlas von Athen Bl. II.

παρὰ ‐‐ κειμένους : ‘lying by’ or ‘near the executioner’; not of course ‘at the executioner's’ as has been suggested. When seen by Leontius, the hangman was engaged in throwing the bodies into the pit ( ὄρυγμα or βάραθρον, from which he was often called ὁ ἐπὶ s. πρὸς τῷ ὀρύγματι). The βάραθρον into which the bodies of executed criminals were thrown, was a deep ravine outside the walls, in the deme Κειριάδαι. Leontius would pass near it, just before entering the city (probably by the Μελιτίδες πύλαι): see Curtius u. Kaupert l.c. Bl. II. The place is still pointed out to visitors to Athens on the western declivity of the Hill of the Nymphs. For the ancient authorities see Milchhöfer l.c. pp. I—II. Various suggestions have been made for δημίῳ. Valckenaer's δημιείῳ is a coinage of his own, and otherwise objectionable; Αυκείῳ (also Valckenaer) is topographically impossible, and so is Διομείῳ (Hemsterhuis), if it has anything to do with the Διομηΐς πύλη. The explanation which I have given seems also to have been held by Milchhöfer, for he quotes the present passage among the authorities for the βάραθρον .

ὦ κακοδαίμονες . ‘Confound you!’

τὴν ὀργήν . q reads τὸν θυμόν, which Ast and others have preferred. But, as Schneider observes, ὀργή is to θυμός, as ἐπιθυμίαι to ἐπιθυμητικόν. If anger fights with desire, the source of anger, θυμοειδές, must be different from that of desire, ἐπιθυμητικόν. This is the whole moral of the anecdote, which is intended to establish the difference between θυμοειδές and ἐπιθυμητικόν only, not also λογιστικόν.

ταῖς δ) ἐπιθυμίαις κτλ. αὐτόν is τὸν θυμόν. ἀντιπράττειν “ad singularem aliquam actionem referendum est, quam ratio suscipere eaque in re sibi repugnare prohibeat, quasi dictum sit: μὴ δεῖν τι πράττειν καὶ τοῦτο δρῶντα ἀντιπράττειν” (Schneider). The words γενομένου τοῦ τοιούτου refer to ταῖς— κοινωνήσαντα. The anacoluthon is an easy one. Plato means merely that θυμός does not unite with the desires against the reason. For αἱροῦντος λόγου cf. X 604 C note On other views on this passage consult App. IV.

ὅσῳ ‐‐ ᾐ . The restriction will be noted. It is not οἱ γενναῖοι who, as the saying is, hate those whom they have injured.

οὐκ ἐν τούτῳ κτλ. : ‘does not he then fume and chafe—and fight on the side of what he believes to be just—both at hunger and at cold and all such inflictions, and bide his ground and conquer, abating not his noble indignation, until he has either achieved his purpose, or perished, or has been called back and soothed by the reason within him, as a herdsman recalls his dog?’ The words καὶ διὰ τὸ πεινῆν—πάσχειν must be taken with ζεῖ τε καὶ χαλεπαίνει, but possibly καὶ ξυμμαχεῖ τῷ δοκοῦντι δικαίῳ has been displaced, and we should read ζεῖ τε καὶ χαλεπαίνει καὶ διὰ τὸ πεινῆν—πάσχειν, καὶ ξυμμαχεῖ τῷ δοκοῦντι δικαίῳ, καὶ ὑπομένων κτλ. νικᾷ is not merely ‘tries to conquer’ or ‘perseveres’ (Schneider), but ‘conquers,’ in spite of the pardonable inconsistency of this translation with τελευτήση— πραυνθῆ. τῶν γενναίων cannot mean “in the case of the noble’ (P. Shorey A. J. Ph. XVI p. 237), unless θυμός is the subject of λήγει, which is not, I think, the case. The meaning is caught the more readily by reason of οὐχ ὅσῳ ἂν γενναιότερος ᾖ in C, and we ought not to substitute ἀγανακτῶν or the like with Richards. See on the whole passage App. V.

καίτοι γε =‘and surely’ has no adversative force here. See Kugler de part. τοι etc. p. 18. Hartman emends, but see on I 331 E.

ἦ. See cr. n. εἰ in direct interrogation is unclassical, and ἐρωτῶ cannot be supplied. Nor can εἰ well be taken as conditional (with Stallbaum) and τὸ ποῖον as a sudden interruption. For the confusion of ει and η see Introd. § 5.

ἀρτίως . 439 E.

τίθεσθαι κτλ. : ‘defends the rational element.’ I have retained the accusative on the strength of CIA II 317. 9 λαβόντος τοῦ δήμου τὰ ὄπλα ὑπὲρ τῆς ἐλευθ <ερί> ας καὶ παρακαλοῦ <ν> τος καὶ τοὺς στρατιώτας τίθεσθαι πρὸς τὴν πόλιν. The inscription dates from about 280 B.C. Other editors read τοῦ λογιστικοῦ (with Ξ alone among the MSS), but do not cite any example of the phrase τίθεσθαι τὰ ὄπλα πρός τινος, although πρός with the genitive is common enough in similar expressions. Thucydides (II 2. 4) has παρά with the accusative like πρός here. The original meaning of the idiom was to take up a position in arms by the side of: see Schneider's Xen. Anabasis pp. 537—540 and the commentators on Thuc. l.c. For the metaphor cf. Arist. Pol. Ath. 8. 5. F. K. Hertlein (quoted in Hartman) also defends the accusative, citing Aen. Polior. 4. 3 ἐτίθεντο τὰ ὅπλα παρὰ τοὺς πολεμίους ὡς παρὰ φιλίους.

ἐὰν μὴ κτλ. See App. IV.

ἐκεῖ : ‘in the other place,’ viz. III 390 D. If Kühn is to be trusted, Galen (Vol. V p. 500) does not, as Hartman asserts, omit the word; and there is no good reason for suspecting corruption.

ὁμολογεῖται . ὡμολογεῖται (sic) q^{1}: ὡμολόγηται q^{2} (with Stob. Flor. 9. 64). The present, ‘we pretty well agree,’ is satisfactory enough.

ἀνδρείαν . See cr. n., and for the error in A cf. IX 573 B note

ἔχειν is intransitive, and not transitive as D. and V. suppose.

ὥσπερ ἐλέγομεν . III 411 E —412 A. This passage enables us to identify the λογιστικόν with the φιλόσοφον of Books II—III. See on 439 D, and cf. Krohn Pl. St. p. 57.

τὸ μέν : i.e. τὸ λογιστικόν, as τὸ δέ is τὸ θυμοειδές. As the subject to the participles is κρᾶσις, we see again that Plato did not intend ‘Music’ and Gymnastic each to affect one part of Soul exclusively. It is curious however that the participles here describe the effect of music only: for it is music ( not gymnastic) which ἐπιτείνει τὸ φιλόσοφον: see on ὄπως ἂν—προσήκοντος III 411 E. The partial ignoring of gymnastic in this passage is perhaps premonitory of the intellectualism of VI and VII: cf. on 439 D and E.

ἀνλεῖσα κτλ. : ‘slackening the other by soothing address, taming it,’ etc. The three participles are not coordinate, but παραμυθουμένη explains the action of ἀνιεῖσα. It is unnecessary to desert the best MSS (as I once did) and read ἀνιεῖσα, παραμυθουμένη καὶ ἡμεροῦσα with Ξ v and the older editors.

προστατήσετον κτλ. Bekker's emendation—see cr. n.—is now generally accepted. τούτω means λογιστικόν and θυμοειδές: so also in B below. On ὃ δὴ πλεῖστον etc. see II 379 C note

καλουμένων κτλ. καλουμένων is said because such pleasures are no true pleasures: cf. I 336 A note and (for the implication itself) IX 583 B ff., Phil. 36 C ff. On οὐκ αὖ see 426 E note The imagery of this passage suggests that the ἐπιθυμητικόν is a sort of θηρίον: cf. IX 588 E ff.

ὧν ‐‐ γένει : sc. ἄρχειν. “Dativus causam indicat, cur tertiae parti non conveniat duabus reliquis praeesse et imperare, eamque in ipsius genere et indole positam demonstrat” (Schneider). If this is the meaning, we should expect φύσει rather than γένει. Perhaps Plato wrote γενῶν (so q Flor. U, Stallbaum etc.): cf. γένη in 441 C. To προσῆκον Campbell prefers προσῆκεν, but the present ( προσῆκον sc. ἐστίν) is better here.

φυλαττοίτην . The two higher parts of soul are to be φύλακες both of the lower part and (in a different sense) ‘also’ ( καί) ‘of external enemies’: cf. III 415 D, E. Dobree's φυλαττοίσθην fails to give its proper force to καί before τοὺς ἔξωθεν. For φυλάττω used in this way cf. II 367 A οὐκ ἂν ἀλλήλους ἐφυλάττομεν μη ἀδικεῖν, ἀλλ’ αὐτὸς αὑτοῦ ἦν ἕκαστος φύλαξ.

ὑπὸ τοῦ λόγου . In this particular the analogy between the city and the soul is not quite exact, otherwise it would be the rulers in the city who prescribe τὸ δεινόν τε καὶ μή, whereas it is the legislator (see on ὁ νομοθέτης 429 C). This point is emphasized, perhaps unduly so, by Krohn ( Pl. St. p. 43). Unless Plato made the Deity the οἰκιστής of the soul, as the original legislator is of the city, it was impossible for him to avoid placing the λογιστικόν in a position of even greater authority than the rulers. In Books VI and VII the inequality is redressed by making the power of the Rulers in the city commensurate with that of λόγος in the soul: see VI 497 C note

ἦρχέν τε κτλ. : ‘ruled within him and issued these instructions.’ The imperfect is used because the instructions must be given before they can be obeyed by θυμοειδές, as described in the last sentence. J. and C. say that ἦρχε refers to 428 E; but Plato is not there speaking of the individual, only of the State. Although a reference to 439 C or 441 E is barely possible, it is much simpler to regard the imperfect as real, and not ‘philosophic.’ See above on III 406 E. Schneider, to judge from his translation, takes the same view. With σμικρῷ μέρει cf. Arist. Eth. Nic. X 7. 1177^{b} 34 ff. εἰ γὰρ καὶ τῷ ὄγκῳ μικρόν ἐστι (sc. τὸ κράτιστον τῶν ἐν αὐτῷ) κτλ.

αὖ κἀκεῖνο κτλ. αὖ καί has been interpreted (1) as implying that the θυμοειδές also has a sort of knowledge: cf. 429 C and 439 E note: (2) as ‘like the rulers in the State’: cf. 428 B ff. The first view is slightly more natural on linguistic grounds, but I think Plato would hardly have attributed ἐπιστήμη in any shape to the θυμοειδές. Probably (2) is right, for the analogy between the city and the soul is in Plato's mind all through this section: see 441 C, D, and 442 D.

ᾧ πολλάκις κτλ. : ‘in virtue of our oft-repeated maxim and in that way’: i.e. τῷ τὰ αὑτοῦ πράττειν. Ficinus seems to have read καὶ ὡς after λέγομεν. At first sight καὶ οὔτως appears to demand the insertion; but Plato is speaking with less formality and precision than in 441 C, D. The reading of Vind. E καὶ οὗτος (for καὶ οὔτως), i.e. ‘the individual, as well as the city,’ is attractive, but unnecessary. Hartman proposes ᾧ <πόλιν> πολλάκις <ἐ> λέγομεν, τούτῳ καὶ οὗτος ἔσται, a solution which will commend itself to few.

μή πῃ ‐‐ εἶναι . ‘Do we find Justice growing dimmer in any way? Does it appear something different from what it was discovered to be in the city?’ lit. ‘blunted, so as to appear’ etc. In the language of 434 D (to which Socrates' question refers) Justice has now ‘passed into’ the Individual; and no feature has been blunted, or lost its clearness of outline. We are therefore confirmed in our view of Justice, both civic and individual. Hartman would read ἀπήμβλυνται, taking ἡμῖν as ‘by us,’ but the present is more expressive, and (with ἡμῖν) represents us as in a certain sense spectators of the selfevolution of Justice: cf. ἐὰν μὲν ἡμῖν καὶ εἰς ἕνα ἰὸν τὸ εἶδος τοῦτο κτλ. 434 D. ἀπαμβλύνεται=‘retunditur’ (Schneider).

τὰ φορτικά . Plato tests his view of Justice by four criteria taken as it were de foro and turning on various popular associations of the word: cf. IX 573 B ff. Of these the first three are concerned with honesty and trustworthiness in public and private life; while the last ( μοιχεῖαι—ἀθεραπευσίαι) refers to morality in general, including the service of the gods. Taken together, they sum up the leading features of the perfect character, and shew that Plato's conception of private, as of political, Justice is in reality Righteousness or Moral Perfection, whereof the other virtues are the fruit. Plato's innovation lay in interpreting Righteousness as τὰ αὑτοῦ πράττειν, or rather in the peculiar meaning which he attached to this phrase: see on 434 C and infra 443 B note

παρακαταθήκην χρυσίου κτλ. Honesty and truthfulness were generally recognised as characteristic of the δίκαιος ἀνήρ: see the passages collected by Nägelsbach Nachhom. Theol. pp. 240—246.

τοῦτο αὐτόν . See cr. n. “Fortasse Plato τοῦτον αὐτό scripsit” (Schneider).

ἱεροσυλιῶν ‐‐ προδοσιῶν . See Nägelsbach l.c. pp. 293 ff., 298 f.

ἄπιστος ‐‐ κατἁ ὅρκους . εὐορκία was an indispensable element in Greek morality: see Nägelsbach l.c. p. 242, and the interesting monograph of Augustin Der Eid im Gr. Volksglauben u. in d. Pl. Ethik Elbing 1894.

μοιχεῖαι ‐‐ ἀθεραπευσίαι . Nägelsbach l.c. pp. 264 ff., 275 ff., 191 ff. The virtue of εὐσέβεια was commonly regarded as δικαιοσύνη ἡ τερὶ τοὺς θεούς (e.g. Euthyph. 12 E ), and εὐσέβεια is concerned with θεῶν θερατεία. See Euthyph. l.c. and cf. also the Stoic Zeno in D. L. VII 119 εἶναί τε τὴν εὐσέβειαν ἐπιστήμην θεῶν θεραπείας, and Sext. Emp. adv. Math. IX 123.

τέλεον κτλ. The language is suggested by Homer's οὐκ ὄναρ, ἀλλ’ ὕπαρ ἐσθλόν, ὅ τοι τετελεσμένον ἔσται ( Od. XIX 547). ὅ is a vague internal accusative: see on ἣν ᾠήθημεν in 434 D.

ἔφαμεν κτλ. The reference is to 433 A. On οἰκίζειν see III 407 B note For κινδυνεύομεν Hartman suggests ἐκινδυνεύομεν; but presents do not of course become imperfects in indirect.

τὸ δέ γε : ‘yes, but in point of fact.’ For τὸ δέ in this sense cf. I 340 D note τὸ δέ γε ἀληθές below expresses the same meaning more fully and emphatically.

δἰ ὃ ‐‐ οὕτως . The imperfect ὠφέλει (see cr. n.), ‘for which reason also it was of service to us,’ viz. in discovering the real or original justice, seems to me better than the present. See II 368 D ff. Plato is justifying himself for having taken so much trouble about a mere εἴδωλον; it was in order to learn the original through the copy. So also Hartman. The present could only mean ‘benefits the city’ (so Schneider, Rettig and others). Madvig, strangely enough, suspects the whole phrase. Civic Justice is an εἴδωλον of Justice in the soul as being its reflection in outward conduct. See also on 443 B ff. above.

τοιοῦτο takes its meaning from τὸ τὸν μὲν σκυτοτομικόν etc. ‘Justice was indeed something of this kind’ (i.e. a sort of τὰ αὑτοῦ πράττειν), but not περὶ τὴν ἔξω πρᾶξιν. The warning conveyed by εἶδος and τρόπον τινά in 433 A (where see note), 433 B and 432 E is now justified: for Justice is said to be περὶ τὴν ἐντὸς πρᾶξιν, and is therefore not, strictly speaking, that which we have called ‘Civic Justice.’

ὡς ἀληθῶς should be construed with περὶ ἑαυτόν. The soul is the true self, as Socrates continually maintained. It is better to regard περί before ἑαυτόν as coordinate with περί in περὶ τὴν ἐντός, than to translate “with internal actions which are in very truth concerned with himself” (J. and C.). ὡς ἀληθῶς περὶ ἑαυτόν etc. merely emphasizes and explains περὶ τὴν ἐντός.

ἕκαστον . Ast would read ἕκαστον τῶν; but the meaning is easily caught after τὰ ἑαυτοῦ just before.

τῷ ὄντι κτλ. : ‘having set his house in order in the truest sense.’ So Schneider, rightly. For οἰκεῖα cf. III 405 B note

ξυναρμόσαντα ‐‐ ἡρμοσμένον . Cf. 432 A, where a similar image is employed. The figure here is taken from the Octachord, the λογιστικόν being represented by the ὑπάτη or highest string (which gave out the lowest note), the ἐπιθυμητικόν by the νεάτη (an octave higher in pitch), and the θυμοειδές by the μέση or fourth. See Dict. Ant. II p. 195 or Gleditsch Die Musik d. Gr. p. 860. The single notes of a ἁρμονία could be called ὅροι because they were in reality terms in a proportion and depended on the relative length of the string: cf. Tim. 35 B , C. Hartman's correction of νεάτης, ὑπάτης, μέσης to νεάτην, ὑπάτην, μέσην is very attractive: for the genitives can only be explained as ὅρον νεάτης etc., and the effect is unpleasing, especially with ἁρμονίας coming between. Retaining the Greek nomenclature, we may translate: ‘having harmoniously joined together three different elements, just like three terms in a musical proportion or scale, lowest and highest and intermediate,’ etc. In ἄλλα ἄττα μεταξύ Plato indicates (as J. and C. observe) that his threefold division of soul may not be ‘strictly exhaustive’ (cf. VIII 548 D note). The missing faculties would thus correspond to the notes intervening between the ὑπάτη and μέση, and the μέση and νεάτη. It will be noted that the unity resulting is not that of unison, but that of a scale or mode. Nevertheless it is clear from the language used that the ἁρμονία which Plato describes is, as before, σωφροσύνη: cf. ἄρξαντα αὐτὸν αὑτοῦ with 431 A, B, φίλον γενόμενον with 442 C; κοσμήσαντα too suggests κοσμιότης, and the word σώφρονα itself is finally employed. Cf. 434 C note A different explanation is given by the Scholiast. Holding that Plato is referring to a system of two octaves ( δὶς διὰ πασῶν) he explains νεάτη, μέση and ὑπάτη as e.g. A', A, and b (not a, which is the προσλαμβανόμενος). His note is as follows: νεάτη ἤγουν νήτη ὑπερβολαίων ἡ ἐσχάτη χορδὴ τοῦ δὶς διὰ πασῶν συστήματος, ὑπάτη δὲ ἡ μετὰ τὸν προλαμβανόμενον (leg. προσλαμβανόμενον) φθόγγον πρώτη χορδὴ τοῦ αὐτοῦ τοῦ δὶς διὰ πασῶν συστήματος. μέση δὲ καὶ αὐτὴ ἤτοι φθόγγος ἢ χορδὴ ἡ τελευταία μὲν τοῦ πρώτου διὰ πασῶν, ἀρχὴ δὲ τοῦ δευτέρου, ὡς εἶναι τούτων κοινήν, ὡς Πτολεμαῖός τέ φησι καὶ οἱ ἄλλοι μουσικοί. But in the δὶς διὰ πασῶν σύστημα, the ὑπάτη is not σύμφωνος with the νήτη ὑπερβολαίων, although the προσλαμβανόμενος of course is (see Gleditsch l.c. p. 861 and Euclid Sect. Can. 10 ed. von Jan): so that according to the Scholiast there is a serious breach of συμφωνία. It seems to me quite clear that in ὥσπερ ὅρους τρεῖς—μέσης Plato is thinking of three ξύμφωνοι φθόγγοι, and in the single octave or διὰ πασῶν, the ὑπάτη, μέση or fourth, and νεάτη were σύμφωνα ἀλλήλοις: see Cleonid. Isag. Harm. 5 ed. von Jan. In 432 A also, Plato contemplates only a single octave: see note ad loc.

ἕνα ‐‐ πολλῶν . Cf. 423 D note and [ Epin.] 992 B ἐκ πολλῶν ἕνα γεγονότα. The phrase εἷς ἐκ πολλῶν is a sort of Platonic motto or text (like the φωναί of post-Aristotelian ethics).

οὕτω δὴ ‐‐ ἤδη : emphatic, as Hirzel points out ( Hermes VIII p. 393): for the just man will not take part in practical affairs until he has ordered his own soul aright. Cf. Alc. I 113 B ff., Ap. 36 C , and Xen. Mem. III 7. 9 , IV 3. 1.

ἢ καί . Stallbaum and others add περί (with Ξ) before πολιτικόν, but τι πράττῃ περὶ πολιτικόν τι is very unpleasing. πολιτικόν depends directly on πράττῃ and is equivalent to περὶ πόλιν. The slight variety of expression is easy and elegant after ἢ καί ‘aut etiam.’

ἐπιστήμην ‐‐ δόξαν . This is, as Krohn points out ( Pl. St. p. 68), the first precise and explicit separation of ἐπιστήμη and δόξα in the Republic. Each of them, however, is still concerned with conduct, and not, as in the end of V, with the theory of knowledge.

τυγχάνει ‐‐ ὄν =‘really is’: I 337 B note

ἀδικίαν . Now that we have discovered Justice, it is necessary to look for Injustice, in order that we may compare the two and decide the question at issue, viz. πότερον δεῖ κεκτῆσθαι τὸν μέλλοντα εὐδαίμονα εἶναι, ἐάν τε λανθάνῃ ἐάν τε μὴ πάντας θεούς τε καὶ ἀνθρώπους (427 D: cf. II 368 E note). The full exposition of Injustice is reserved for Books VIII and IX, where Plato takes the subject in its proper order, considering civic injustice first, and afterwards that of the individual. At present he contents himself with a preliminary or exoteric sketch of Injustice in the soul, representing it as unrighteousness in general, just as Justice, both in the State and in the individual, has been identified with righteousness or moral perfection (434 C, 442 E notes).

ἀλλὰ τοιούτου ‐‐ ὄντι . See cr. n. The reading of Ξ and other inferior MSS, which (in common with all the editors) I have printed above, seems to be an attempt to emend the older and more difficult reading preserved in Α and Π. Stallbaum supposes that A here represents a corruption of Ξ, αὖ δουλεύειν being presumably a correction (of αὐτῳ δουλεύειν) which has crept into the text; but this is unlikely in itself, and also leaves τοῦ δ) before αὖ δουλεύειν unexplained. The text of Ξ is not in itself quite satisfactory, as Richards has pointed out. οἵου πρέπειν αὐτῷ for ὥστε πρέπειν αὐτῷ seems unexampled, although οἵου δουλεύειν would of course be right. The expression τῷ τοῦ ἀρχικοῦ γένους ὄντι, ‘that which is of the ruling class,’ is also curious for the more direct and accurate τῷ ἀρχικῷ γένει. The reading of A and Π yields no tolerable sense, and certainly cannot come from Plato. Madvig (with Vind. E) proposes οἵου πρέπειν αὐτῷ δουλεύειν, τὸ δ’ αὖ <μὴ> δουλεύειν ἀρχικοῦ γένους ὄντι, which is intelligible, if weak. I have thought of οἵου πρέπειν αὐτῷ δουλεύειν, τῷ δ’ αὖ βουλεύειν (or δεσπόζειν, after Schneider) ἀρχικοῦ γένους ὄντι, but there are obvious objections. I should not be surprised if the whole clause ἀλλὰ —ὄντι, as it appears in A and Π, is only an attempt by some illiterate scribe to work out the antithesis of προσῆκον: lit. ‘being by nature such as to be proper for it to be a slave, and the slavery again to that which is of the ruling class.’ The clause, even as read in Ξ, adds nothing to the sense, and the references in τοιαῦτ’ ἄττα and τούτων just below are caught more easily without the obnoxious words. See 442 B ἄρχειν ἐπιχειρήσῃ ὧν οὐ προσῆκον αὐτῷ γένει. Cf. III 413 C note

τούτων : sc. τῶν γενῶν.

ταὐτὰ ‐‐ ταῦτα . μὲν οὖν corrects τοιαῦτ’ ἄττα: “immo haec eadem” (Schneider). For ταὐτά some prefer, with one MS of Stobaeus Flor. 9. 64, αὐτά (‘immo haec ipsa’), others τοιαῦτα, but there is not sufficient reason for deserting A.

δίκαια ‐‐ ἐμποιεῖ . Krohn ( Pl. St. p. 59) reminds us of Arist. Eth. Nic. II 1. 1103^{a} 34 ff. τὰ μὲν δίκαια πράττοντες δίκαιοι γινόμεθα κτλ. On the Socratic analogy between body and soul cf. II 380 B note

ὑγίειαν ποιεῖν . Ξ (with a few other MSS) reads ἐμποιεῖν; and Stallbaum and others adopt this reading. ποιεῖν, ‘to produce,’ is however satisfactory: cf. 422 A.

τὸ δὲ νόσον ‐‐ ὑπ) ἄλλου . Here and in Tim. 82 A ff. Plato adopts the Hippocratean theory of the origin of disease: see de nat. hom. VI p. 40 c. 4 Littré ὑγιαίνει μὲν οὖν μάλιστα, ὁκόταν μετρίως ἔχῃ ταῦτα (sc. αἷμα καὶ φλέγμα καὶ χολὴ ξανθή τε καὶ μέλαινα) τῆς πρὸς ἄλληλα κρήσιός τε καὶ δυνάμιος καὶ τοῦ πλήθεος, καὶ μάλιστα μεμιγμένα ᾖ. ἀλγέει δὲ ὁκόταν τι τουτέων ἔλασσον ἢ πλέον ᾖ ἢ χωρισθῇ ἐν τῷ σώματι καὶ μὴ κεκρημένον ᾖ τοῖσι ξύμπασιν (Poschenrieder die pl. dial. in ihr. Verhältnisse zu d. Hippokr. Schr. p. 37). Cf. also Pl. Symp. 186 D with Hug ad loc. On κατὰ φύσιν see 443 B note

κάλλος ‐‐ εὐεξία : with reference perhaps to Thrasymachus' statement in I 348 E f. that Injustice is καλόν and ἰσχυρόν.

τὸ δὴ λοιπὸν κτλ. ‘What remains for us now to enquire is whether,’ etc. For the position of ἡμῖν cf. that of πάντων in 445 B. Herwerden suggests τόδε or τόδε δή, neither of which is necessary.

πότερον αὖ κτλ. See I 354 B, C, and note on 444 A.

ἐάν τε λανθάνῃ . Cf. 427 D and II 367 E.

βελτίων ‐‐ κολαζόμενος . II 380 B note

ᾧ ζῶμεν . Cf. I 353 D τί δ) αὖ τὸ ζῆν; ψυχῆς φήσομεν ἔργον εἶναι; μάλιστά γε and note ad loc. βιωτὸν ἄρα ἔσται should not be made interrogative. The sentence means: ‘if life, which men deem unbearable when the bodily constitution decays, even when they are surrounded by every variety of food and drink and wealth and power, shall be, forsooth, when tumult and decay affect the constitution of the very principle whereby we live, worth living, if so be we do what we desire, and take no steps to escape from wickedness and injustice, and acquire justice and virtue.’ Life is not (says Plato) βιωτός to the guilty man who works his will; it may become so if he takes steps to rid himself of vice, i.e. ἐὰν διδῷ δίκην καὶ βελτίων γίγνηται κολαζόμενος. For the sentiment cf. Crit. 47 D , E, Gorg. 477 B —E, Prot. 313 A , B.

ὅσον ‐‐ σαφέστατα . “Quam certissime fieri potest” is Ficinus' rendering, with which Schneider and later editors agree, taking κατιδεῖν as explanatory of ἐνταῦθα. But it is hard to find another instance of ὅσον οἶόν τε, although καθ’ ὅσον οἷόν τε and ὅσον δυνατόν (Thuc. I 22. 2) occur. ὡς οἷόν τε is the almost invariable phrase. For ὅσον Stephanus proposed ὅθεν, Ast ὅπου. I think the meaning is ‘now that we have come far enough to be able most clearly to descry that these things are so,’ ἐνταῦθα being equivalent to ἐπὶ τοσοῦτον, and ὅσον οἷόν τε to ὅσον ἐλθόντας οἷόν τέ ἐστιν.

ἀποκνητέον . I have reverted to the MS reading. Bekker's emendation ἀποκμητέον is very attractive, but ἀποκνητέον gives excellent sense (cf. I 349 A), and there is no real reason why Glauco should repeat the word employed by Socrates (see on V 465 E); nor does there appear to be any instance in Greek literature of the verbal of ἀποκάμνω.

ἄξια θέας . Plato does not claim that his enumeration of degenerate commonwealths is complete. Cf. VIII 544 D.

ἓν ‐‐ κακίας . An old Pythagorean principle, whence the parade with which Plato announces it. See Arist. Met. I 5. 986^{a} 22 ff. (RP.^{7} § 55) and Eth. Nic. I 4. 1096^{b} 6 with Stewart's note.

εἴδη ἔχοντες : ‘having’ (i.e. as we should say ‘forming’) ‘specific kinds’: cf. VIII 544 D.

ἐγγενομένου ‐‐ ἀριστοκρατία . Knowledge, not number, is the criterion of good government: cf. Pol. 292 C . Hitherto however the rulers have always been represented as a plurality, and we have heard nothing of a king. In the later books (from V 473 C onwards) we often hear of kingship; and in IX 576 D (as Newman points out Politics of Aristotle I p. 413 note) the ideal city is called βασιλευομένην, οἵαν τὸ πρῶτον διήλθομεν. With the present passage cf. VII 540 D ἢ πλείους ἢ εἷς and IX 587 D, where the ἀριστοκρατικός and the βασιλικός are identified. The fact is, as Henkel has pointed out ( Stud. zur Gesch. d. gr. Lehr. v. St. p. 57), that “Kingship is only a form of Aristocracy throughout the whole political theory of antiquity, and rests on no distinct and independent basis of its own.” It must be regarded as exceptional when in the Politicus (302 C ff.), probably a later dialogue, Plato distinguishes between kingship and aristocracy and places aristocracy on a lower plane. See also Whibley Gk. Olig. pp. 15 ff.

τῶν ‐‐ νόμων . For the genitive cf. (with Stallbaum) Gorg. 514 A δημοσίᾳ πράξοντας τῶν πολιτικῶν πραγμάτων, and infra VI 485 B.

ἰδιωτῶν ‐‐ κατασκευήν : ‘the organization of the character of the individual soul.’ ψυχῆς was doubted by Ast; but cf. IV 445 C τοσοῦτοι κινδυνεύουσι καὶ ψυχῆς τρόποι εἶναι, and for the collocation of genitives VII 525 C αὐτῆς τῆς ψυχῆς ῥᾳστώνης μεταστροφῆς, VIII 544 D, 559 E, 560 B, Tim. 24 B and other cases in Kühner Gr. Gr. II p. 289. ψυχῆς τρόπου is practically a single word like ‘soul-character’ (“Seelenbeschaffenheit” Schneider).

σμικρὸν κτλ. explains ἐκτείνας τὴν χεῖρα: ‘paullo longius ab Adimanto, quam clandestinum colloquium requirebat’ or ‘paullo remotior, quam reliqui a suis vicinis,’ not “a little further away from Socrates than Adimantus” (J. and C.): for “cur propterea manum protenderit et Adimantum attraxerit, non apparet” (Schneider).

τί μάλιστα κτλ. ‘What particular thing is it that you decline to let off?’ ‘You,’ said he. ‘Because of what particular remark of mine?’ (lit. ‘because I said what in particular?’) There is not, as J. and C. suppose, a play on the two senses of τί μάλιστα — cur potissimum and quid potissimum: for it must be observed that ἀφήσομεν has no expressed object, and Socrates could not have known that it was intended to refer to him. The removal of the commas usually printed after ὅτι and εἶπον restores sense, I think, to the remainder of this passage. ἔτι for ὅτι (see cr. n.) can scarcely stand, for ἔτι ἐγὼ εἶπον cannot mean ‘I repeated’ (Jowett), nor can we read ἔτι, ἐγὼ εἶπον, τί μάλιστα ‘once more, said I’ etc. In none of the parallels hitherto cited does ἔτι mean merely ‘once more’ or ‘again.’ Those who print ὅτι, ἐγὼ εἶπον, τί μάλιστα (Stallbaum) mostly take ὅτι—τί μάλιστα as in I 343 A ὅτι δὴ τί μάλιστα; ἦν δ’ ἐγώ. Ὅτι κτλ. But in such cases (as Schneider points out) there must be a second ὅτι to introduce the answer, and here there is not.

ἐκκλέπτειν = ‘to cheat out of’ as in μὴ—ἐκκλέψῃς λόγον Soph. Trach. 436 f.: see Jebb ad loc.

κοινὰ τὰ φίλων . See IV 423 E, 424 A notes κοινὰ τὰ τῶν φίλων is preferred by Ast and Stallbaum (with two late MSS), but the shorter form is far more racy of the soil, and occurs also in Lys. 207 C , Laws 739 C (Schneider on IV 424 A).

καὶ ὅλῃν κτλ. : i.e. καὶ ἐξηγήσεσθαι or the like, supplied from μνησθήσεσθαι. The construction cannot (as J. and C. suggest) go back to μὴ οὖν παρῇς.

μέγα κτλ. καί after φέρειν=‘or rather’ (atque) as in ὀλίγου τινὸς—καὶ οὐδενός ( Ap. 23 A ). For γιγνόμενον Liebhold proposes γιγνομένην, but see on IV 427 D. The feminine would be awkward after πολιτείαν, and κοινωνίαν—παίδων, though grammatically feminine, is logically neuter.

ἄλλης ‐‐ πολιτείας is explained by ᾖα τὰς ἐφεξῆς ἐρῶν (449 A). Stallbaum makes a curious slip: “quoniam ad alias πολιτείας partes considerandas celeriter accedis.”

τίθετε . Glauco addresses both Polemarchus and Adimantus. There is no occasion to write τίθει (with Hartman).

οἶον ‐‐ εἰργάσασθε κτλ. Chiappelli ( Riv. di Filologia etc. XI p. 195) finds in this and the following sentences a vaticinium ex eventu of Aristophanes' Ecclesiazusae. But the word παρῆκα shews that the ἐσμὸς λόγων does not refer to swarms of adverse criticism, but merely to the topics which Socrates must now discuss. See App. I, and (on the subject in general) Laws 779 E.

τότε . IV 423 E.

ἃ νῦν ὑμεῖς κτλ. : ‘in appealing to these topics now you’ etc. παρακαλοῦντες means literally ‘calling to you’: “das ruft ihr nun herbei” (Schneider). This interpretation is in harmony with ἐπεγείρετε, and gives the right antithesis to ἐάσοι. Neither “excitantes” (Ast), nor “in disputationem vocantes” (Stallbaum) is quite accurate. J. and C. give two alternative renderings (1) “and in now calling in this fresh argument,” (2) “and in now urging me to this.” But the antecedent can only be ταῦτα.

χρυσοχοήσοντας κτλ. Socrates shudders at the swarm of λόγοι to be encountered. ‘Why,’ says Thrasymachus, ‘it was precisely to listen to λόγοι, and not to smelt ore for gold, that we came here.’ χρυσοχοεῖν is a proverbial expression said of those who neglect their proper duty for some more fascinating— if less profitable—pursuit. Cf. Harpocr. s.v. χρυσοχοεῖον: Δείναρχος ἐν τῷ κατὰ Πυθέου· πάλιν παρ’ Αἰσχίνην ἀποφοιτήσας παρὰ τούτῳ δῆλον ὅτι χρυσοχοεῖν ἐμάνθανεν, ἀλλ’ οὐ τὸ προκείμενον αὐτὸ ποιεῖν ἢ πάσχειν. Here τὸ προκείμενον is λόγων ἀκούειν. The origin of the proverb is thus explained. A heap of gold-dust having been discovered on Hymettus, the Athenian populace deserted their usual avocations, and sallied out to seize it. But as it was guarded ὑπὸ τῶν μαχίμων μυκτήρων (cf. Hdt. III 102 ff. with the parallels cited by Stein), they failed. On returning ἔσκωπτον ἀλλήλους λέγοντες ‘σὺ δὲ ᾤου χρυσοχοήσειν.’ Cf. Suidas s.v. and Leutsch u. Schneidewin Paroem. Gr. I p. 464, II pp. 91, 727. A gloss in Bekker's Anec. Gr. I p. 316 (cited by Schneider) explains χρυσοχοεῖν in Dinarchus as proverbial for πορνεύειν; but it cannot have so offensive a meaning here, for (among other reasons) Thrasymachus and Socrates are now reconciled. Ast's explanation “aurum fundere proverbialiter dicitur, quem magna, quam animo conceperat, spes frustratur” expresses only one side of the proverb: the other—neglecting the duty which lies nearest—is more important and relevant here. “To find an Eldorado” (Warren) may perhaps meet the case. Thomas Gray's explanation is not altogether right: “a proverbial expression used of such as are idly employed or sent (as we say) on a fool's errand.”

μέτρον δὲ κτλ. An argumentum ad hominem, for the sentiment is Socratic: cf. VI 504 C. δέ γε=‘yes, but’ helps to bring out this point. ἀκούειν is the common epexegetic infinitive: cf. III 407 B note To insert τοῦ before τοιούτων (with Herwerden and Richards) is both unnecessary and inelegant.

τὸ μὲν ἡμἑτερον ἔα : ‘never mind us’: we are equal to a long discourse (so also J. and C.).

τῷ μεταξὺ χρόνῳ . The interval between γένεσις and παιδεία is nowhere defined in the Republic: in Laws 794 C it is reckoned at six years. For the regulations applying to this period see infra 460 B—D, and cf. Laws 788 D ff.

οὖν . See cr. n. It is admitted that Π is independent of A, so that οὖν (which most MSS have) may well be right. The tendency to confuse οὖν and ἄν may help to explain A's variants οὔκουν and οὐκ ἂν οὖν in I 333 E. Baiter reads δή. The confusion of ἄν and δή occurs no doubt in MSS, but its frequency has been much exaggerated, as for instance in Cl. Rev. VI p. 338.

ὦ εὔδαιμον . Cf. IV 422 E note

μὴ εὐχὴ κτλ. For εὐχή=an impossible aspiration, a Utopian or chimerical proposal, cf. 456 C, VII 540 D and see Susemihl and Hicks on Arist. Pol. B 1. 1260^{b} 29.

ἀγνώμονες : i.q. ἀνεπιστήμονες, as explained (with reference to this passage) in Bekker's Anec. Gr. I p. 334: cf. Phaedr. 275 A . Hence φρονίμοις in E below. The more usual meaning, ‘inconsiderate,’ ‘unkind,’ is less suitable here on account of δύσνοι.

φίλων κτλ. φίλων though neuter is of course intended to balance φίλοις. The conjecture φιλτάτων (Richards, Hartman) destroys the balance and is in itself superfluous: see Kühner Gr. Gr. II p. 23. Note the characteristic chiasmus ἀσφαλὲς καὶ θαρραλέον—φοβερόν τε καὶ σφαλερόν.

ὀφλεῖν κτλ. The infinitive depends on φοβερόν, and is like the infinitive after φοβοῦμαι. In the antithetical clause Plato substitutes the more usual construction with μή. The future indicative ( κείσομαι) is rare after words of fearing (Goodwin MT. p. 132), and represents the danger as imminent. To regard οὔ τι γέλωτα ὀφλεῖν as a reference to the Ecclesiazusae is rash and unjustifiable: see App. I.

προσκυνῶ κτλ. The apology looks forward, and not backward; whence δέ rather than δή (which Herwerden would read).

Ἀδράστειαν . Adrasteia was originally, perhaps, a personification of ἀνάγκη in its relation to humanity and the issues of human conduct. This meaning survived in the Orphic theology (Abel Orph. Fr. 36, 109—111) and appears in Phaedr. 248 C . Specifically, she was viewed as a variety of Nemesis, θεά τις τοὺς ὑπερηφάνους τιμωροῦσα (Schol. on Aesch. Prom. 936 ), and in this sense Aeschylus (l.c.) writes οἱ προσκυνοῦντες τὴν Ἀδράστειαν σοφοί (the first mention of Adrasteia in Greek literature). Adrasteia is in a still more special sense the punisher of proud words; so that προσκυνῶ Ἀδράστειαν becomes, as here, a sort of apologetic preface to a bold assertion or rash utterance: cf. Eur. Rhes. 342 , 468 ( ξὺν δ’ Ἀδραστείᾳ λέγω). See Nägelsbach Nachhom. Theol. p. 47 and Seymour in the Proceedings of the Amer. Philol. Assoc. for July 1891 pp. XLVIII ff.

ἐλπίζω κτλ. ἐλπίζω is ‘I fancy,’ not ‘I expect’: cf. II 383 B note The omission of εἶναι is curious: Madvig would restore it after ἁμάρτημα. I can find no parallel to its omission with ἐλπίζω, but οἴομαι, ἡγοῦμαι and other verbs of thinking often dispense with it. For examples see Schanz Nov. Comm. Pl. p. 34.

καλῶν κτλ. : “concerning noble and good and just institutions” (D. and V.), not “about the beautiful, the good, and the just, in the matter of laws” (J. and C.). The latter explanation gives a good sense, but it is harsh to separate δικαίων from νομίμων, and still harsher to take καλῶν as equivalent to περὶ καλῶν. Schneider was inclined to treat δικαίων as a gloss on νομίμων. But ‘about things beautiful and good and institutions’ is an anti-climax; and, besides, it is of institutions in conjunction with, not as distinct from, justice etc. that Plato is about to speak. In his translation Schneider takes the right view.

εὖ . q has οὐκ εὖ, an obvious but audacious correction, suggested, no doubt, by καλῶς εἶχεν ἡ παραμυθία in 450 D. εὖ is ironical. Glauco had comforted Socrates by saying inter alia that his hearers were friendly ( οὔτε δύσνοι οἱ ἀκουσόμενοι 450 D). Excellent comfort! says Socrates: I had rather, in the circumstances, that they were enemies! Stallbaum and others read οὐκ εὖ, and Hermann οὐ, for εὖ, thinking the irony misplaced; but Glauco's smile ( γελάσας) favours the ironical interpretation, and so does the ‘Socratic irony’ with which the whole sentence is overflowing. I agree with J. and C. in rejecting the pointless alternative rendering ‘you do well to comfort me.’

ὥσπερ φόνου κτλ. See cr. n. καί before καθαρόν is absent from the great majority of MSS and can scarcely, I think, be sound: for the difference in meaning between καθαρόν and μὴ ἀπατεῶνα is hardly enough to carry off the double καί. ὥσπερ belongs to the whole expression φόνου καθαρόν, which is virtually one word. Hartman would expunge καὶ μὴ ἀπατεῶνα ἡμῶν, but it is quite in Plato's way to subjoin the interpretation of a metaphor or simile (cf. 470 C, VIII 553 D, 555 D, and my note on Prot. 314 A ), nor have we any right to excise such expressions wholesale, as many Dutch critics would do (especially J. J. Hartman de embl. in Pl. text. obviis 1898).

ἐκεῖ : viz. in cases of φόνος ἀκούσιος (so Schneider, Stallbaum, etc.), not (with D. and V.) ‘in the next world.’ κἀνθάδε is relevant only if it means ‘in this case too,’ i.e. ἐν τῷ ἀπατεῶνα εἶναι καλῶν τε καὶ ἀγαθῶν κτλ.: and this fixes the meaning of ἐκεῖ.

ὡς ὁ νόμος λέγει is explained by Dem. πρὸς Πανταίνετον 58 καὶ γὰρ ἀκούσιοι φόνοι—καὶ πολλὰ ἄλλα τοιαῦτα γίγνεται· ἀλλ’ ὅμως ἁπάντων τούτων ὅρος καὶ λύσις τοῖς παθοῦσι τέτακται τὸ πεισθέντας ἀφεῖναι, and ib. 59. See also Laws 869 E.

ἀνδρεῖον δρᾶμα κτλ. There is probably a playful allusion to the mimes of Sophron, as was first pointed out by R. Förster in Rhein. Mus. XXX (1875) p. 316. According to Suidas (s.v. Σώφρων) and others, Sophron's mimes were classified as ἀνδρεῖοι μῖμοι and γυναικεῖοι μῖμοι. In the former, as may be inferred from Choricius' Defence of Mimes (first published by Graux in Revue de Philologie 1 pp. 209 ff.) Sophron represented male characters, in the latter female ( μιμεῖται μὲν ἄνδρας, μιμεῖται δὲ γύναια ib. p. 215). This is corroborated by many of the titles of his plays, such as ὁ ἀγροιώτας, ὁ θυνναθήρας, ὁ ἄγγελος contrasted with ταὶ ἀκ<*>στρίαι, ἁ νυμφόπονος, ἁ πενθερά etc. Sophron's mimes are called δράματα (cf. ἀνδρεῖον δρᾶμα) by Demetrius περὶ ἑρμηνείας § 156 σχεδόν τε πάσας ἐκ τῶν δραμάτων αὐτοῦ τὰς παροιμίας ἐκλέξαι ἐστίν. The point here is that just as custom required an ἀνδρεῖος μῖμος to precede a γυναικεῖος—this is not otherwise attested, so far as I can discover—, so it will be proper ( ὀρθῶς ἂν ἔχοι) for Plato's women to come on the stage after his men have played their part. Plato's partiality for Sophron is frequently mentioned by ancient authors, as for example by D. L. III 18, Quintil. I 10. 17: see Schuster in Rhein. Mus. XXIX (1874) pp. 605 ff., where these and other authorities are cited. Susemihl (Bursian's Jahresbericht 1874—1875 III p. 343) doubted whether Plato has Sophron in view here; but the allusion, which was admitted by Graux (l.c. p. 215 note), and successfully reaffirmed by Förster ( Rhein. Mus. for 1880 p. 472), is highly probable. I can see no point in making δρᾶμα γυναικεῖον an ironical reference to the Ecclesiazusae of Aristophanes (with Munk die nat. Ordnung d. Pl. Schr. p. 296, and Chiappelli l.c. p. 196), nor is it likely that the words allude to a dramatic caricature of Plato's policy by some other comedian, as is supposed by Bergk Gr. Literaturgesch. IV p. 462 note 134. On Sophron's prose-mimes as a preparation for the Socratic Dialogue see Hirzel der Dialog I pp. 20—26.

κατ’ ἐκείνην κτλ. : “in following out that original impulse which we communicated to them” (D. and V.).

ὡρμήσαμεν (sc. αὐτούς) is causative, and not intransitive, as Jowett supposes.

ἀγέλης . Cf. II 375 D and infra 460 C, 466 D notes

πλὴν κτλ. One MS inserts ταῖς μέν after πλήν; but, “ ταῖς θηλείαις utpote ex ipsa sententia et ex adjectivo ἀσθενεστέραις facile intelligendum enuntiatum non est” (Schneider). Schneider's explanation is more accurate than to say (with Stallbaum) that ταῖς μέν is idiomatically suppressed, like τὸ μέν before ἄλλο in Prot. 330 A ( ἄλλο, τὸ δὲ ἄλλο): cf. infra 455 E. This passage is thus criticised by Aristotle ( Pol. B. 5. 1264^{b} 4) ἄτοπον δὲ καὶ τὸ ἐκ τῶν θηρίων ποιεῖσθαι τὴν παραβολήν, ὅτι δεῖ τὰ αὐτὰ ἐπιτηδεύειν τὰς γυναῖκας τοῖς ἀνδράσιν, οἷς οἰκονομίας οὐδὲν μέτεστιν. But, from Plato's point of view, the analogy holds; for he regards οἰκονομία as παρὰ φύσιν even for human beings, and aims at abolishing it.

μουσικὴ μέν . The particle μέν “Latino atqui non multo debilius” (Schneider, comparing I 339 B and III 412 C ὅτι μὲν πρεσβυτέρους τοὺς ἄρχοντας δεῖ εἶναι—δῆλον). Richards conjectures μήν, which would certainly be more usual (cf. 465 B): but no change is necessary. Although the position of τε (which a few inferior MSS omit) is irregular, we ought not to read γε: cf. infra 465 E note, and (with Schneider) Laws 800 A, 966 A ( ἢ καὶ ὅπως ἕν τε καὶ ὅπῃ). In these cases τε suffers hyperbaton, being attracted forward by καί. The reverse kind of hyperbaton is more usual with this word: see Prot. 316 D , with my note ad loc. Here it would be awkward to place τε after either μουσική or μέν. For ἐδόθη Richards proposes ἀπεδόθη, to correspond with ἀποδιδῷς above; but cf. δοτέον infra 457 A and see on I 336 E.

παρὰ τὸ ἔθος : ‘contra consuetudinem,’ not ‘respectu consuetudinis’ as Hartman thinks. The phrase specifies the particular variety of γελοῖα intended by Plato: ‘many ludicrous breaches of etiquette.’ It is not quite easy (with J. and C.) to understand πραττόμενα.

πράξεται . q has πεπράξεται, which is tempting, and may be right; but, as Schneider points out, ‘si peragentur’ is somewhat more appropriate than ‘si peracta fuerint.’ πράξεται as passive seems to occur only here in Attic.

ἤδη = demum adds emphasis to ἀλλὰ καί. We may translate: ‘but positively also the older women.’ On this use of ἤδη (‘now that we have reached this point’) and kindred words see Cope Aristotle's Rhetoric Vol. I pp. 13 ff. J. and C. (with other editors) suppose a hyperbaton for τὰς ἤδη πρεσβυτέρας (which Herwerden would actually read): but the hyperbaton is harsh, and no parallel has yet been adduced. The rules laid down by Plato in this passage are an exaggeration of Spartan usage: cf. Plut. Lyc. 14 and the passages cited by Paley on Eur. Androm. 596 ff. Σπαρτιάδων —αἳ ξὺν νέοισιν ἐξερημοῦσαι δόμους | γυμνοῖσι μηροῖς καὶ πέπλοις ἀνειμένοις | δρόμους παλαίστρας τ’ οὐκ ἀνασχέτους ἐμοὶ | κοινὰς ἔχουσι, and by Blaydes on Ar. Lys. 82 : cf. also Laws 813 E ff., 833 C ff. and infra 457 A. The words ὅταν ῥυσοὶ— φιλογυμναστῶσιν are a characteristically Hellenic touch: cf. Theaet. 162 B .

τῶν χαριέντων . It is tempting to see in this an allusion to the author of the Ecclesiazusae (with Krohn Pl. St. p. 81 and Chiappelli Riv. di Filol. XI p. 198). If —with the majority of modern critics— we hold that the Ecclesiazusae is earlier than Book V, and if we consider the play as at least in some measure directed against theories on communism and the position of women with which the Socratic school sympathised, it is easy to interpret Plato here as addressing a rebuke to the comic stage in the form of a further challenge. In any case, however, the words οὐ φοβητέον—ὀχήσεις are not a vaticinium ex eventu, for the Ecclesiazusae does not touch on any of the points specifically mentioned here. See also on 452 D, 455 A, 457 B, 464 B, and 473 E f. In each of these passages there is some prima facie ground for suspecting a personal or polemical motive of some kind. See on the whole subject App. I.

τὰ αὑτῶν πράττειν : i.e. παίζειν. Herwerden's conjecture τὰ τοιαῦτα παίζειν is both needless and inelegant.

οὐ πολὺς χρόνος κτλ. Stallbaum cites Hdt. I 10 παρὰ γὰρ τοῖσι Λυδοῖσι, σχεδὸν δὲ καὶ παρὰ τοῖσι ἄλλοισι βαρβάροισι, καὶ ἄνδρα ὀφθῆναι γυμνὸν ἐς αἰσχύνην μεγάλην φέρει, and Thuc. I 6 ἐγυμνώθησάν τε πρῶτοι ( Λακεδαιμόνιοι ) κτλ.

γυμνασίων is used in its strict etymological sense of γυμνοὶ ἀγῶνες: we ought not to insert γυμνῶν (with Richards) or τοιούτων (with Herwerden) before γυμνασίων.

πρῶτοι ‐‐ Κρῆτες . Plato contradicts Thucydides l.c. Cf. [ Minos] 318 D, where Spartan institutions are derived from Crete, and see Hermann-Thumser Gr. Staatsalt. p. 141 notes

καὶ κτλ. καί begins the apodosis: ‘then too’ etc. The general idea is that when experience proved that it was better to take exercise in a nude condition, nudity also ceased to be ludicrous. Plato thus prepares the way for the identification to be presently made (see next note). The particle δή (‘forsooth’) hints that the eye is less trustworthy than the reason; and the contrast is further accentuated by the somewhat artificial balance between ἐν τοῖς ὀφθαλμοῖς and ἐν τοῖς λόγοις. D. and V. wrongly make the apodosis begin with καὶ τοῦτο (where τοῦτο is of course nominative).

μάταιος κτλ. I have (with the Oxford editors) retained the text of A. It at least affords an intelligible sense, and none of the numerous variants or emendations is at all convincing. The general drift of the passage is clear enough. Nothing is γελοῖον except what is κακόν ( μάταιος—κακόν), and, conversely, nothing is σπουδαῖον except what is ἀγαθόν ( καὶ καλοῦ—ἀγαθοῦ. σπουδαῖον is involved in σπουδάζει). Both inferences are expressed in such a way as to suggest a personal reference: cf. χαριέντων in B, and see App. I. γελωτοποιεῖν, especially after κωμῳδεῖν just above, points to the comic stage: and Aristophanes is perhaps intended. See on 452 B. The whole sentence means: ‘Foolish is the man who identifies the laughable with anything but the bad, and he who attempts to raise a laugh by looking at any spectacle as laughable except the spectacle of folly and evil aims in all seriousness also at another standard of beauty, which he has set up for himself, than the standard of the good.’ The analysis of τὸ γελοῖον, so far as it goes, is in harmony with Phil. 48 A ff.: cf. especially 49 A. With στησάμενος we must supply αὐτόν, i.e. τὸν σκοπόν. On the difficulties of this passage see App. II.

ἡ ἀνθρωπίνη was objected to by Cobet; but ἡ θήλεια alone would be too general: we are dealing only with ‘female human nature.’

καὶ κάλλιστα . Dobree conjectured κάλλιστα καί, neatly, but needlessly, for καλὴ τελευτή, like καλὴ ἀρχή, may be treated as a single notion. Cf. III 404 B note

κατοικίσεως : sc. τῆς πόλεως, but the antecedent is attracted into the relative clause ( ἣν ᾠκίζετε πόλιν), as often: cf. I 350 C note

ὡμολογεῖτε . II 369 E ff.

οὐ γὰρ εὐκόλῳ κτλ. The MS reading has been defended in two ways. Schneider prints a colon after ἔφη, and explains οὐ μὰ τὸν Δία as “minime, per Jovem, ”; but it is exceedingly difficult to supply the words in brackets. This difficulty induced Apelt ( Obs. Crit. p. 12) to suggest οὐ <μάτην> μὰ τὸν Δία, ἔφη· οὐ γὰρ κτλ. Others explain the oath as emphasizing οὐ γὰρ εὐκόλῳ ἔοικεν, and compare X 605 E οὐ μὰ τὸν Δἴ, ἔφη, οὐκ εὐλόγῳ ἔοικεν and Parm. 131 E οὐ μὰ τὸν Δία, φάναι, οὔ μοι δοκεῖ εὔκολον εἶναι τὸ τοιοῦτον διορίσασθαι. But the whole difficulty centres round γάρ, and γάρ is absent from each of these passages. Hartman strangely explains γάρ as ‘profecto’; while Stallbaum inclines to cut it out. Groen van Prinsterer ( Prosop. Plat. p. 209) proposed to read οὐ γὰρ εὐκόλῳ ἔοικεν. Οὐ μὰ τὸν Δία, ἔφη. Οὐ γάρ, εἶπον. It appears to me that the emphatic οὐ μὰ τὸν Δία is more appropriate in the mouth of Socrates, who is continually dwelling on the difficulty of his task, and I therefore think that Plato wrote Οὐ γὰρ εὐκόλῳ ἔοικεν, ἔφη. Οὐ γάρ, εἶπον, οὐ μὰ τὸν Δία, although I have not ventured to change the text. εὐκόλῳ is of course neuter, not masculine, as Richter supposed ( Fl. Jahrb. 1867 p. 143).

κολυμβήθραν : a swimming tank. See Blümner Privatalt. p. 210 note 2. In what follows we have the first suggestion of the wave metaphor, which dominates nearly the whole of Book V: see on 449 A.

ἄπορον . As ἄλλην here means ‘other’ and not ‘else,’ the epithet ἄπορον (‘difficult to procure,’ cf. II 378 A) must be applicable to the dolphin also. The Platonic litotes seems delicately to suggest that the miraculous story of Arion and the dolphin is not above suspicion. Herwerden conjectured ἄτοπον, but no change is necessary.

κατηγορεῖτε . Socrates identifies his audience with the imaginary opponents of 453 A—C, and Glauco replies in their name. As ἡμῶν means primarily Socrates and Glauco (453 B), the situation is somewhat confusing: and some may wish to read κατηγορεῖται, as I formerly printed (with Vind. F, Flor. R T, Ficinus and Hartman). The confusion of ε and αι is of course common (see Introd. § 5), but it is better to adhere to the best MSS. Cf. VI 489 B.

ἀντιλογικῆς τέχνης . ἀντιλογική is defined in Soph. 225 B as a variety of ἀμφισβητητικόν: viz. τὸ ἐν ἰδίοις—opposed to τὸ δικανικόν, which is δημοσίᾳ—αὖ καὶ κατακεκερματισμένον ἐρωτήσεσι πρὸς ἀποκρίσεις. It is described in Phaedr. 261 D ff., and practical illustrations are given in the sophisms of Euthyd. 275 C ff. The Ἀντιλογικοί are spoken of as almost a distinct sect in Plato's time: see Lys. 216 A and Isocr. περὶ ἀντιδόσεως 45 ἄλλοι δέ τινες περὶ τὰς ἐρωτήσεις καὶ τὰς ἀποκρίσεις—οὓς ἀντιλογικοὺς καλοῦσιν. Here Plato probably has in view some of the ‘Sophists’ (as in VI 499 A) as well as the Megarian school, whose well-known puzzles— ὁ ψευδόμενος, ὁ διαλανθάνων, Ἠλέκτρα, ὁ ἐγκεκαλυμμένος: see D. L. II 108—are excellent examples of verbal fallacies. The same class of people are also called ἐριστικοί and ἀγωνιστικοί: see Men. 75 C and cf. Theaet. 167 E , Phil. 17 A and Isocr. in Soph. 20 τῶν περὶ τὰς ἔριδας καλινδουμένων—τοιαῦτα λογίδια διεξιόντες οἷς εἴ τις ἐπὶ τῶν πράξεων ἐμμείνειεν, εὐθὺς ἂν ἐν πᾶσιν ε<*>η κακοῖς. On the history and place of Eristic in Greek philosophy see E. S. Thompson's elaborate excursus in his edition of the Meno pp. 272—285.

κατ) εἴδη διαιρούμενοι . εἴδη is not of course ‘the Ideas’: but ‘species’ ‘kinds’: cf. Pol. 285 A κατ’ εἴδη— διαιρουμένους and Soph. 253 D κατὰ γένη διαιρεῖσθαι. That κατὰ γένη (s. εἴδη) διαλέγειν is the peculiar province of dialectic was the iew of Socrates as well as of Plato: see Xen. Mem. IV 5. 12 ἔφη δὲ καὶ τὸ διαλέγεσθαι ὀνομασθῆναι ἐκ τοῦ συνιόντας κοινῇ βουλεύεσθαι διαλέγοντας κατὰ γένη τὰ πράγματα. See also on III 402 C.

κατ) αὐτὸ ‐‐ ἐναντίωσιν : lit. ‘pursue the contradiction of what has been said according to the name and nothing more’ i.e. ‘aim at the merely verbal contradiction of what has been said.’ We are told by Clement ( Strom. II 7. 968 B ed. Migne) that Critolaus called such persons ὀνοματομάχοι. With διώκειν ἐναντίωσιν cf. III 410 B γυμναστικὴν διώκειν. The implied antithesis to κατ’ αὐτὸ τὸ ὄνομα is κατ’ αὐτὸ τὸ πρᾶγμα: cf. Soph. 218 C δεῖ δὲ ἀεὶ παντὸς πέρι τὸ πρᾶγμα αὐτὸ μᾶλλον διὰ λόγων ἢ τοὔνομα μόνον συνομολογ(ήσασθαι χωρὶς λόγου.

ἔριδι ‐‐ διαλέκτῳ : a common opposition: cf. Men. 75 C ff. and Phil. 17 A .

τὸ μὴ τὴν αὐτὴν κτλ. See cr. n. The omission of μή was perhaps due to the erroneous idea that διώκομεν below meant ‘to attack.’ In reality, it means ‘we are pursuing’ (the proposition that), i.e. ‘we are insisting that.’ The way for this somewhat strained use is prepared by διώκειν τὴν ἐναντίωσιν just above. Plato is in fact applying the expression τοῦ λεχθέντος τὴν ἐναντίωσιν to the special case before us. τὸ λεχθέν would in this case be that ‘different natures are to follow the same pursuits’ (453 E τὰς ἄλλας φύσεις τὰ αὐτά φαμεν νῦν δεῖν ἐπιτηδεῦσαι). Its ἐναντίωσις is that ‘different natures are not to have the same pursuits.’ For τὴν αὐτήν we must therefore read either <μὴ> τὴν αὐτήν or else τὴν ἄλλην (with Baiter). I prefer the former, both because it has some MS authority, and also because, if Plato had chosen to use ἄλλος, he would probably have written τὰς ἄλλας φύσεις as in 453 E. It is also true, as J. and C. observe, that “the opposition of μὴ τὴν αὐτήν, οὐ τῶν αὐτῶν is more like Plato than the conjectural reading τὴν ἄλλην.” Translate ‘we cling to the verbal point and insist that what is not the same nature ought not to have the same pursuits.’

ὡς ἔοικεν marks the irony. For ἡ ἐναντία in the next line a few MSS have ἐναντία, which Hartman approves. If ἡ αὐτή were predicative, Plato would have written ἐναντία, but, as it is, ἡ ἐναντία is correct, being, like ἡ αὐτὴ φύσις, the subject to an ἐστίν understood.

πρὸς ‐‐ τεῖνον corresponds to πρὸς τί τεῖνον in B above. On the corruption in A see Introd. § 5.

ἰατρικὸν κτλ. Plato is illustrating that particular variety of ὁμοίωσις and ἀλλοίωσις which πρὸς αὐτὰ τείνει τὰ ἐπιτηδεύματα. As an instance of ὁμοίωσις he gives two ἰατρικοί (cf. I 350 A): these clearly have the same nature πρὸς αὐτὰ τὰ ἐπιτηδεύματα, i.e. in this case πρὸς τὸ ἰατρεύεσθαι. ἀλλοίωσις he illustrates by the difference between an ἰατρικός and a τεκτονικός: these have different natures πρὸς τὰ ἐπιτηδεύματα, for the one is qualified ἰατρεύεσθαι, the other τεκταίνεσθαι. Nothing could be more clear; but the text has been plunged into confusion by the introduction of the words τὴν ψυχὴν ὄντα after the second ἰατρικόν. The reading of A—see cr. n.—is indefensible; and the majority of recent editors print ἰατρικὸν μὲν καὶ ἰατρικὸν τὴν ψυχὴν ὄντα with q. But τὴν ψυχὴν ὄντα adds nothing to ἰατρικόν. It has indeed been thought that ἰατρικόν by itself suggests a doctor in actual practice, whereas an ἰατρικὸς τὴν ψυχήν need not practise. If so, we may fairly doubt whether the two have the same nature; and at all events the difference between them renders them inapt illustrations of Plato's argument. Jowett and Campbell attempt to escape these difficulties by taking τὴν ψυχὴν ὄντα with the first ἰατρικόν as well as with the second; but the Greek does not permit of this solution. Similar objections apply to the readings of Bekker (and apparently Ficinus) ἰατρὸν μὲν καὶ ἰατρικὸν τὴν ψυχὴν ὄντα, of Stephanus and other early editors ἰατρικὸν μὲν καὶ ἰατρικὴν τὴν ψυχὴν ἔχοντα (partly supported by Θ), and also, with some modifications, to Richards' otherwise unhappy proposal ἰατρικὸν μὲν καὶ ἰατρικὸν <εὐφυᾶ> τὴν ψυχὴν ὄντα. Hermann reads ἰατρικὸν μὲν καὶ ἰατρικὴν τὴν ψυχὴν ὄντας, but the introduction of women is of course premature. I regard τὴν ψυχὴν ὄντα as a relic of ἰατρὸν τὴν ψυχὴν ὄντα, a marginal annotation on ἰατρικόν. Cf. Introd. § 5.

ἐλέγομεν : ‘we were saying,’ i.e. ‘we meant.’ Nothing of the sort was actually said before.

διαφέρον : ‘excelling’ rather than ‘differing’ (D. and V.): hence τοῦτο δὴ— ἀποδιδόναι. Richards proposes διαφέρειν, to avoid the singular. But the subjects are distributed, as appears from καὶ τό— καὶ τό, as well as from ἑκατέρῳ; and the infinitive is somewhat less suitable here than it is below. Translate ‘if either the male or the female sex plainly excels the other’ etc.

οὐκοῦν κτλ. ‘Is not our next step to invite?’ &c. Ξ reads κελεύωμεν, which may be right, but the indicative is quite defensible. With τὸν τὰ ἐναντία λέγοντα cf. infra 455 A. It is not likely that a specific allusion to Aristophanes is here intended (see Chiappelli Riv. di Filolog. XI p. 200), but there is some plausibility in the conjecture that the coming argument may be inspired in some measure by the Ecclesiazusae, where the essentially domestic qualities of women are contrasted with their incapacity for government. See App. I.

ὀλίγον πρότερον . 453 C.

τοῦ ‐‐ ἀντιλέγοντος . 454 E note

ἢ μακρολογῶμεν κτλ. Socrates is unwilling to bore us ( μακρολογεῖν) by enumerating the exceptions, which are —he implies—quite trivial. Cf. Xen. Mem. III 9. 11 ἐν δὲ ταλασίᾳ καὶ τὰς γυναῖκας ἐπεδείκνυεν ἀρχούσας τῶν ἀνδρῶν, διὰ τὸ τὰς μὲν εἰδέναι ὅπως χρὴ ταλασιουργεῖν, τοὺς δὲ μὴ εἰδέναι. It is hinted in οὗ—ἡττώμενον that, even in these, women may sometimes be excelled by men; but the general rule was the other way, otherwise the ridicule would be pointless. Grote somewhat exaggerates the significance of the clause οὗ —ἡττώμενον, when he suggests that Plato may have seen finer webs in Egypt— where weaving was performed by men— than in Greece (Plato III p. 200 note). Cf. Proclus in remp. I pp. 242, 253 ed. Kroll.

κρατεῖται is construed like ἡττᾶται, μειοῦται, νικᾶται and the like; but a parallel instance is hard to find. (In Aeschin. F. L. 152, cited by J. and C., the reading is ποίᾳ κρατηθεὶς ἡδονῇ;). Richards proposes κρατεῖ, in which case τὸ γένος would be the male sex—an awkward change of subject.

ὡς ἔπος εἰπεῖν . See I 341 B note The sentiment is illustrated by J. and C. from Crat. 392 C πότερον οὖν αἱ γυναῖκες ἐν ταῖς πόλεσιν φρονιμώτεραί σοι δοκοῦσιν εἷναι ἢ οἱ ἄνδρες, ὡς τὸ ὅλον εἰπεῖν γένος; Οἱ ἄνδρες.

πάντων μὲν κτλ. Plato, in short, makes government a question of capacity, and not of sex. With what follows cf. the passages cited above on 451 C. For the relative weakness of woman cf. infra 457 A and Laws 781 A.

ἐπὶ πᾶσι is doubted by Herwerden, who proposes ἐν πᾶσι or ἐν ἅπασι. ἐπί may however mean ‘with a view to,’ ‘for,’ as in 471 A.

ἡ δ) οὔ . ἡ μέν is idiomatically omitted: see 451 D note

ἄρα is better, I think, than ἆρα, though somewhat more difficult: the interrogative ἆρα is moreover generally elided before οὐ. The sentence (as J. and C. remark) is “an ironical negation with an interrogative tone.” The irony in this passage lies in ἄρα. As might be expected from the accumulation of negatives, late MSS shew a great variety of readings. Bekker follows q and reads καὶ γυμναστικὴ ἄρα καὶ πολεμική—an obvious but wholly superfluous attempt to simplify the authoritative text.

πλὴν ὅσα κτλ. For ὅσα Eusebius ( Praep. Ev. XII 32. 5) read ὅσῳ followed by ἀσθενεστέρα, ἡ δὲ ἰσχυροτέρα ἐστί, and the dative was also preferred by Schneider ( Addit. p. 38). The neuter plural of ὅσος is however used adverbially as well as the neuter singular; and the dative of ‘amount of difference’ is scarcely to the point. Instead of ἀσθενεστέρα ἢ ἰσχυροτέρα we might read (with A^{2}) ἀσθενεστέρα ἰσχυροτέρας. But the reading in the text is preferable, because it lays more stress on the identity of the male and female nature. It is the same nature, only it is stronger in men, and weaker in women. ἤ=‘or’ and not ‘than.’

ξυγγενεῖς ‐‐ τὴν φύσιν . J. and C. remark that “in the Politicus and Laws, on the other hand, the aim of the legislator is rather to unite in marriage opposite natures that they may supplement each other: Pol. 309, 310, Laws 773 ff.” Such a marriage law is unnecessary in the Republic, where the opposite qualities of strength and sensibility are already united in the character of each of the parents. See on II 375 C.

εὐχαῖς ὅμοια . Cf. 450 D note

κατὰ φύσιν . 449 A notes Plato's proposals — so he asserts — are ‘natural,’ because in harmony with the natural endowments of gifted women; and it is because they are natural that he calls them possible. The definition of δυνατόν is interesting and noteworthy: see 466 D and 471 C note Grote ( Plato III p. 201) has observed that Plato is here refuting a current objection to his theories: in the next sentence he turns his adversaries' weapon against themselves.

ἦν . 452 E.

ἀρετὴν ‐‐ ἀμφιέσονται : ‘they will clothe themselves with excellence instead of garments,’ viz. by thus stripping for exercise, because τοῦ βελτίστου ἕνεκα γυμνάζονται: see B below. Jowett's translation “for their virtue will be their robe” is incorrect, and would require the future perfect instead of ἀμφιέσονται. The correct explanation is given by Schneider on p. 300 of his translation. ἀμφιέσονται (for the usual Attic ἀμφιοῦνται, which Herwerden would write) has a certain archaic effect (cf. I 330 B note), and the saying may be borrowed or adapted from some earlier author. The same metaphor is found in Plutarch Praec. Coniug. 10. 139 C τοὐναντίον γὰρ ἡ σώφρων ἀντενδύεται τὴν αἰδῶ (with reference to Hdt. I 8, a passage which is hardly likely—as Ast supposed— to have suggested Plato's phrase), but Plutarch's meaning is different from Plato's. So—except for the metaphor— is Tennyson's in the line quoted by Warren from Godiva “Then she rode forth, clothed on with chastity.”

κοινωνητέον πολέμου κτλ. The wives of the Sauromatae are described by Herodotus (IV 116) as ἐπὶ θήρην ἐπ’ ἵππων ἐκφοιτέουσαι ἅμα τοῖσι ἀνδράσι καὶ χωρὶς τῶν ἀνδρῶν, καὶ ἐς πόλεμον φοιτέουσαι καὶ στολὴν τὴν αὐτὴν τοῖσι ἀνδράσι φορέουσαι. Cf. also Laws 804 E—806 B. See also on 451 C ff.

δοτέον . There is no reason whatever for thinking (as some critics have thought) that Plato is not serious in making these regulations. Stobaeus ( Flor. 43. 100) has ἀποδοτέον: but see 452 A note

γελῶν ἀνήρ . ἀνήρ is said with a fine touch of scorn. It is difficult to read this passage without suspecting a personal reference, perhaps to some representative of the comic stage. J. and C. remark that jests of the kind objected to by Plato occur in Ar. Lys. 80 —83. See also next note and App. I. Spartan precedents are cited by Hermann-Thumser Gr. Staatsalt. p. 180 note 3.

ἀτελῆ ‐‐ καρπόν : ‘plucking unripe fruit of laughter.’ Pindar ( Fr. 209 B ergk) satirised physical speculation ( τοὺς φυσιολογοῦντας) in the words ἀτελῆ σοφίας δρέπων καρπόν, where σοφίας is a defining genitive, denoting not the tree, but the fruit. Pindar means that their σοφία is ἀτελής or inconsummatc—misses its mark —is no real σοφία at all. More suo Plato adapts the Pindaric fragment to his own purpose. The object of his attack is Comedy, and Comedy cultivates, not σοφία, but τὸ γελοῖον. Hence—according to the reading of the text—Plato replaces Pindar's σοφίας by the words τοῦ γελοίου. The humour of his adversary is ἀτελές or inconsummate—no real humour at all: for οὐδὲν οἶδεν—ἐφ’ ψ(= γελᾷ οὐδ’ ὅ τι πράττει. Cf. 452 D μάταιος ὃς γελοῖον ἄλλο τι ἡγεῖται ἢ τὸ κακόν. This interpretation assumes that σοφίας in Plato is a gloss interpolated to complete the quotation. See cr. n. and App. III.

κάλλιστα κτλ. The doctrine of this famous sentence, which sounds like a manifesto, and was characteristically selected by Grote as one of the mottoes to his Plato, is essentially Socratic: see especially Xen. Mem. IV 6. 8 , 9 and other passages quoted by Zeller^{4} II 1. pp. 149—153. Utilitarianism of this kind pervades the Republic, as Krohn has amply proved ( Pl. St. p. 370), and asserts itself even in the highest flights of Plato's idealism ( ἡ τοῦ ἀγαθοῦ ἰδέα—ᾗ δίκαια καὶ τἄλλα προσχρησάμενα χρήσιμα καὶ ὠφέλιμα γίγνεται VI 505 A). But even Socrates ennobles his utilitarianism by placing soul far above body in dignity and worth. In Plato utilitarianism becomes transfigured by Idealism and the doctrine of Immortality. Here it should be noted that καλόν combines, as often, the ideas of artistic fitness or propriety, and propriety of conduct. The moral sense of the Greeks lay in their appreciation of the beautiful.

διαφεύγειν . The present is less presumptuous than διαφυγεῖν conjectured by Herwerden. It is proved to be right by διαφεύγεις below, which Herwerden more suo ejects.

γυναικείου ‐‐ νόμου . If γυναικεῖος is equivalent only to περὶ γυναικῶν, it is strangely used. I suspect that Plato is playing on the musical sense of νόμος, as in VII 532 A: cf. IV 424 D, E notes γυναικείου νόμου—a melody sung by women— is thus exactly parallel to the γυναικεῖον δρᾶμα (451 C note), which it is clearly intended to recall.

λέγε is changed to φέρε by Cobet, to ἄγε by Richards. ἄγε may of course be right: the confusion occurs in the MSS of Plato Theaet. 162 D and 169 C (see Schanz's critical notes on these two passages), and doubtless elsewhere also. But in default of MS authority, it is safer to retain λέγε. Praestat lectio difficilior. ‘Say on: let me see it’ gives an excellent meaning, and could not have been otherwise expressed. The hortatory subjunctive of the first person is occasionally used after imperatives other than ἄγε and φέρε, as in Eur. Hipp. 567 . See Kühner Gr. Gr. II p. 185.

τὰς γυναῖκας κτλ. Plato imitates the emphasis and precision of a legal enactment. The Aristophanic parallel is καὶ ταύτας γὰρ κοινὰς ποιῶ τοῖς ἀνδράσι συγκατακεῖσθαι ( Eccl. 614, 615). See App. I.

οὐκ οἶμαι κτλ. Aristotle disappointed Plato's expectations: for he will not admit that such arrangements are even ὠφέλιμα ( Pol. B 1. 1261^{a} 2 ff.).

πλείστην ἂν κτλ. On the omission of ἄν see IV 437 B note and Prot. 316 C , with my note ad loc. Without ἄν, the reference must, I think, be to the past, in which case πλείστην—γενέσθαι will allude to some controversy which the doctrine of the community of wives may have occasioned before these words were written. But εὖ μάλ’ ἂν ἀμφισβητηθείη makes it pretty clear that Plato is thinking of the future.

λόγων σύστασιν : “sermonum conspirationem” Ficinus, rightly. The passage which follows is an excellent example of Socratic εἰρωνεία .

ὑφεκτέον ‐‐ δίκην : ‘I must pay the penalty,’ viz. for trying to run away. The natural penalty for running away is of course to have to stay and fight. Herwerden misses the point when he proposes to excise δίκην and understand λόγον .

ἔασόν με κτλ. For the metaphor in ἑορτάσαι and ἑστιᾶσθαι see I 354 A note ἑστιᾶσθαι ὑφ’ ἑαυτῶν is like our ‘castles in the air.’

πρὶν ἐξευρεῖν κτλ. Cf. Men. 86 E .

θέντες κτλ. εἶναι goes with θέντες: “das Dasein des gewünschten als gegeben annehmend” (Schneider). A few inferior MSS omit εἶναι; but “apparet εἶναι facile supervacaneum, minime vero explicationis gratia addendum videri librariis potuisse” (id.). To write ἤδη for εἶναι (with Vind. E: cf. also Postgate in J. Ph. XV p. 113) is too great a change, and otherwise objectionable, in view of the ἤδη which follows.

καὶ ὕστερον . καί is ‘and’ (Jowett), not ‘also’ (Campbell).

ᾗ δυνατά . Stallbaum (with q and a few late MSS) reads εἰ δυνατά, which is more accurate, no doubt. But in saying ‘how it is possible’ instead of ‘whether it is possible’ Socrates hints that he will be able to prove the possibility of his scheme. We have here in fact a sort of prophecy of 473 B ff. Schneider ( Addit. p. 39) cites a close parallel from Tim. 27 C ᾗ γέγονεν ἢ καὶ ἀγενές ἐστιν.

ὅτι πάντων κτλ. Cf. Ar. Eccl. 583 καὶ μὴν ὅτι μὲν χρηστὰ διδάξω πιστεύω. See App. I.

αὐτοὺς ‐‐ νόμοις . In issuing their commands, the rulers will either themselves obey the laws (i.e. issue such orders as the laws direct) or act in accordance with the spirit of the laws: see next note. αὐτούς=ipsos sc. as well as τοὺς ἀρχομένους. The reading αὐτοῖς (K and Ficinus) is intrinsically good, and may be right: for it accentuates the contrast between cases prescribed for by actual law, and such as are left to the rulers' discretion. But there is hardly sufficient ground for deserting A.

μιμουμένους : sc. τοὺς νόμους. In matters not actually prescribed for by legislative enactment, the rulers will ‘imitate,’ i.e. will issue commands in harmony with the spirit of, such laws as do exist. The reading of q^{2}, μὴ πειθομένους, recommended by Herwerden, gives a poor, if not actually an erroneous, meaning.

ὁμοφυεῖς . See on 456 B.

ἀναμεμιγμένων . ἀναμεμιγμένοι would be more usual, but the genitive lays more stress on the participial clause: cf. Thuc. III 13. 6 βοηθησάντων δὲ ὑμῶν προθύμως πόλιν τε προσλήψεσθε κτλ., and other examples quoted in Kühner Gr. Gr. II p. 666. See also infra on 459 C. Here, too, it should be noted that the addition of a parenthetical οἶμαι helps to render ἀναμεμιγμένων independent of ἔσονται. The genitive absolute in ἰδίᾳ δὲ—κεκτημένου may also, as Jackson suggests, have influenced Plato's choice of construction in this clause. Plato perhaps thought of Sparta when he wrote the present sentence: cf. Plut. Lyc. 15. 1 ἦν μὲν οὖν καὶ ταῦτα παρορμητικὰ πρὸς γάμους· λέγω δὲ τὰς πομπὰς τῶν παρθένων καὶ τὰς ἀποδύσεις καὶ τοὺς ἀγῶνας ἐν ὄψει τῶν νέων, ἀγομένων οὐ γεωμετρικαῖς, ἀλλ’ ἐρωτικαῖς, ὥς φησιν ὁ Πλάτων, ἀνάγκαις.

γεωμετρικαῖς γε : sc. ἀναγκαῖα, with which the dative goes, as in Soph. 252 D ταῖς μεγίσταις ἀνάγκαις ἀδύνατον (cited by J. and C.). We have here one of the earliest assertions of the famous doctrine which has played so large and important a part in the history of philosophy—the doctrine of the so-called ‘necessity’ of mathematical reasoning. See for instance Mill's Logic Book II c. 5. In the rest of this sentence Schneider suspects that Glauco is paraphrasing some passage of poetry. τὸν πολὺν λεών certainly sounds tragic.

γάμους ‐‐ ἱερούς . Cf. Laws 841 D ταῖς μετὰ θεῶν καὶ ἱερῶν γάμων ἐλθούσαις εἰς τὴν οἰκίαν. The nuptials of Zeus and Hera were known as the Θεογαμία, or ἱερὸς γάμος, and were celebrated by a special festival in Athens and elsewhere: see H. Graillot's article on ἱερὸς γάμος in Daremberg and Saglio's dictionary, where the authorities are cited, or Farnell's Cults of the Greek States I pp. 184—192. To Greek religious sentiment the marriage of Zeus and Hera was (as Graillot says) the ideal type of all human marriages, and for this reason Plato characteristically applies the expression ἱερὸς γάμος to his ideal of marriage in his ideal city. Cf. also Proclus in Tim. 16 B τῶν ἐν ἀπορρήτοις λεγομένων ἱερῶν γάμων, οἶς καὶ ὁ Πλάτων εἰς δύναμιν ἐξομοιῶν περὶ τοὺς πολίτας καὶ τοὺς τῶνδε γάμους ἱεροὺς γάμους προσηγόρευσε, and see Abel Orphic. p. 243. It is clear from Plato's words that he would have repudiated with scorn the charge of seeking to abolish marriage. We have already seen that he endeavours to make the State into one vast family (457 B note); and it is in the same spirit that he now tries to raise marriage from a private into a public institution, without sacrificing any of the religious ceremonies and associations by which the union of the sexes was hallowed in the eyes of his contemporaries: cf. 459 E. If his vaulting idealism “o'erleaps itself and falls on the other,” that is no reason why we should impugn his motives, or refuse our homage to his unquenchable faith in the possibilities of human nature.

κύνας θηρευτικοὺς κτλ. Cf. 451 D and Plut. Lyc. 15. 12 πολλὴν ἀβελτερίαν καὶ τῦφον ἐνεώρα τοῖς περὶ ταῦτα τῶν ἄλλων νομοθετήμασιν· οἳ κύνας μὲν καὶ ἵππους ὑπὸ τοῖς κρατίστοις τῶν ὀχείων βιβάζουσι—τὰς δὲ γυναῖκας ἐγκλεισάμενοι φρουροῦσιν κτλ. See also on 451 C, 460 C.

γίγνονται : ‘prove themselves to be’ (J. and C.), rather than ‘grow to be’ (D. and V.): cf. III 412 C οἱ δὲ γεωργῶν ἄριστοι ἆρ’ οὐ γεωργικώτατοι γίγνονται;

γεννᾶται : viz. τό τε—γένος, not τὸ γεννώμενον (suggested as an alternative explanation by J. and C.). For the sense cf. Xen. Mem. IV 23 (Jackson).

ἄκρων εἶναι . εἶναι is omitted by q and Flor. U. Without it, however, as Schneider points out, σφόδρα might be taken with δεῖ; whereas the sense requires it to go with ἄκρων. We should expect ἄκρους εἶναι—τοὺς ἄρχοντας, but the accusatives are attracted into the genitive by δεῖ. For an analogous idiom see III 407 B note

ἰατρὸν δέ που κτλ. μέν after δεομένοις balances δέ in ὅταν δὲ δή, and not ἀλλὰ—ὑπακούειν, which merely explains φαρμάκων by stating its antithesis. There is consequently no sufficient reason for changing ἐθελόντων into ἐθέλουσιν (with some inferior MSS, Stephanus, Madvig, and others), although ἐθέλουσιν would no doubt be more usual. For the genitive absolute cf. 458 D note I agree with Schneider and Campbell in taking the participle as neuter and not masculine (so Stallbaum and Jowett). ὑπακούειν is not ‘submit to’ but ‘respond to’—‘are willing to respond to,’ i.e. be cured by ‘a course of treatment without drugs’: cf. Prot. 325 A , and for δίαιτα contrasted with drugs III 406 D.

ἡγούμεθα κτλ. See cr. n. If εἶναι is retained after ἡγούμεθα, we must (with Ast in his second edition) understand Plato to mean ἡγούμεθα ἐξαρκεῖν καὶ φαυλότερον εἶναι, i.e. καίπερ φαυλότερον ὄντα, or else suppose that ἰατρὸν καὶ φαυλότερον εἶναι is an accusative and infinitive forming the subject to ἐξαρκεῖν. Neither explanation is simple or natural; and Stephanus, Madvig, and others have in my judgment some reason for expunging εἶναι, although its intrusion is not altogether easy to explain. It is possible enough that Plato wrote ἀνδρειοτέρου δεῖ <εἶναι> τοῦ ἰατροῦ in line 19 (cf. δεῖ ἄκρων εἶναι τῶν ἀρχόντων in B above); and the possibility is raised, I think, into a probability, when we thus obtain a natural explanation of the erroneous εἶναι after ἡγούμεθα. εἶναι following δεῖ appeared difficult, and was omitted, as it is in B above by q; a later scribe reinserted it in the wrong place. I have therefore ventured to transpose the word.

ἀνδρειοτέρου . It needs more courage to use drugs than to prescribe a regimen, because the risk is greater. Nothing could be more appropriate than Plato's use of the word, although it has been doubted by Richards. who proposed αὖ δριμυτέρου at first, and afterwards ἀνδρικωτέρου. With the general sentiment Poschenrieder ( die Plat. Dial. in ihrem Verhältnisse zu den Hippokr. Schr. p. 57) compares [Hippocr.] de victus ratione VI p. 592 c. 67 Littré προκαταλαμβάνειν τὴν <*>γίειαν, ὥστε τὰς νούσους μὴ προσπελάζειν, εἰ μή τις μεγάλα πάνυ ἐξαμαρτάνοι καὶ πολλάκις· ταῦτα δὲ φαρμάκων δέεται ἤδη.

ἔφαμεν . III 389 B. Cf. also II 382 C, D.

τὸ ὀρθὸν τοῦτο : i.e. this which you call right, viz. τὸ ψεῦδος. The medicinal lie frequently appears ( γίγνεται οὐκ ἐλάχιστον) in connexion with the marriages of the guardians, as Plato proceeds to shew. τοῖς γάμοις should not be made general; the reference is specific.

δεῖ μὲν κτλ. “The case resembles that of a breeding stud of horses and mares, to which Plato compares it: nothing else is wanted but the finest progeny attainable” Grote Plato III p. 205. It is worth while to compare Plato's arrangements with those of Aristophanes in Eccl. 616—634, in spite of the comedian's lewdness and buffoonery.

τῶν δὲ μή . Cf. 460 C and 461 C. It seems to me certain from these passages that Plato in this book lends his sanction to infanticide. This has often been denied, but without sufficient reason. The subject is discussed in App. IV.

ἀκρότατον . Cf. (with Schneider) σφόδρα ἄκρων in B above and ὡς ἀκρότατον in Laws 730 E. Stephanus' ἀκρατότατον is neat, but unnecessary, in spite of καθαρόν in 460 C.

ἀγέλη , like ποίμνιον, is intended “to recall the analogy of the lower animals” (J. and C.). Cf. 451 C note αὖ serves the same purpose, by suggesting that ἀγέλη has another and a more primitive signification.

ἑορταὶ κτλ. As the ἱερὸς γάμος was celebrated with a procession and sacrifices, ending with the κλίνη τῆς Ἥρας, so Plato's ἱεροὶ γάμοι are attended with religious rites and ceremonies: see 458 E note Plato apparently does not intend these State-marriages to last beyond the duration of a single festival. At each successive festival fresh unions would be tried.

τὸν αὐτὸν ἀριθμόν . See IV 423 A note

γέρα κτλ. Special privileges seem to have been awarded at Sparta for bravery in the field (cf. Tyrtaeus Fr. 12. 35—44): it is certain at all events that cowardice was visited with every mark of disgrace (Xen. Rep. Lac. 9. 4—6 and other references in Gilbert's Gk. Const. Ant. E. T. p. 77). γέρα must be nominative, and δοτέον passive, in spite of its singular number: cf. Symp. 188 B πάχναι καὶ χάλαζαι καὶ ἐρυσῖβαι— γίγνεται. Examples like Crat. 410 C αἱ μὲν δὴ ὧραι Ἀττικιστὶ ὡς τὸ παλαιὸν ῥητέον (cited by Schneider and others) are not to the point, because αἱ—ὧραι means τὸ ὄνομα ‘αἱ ὦραι.’ It is scarcely possible to take δοτέον as active, and understand from it a passive δοτέα with ἐξουσία, because the connexion between γέρα, ἆθλα, and ἐξουσία—note ἄλλα τε καὶ κτλ.—is too close to permit of γέρα being in the accusative case.

ἐπὶ τούτων . For the construction cf. Dem. F. L. 298 τοὺς ἐπὶ τῆς πολιτείας ἐφεστηκότας and de Cor. 247 τοὺς ἐπὶ τῶν πραγμάτων.

ἀμφότερα . q and some other MSS read ἀμφότεραι, which is quite wrong: cf. Lach. 187 A πείθωμεν ἢ δώροις ἢ χάρισιν ἢ ἀμφότερα.

καὶ ἀρχαί : sc. as well as the other duties of guardians. It has not yet been specifically said that magistracies are to be open to women as well as men. J. and C. observe that “Plato seems to betray a certain consciousness that the office immediately in question might be specially suitable for women.’ Kindred duties are actually assigned to a female vigilance committee in Laws 784 A, 794 A ff.

τὸν σηκόν . A σηκός is an enclosed pen or fold in which the young of animals may be reared. Hartman prefers τινα σηκόν (with q and a Florentine MS), because the σηκός has not been mentioned before. The way has, however, been prepared for it by 459 A, 459 B ( τί δὲ ἵππων κτλ.), ποίμνιον (459 E), ἀγέλη (ib.), and σύνερξις (460 A). The comparison with a sort of ‘breeding-stud’— see above on 459 D—runs through all this passage and supplies the metaphors. See also on 460 E. The whole discussion affords an excellent example of the uncompromising rationalism with which Plato carries out his theories to their logical conclusion.

ἀνάπηρον . Pollux (II 61) explains this word as ὁ πᾶν τὸ σῶμα πεπηρωμένος; but it is little more than πηρός: cf. ἀνάπλεως, ἀναπιμπλάναι etc. The present passage is not inconsistent with III 415 B, for ὑπόχαλκος and ὑποσίδηρος do not imply deformity.

ἐν ἀπορρήτῳ κτλ. is a euphemism for infanticide: see App. IV. Compare the Spartan usage: εἰ δ’ ἀγεννὲς καὶ ἄμορφον, ἀπέπεμπον εἰς τὰς λεγομένας Ἀποθέτας, παρὰ Ταΰγετον βαραθρώδη τόπον (Plut. Lyc. 16. 1). (The word for the exposure of infants was ἀπόθεσις.) See also Whibley Greek Oligarchies p. 113 notes

εἴπερ μέλλει . μέλλοι (see cr. n.) might possibly be defended, if it referred to a previous statement to the same effect (‘if, as we saw,’ etc.). But there has been no such statement, unless with Stephanus we write ἀκρατότατον for ἀκρότατον in 459 E. εἴπερ γε μέλλει, conjectured by Herwerden, would be in danger of meaning κατακρύψουσιν, εἴπερ μέλλει κτλ. As it is, εἴπερ μέλλει qualifies ὡς πρέπει ‘as is proper.’ Gl. ‘If the class of guardians is to be kept pure.’ Glauco, in fact, takes the words out of Socrates' mouth. On the meaning and usage of εἴπερ in Greek see E. S. Thompson's edition of the Meno pp. 258—264.

πᾶσαν μηχανὴν κτλ. Aristotle ( Pol. B 3. 1262^{a} 14 ff.) argues that no precautions would prevent parents from occasionally recognising their children. In such cases Plato might reasonably hope that the general weakening of parental sentiment would secure his city against serious harm.

ἄλλας . The mothers of the children who have been exposed.

αὐτῶν τούτων : viz. the mothers. This provision is conceived in their interests, rather than in the interests of the children, as the next clause also shews.

θηλάσονται . θηλάσωνται has more MS authority than θηλάσονται; but the future indicative (and not the aorist subjunctive) is the regular construction after ὅπως in semi-final clauses: cf. IV 429 D, VII 519 E. The exceptions are—besides this passage— Symp. 198 E , Phaed. 91 A , Gorg. 480 A , B, 510 A. In most of these places there is inferior MS authority for the future, which editors now for the most part read. See Weber Entwickelung d. Absichtssätze in Schanz's Beiträge II 2. p. 66; and for the confusion in Paris A of ο and ω Introd. § 5.

προὐθέμεθα . See cr. n. προὐθέμεθα is intrinsically so much better than προθυμούμεθα that we can hardly refuse to regard this as one of the passages in which v has preserved the right reading. See Introd. § 5.

ἔφαμεν . Cf. 459 B.

ἐξ ἀκμαζόντων . The same principle was observed in Sparta (Xen. Rep. Lac. 1. 6 and Plut. Lyc. 15. 4). It is possible, though I believe incapable of proof, that Plato's limits of age were in agreement with Spartan usage.

τὰ εἴκοσι ἔτη κτλ. A woman's ἀκμή lasts ‘the twenty,’ a man's the thirty' years. Glauco asks ‘which twenty and which thirty?’ and Socrates then explains. τά before εἴκοσι is correctly explained by Stallbaum: “articulum ponit de certo quodam cogitans temporis spatio quod deinceps definit accuratius.” The antecedent to αὐτῶν is not simply ἔτη (so J. and C., with the English translators), but the duplicate expression εἴκοσι ἔτη and τριάκοντα. In γυναικὶ μὲν κτλ. Socrates proceeds as if Glauco had not interrupted: the construction is μέτριος χρόνος ἀκμῆς—γυναικί, ἀνδρὶ δὲ τὰ τριάκοντα, γυναικὶ μὲν— τίκτειν, ἀνδρὶ δὲ—γεννᾶν. τὸ ποῖον, τὰ ποῖα and the like are idiomatically used in asking for further specification, and are sometimes only impatient interruptions, intended to draw attention to the important point and add liveliness to the style: see Kühner Gr. Gr. II p. 540. Schneider takes τὰ εἴκοσι ἔτη and τὰ τριάκοντα as twenty and thirty years of age respectively, comparing τῶν ἐνενήκοντα ἐτῶν in Tim. 21 A , but χρόνος in χρόνος ἀκμῆς means duration, as is clear from ἀμφοτέρων—φρονήσεως below. It should be observed that in the Laws Plato fixes the inferior limit for men sometimes at 25 (772 D), sometimes at 30 (721 A, 785 B). By thirty-five he expected them to be married (ib.). Girls are to marry between 16 (785 B) or 18 (833 D) and 20 (ib.). Cf. Hesiod OD. 696 ff., pseudo-Solon Fr. 27. 9 and Arist. Pol. H 16. 1335^{a} 28. The Greeks seem generally to have recommended men to marry a little under or a little over thirty. See on this subject Blümner Privatalterthümer p. 36 note 1.

τίκτειν τῇ πόλει ‐‐ γεννᾶν τῇ πόλει . These phrases express concisely the Platonic view of marriage. They are equally applicable to the Spartan ideal, and may have been borrowed from Sparta. Cf. Plut. Pyrrh. 28. 5 τῶν δὲ πρεσβυτέρων τινὲς ἐπηκολούθουν βοῶντες· Οἷχε, Ἀκρότατε, καὶ οἷφε τὰν Χιλωνίδα· μόνον παῖδας ἀγαθοὺς τᾷ Σπάρτᾳ ποίει. “What Lucan observes about Cato of Utica, is applicable to the Guardians of the Platonic Republic: — Venerisque huic maximus usus | progenies: Urbi pater est, Urbique maritus” ( Phars. II 387 f.) Grote.

ἐπειδὰν ‐‐ ἀκμήν : ‘when he has outlived his swiftest prime of running.’ The expression ὀξυτατην δρόμου ἀκμήν is doubtless borrowed from some epinikian poet, perhaps Bacchylides or Pindar. The dactylic rhythm is not in itself enough to justify us in assigning the phrase (with Herwerden) to epic or elegy. The author of the quotation was probably speaking not of a man, but of a race-horse. By applying the phrase (of course in a metaphorical sense) to his bridegrooms, Plato contrives again to suggest the now familiar analogy of a ‘breeding-stud of horses and mares’: see on 460 C. The comparison gains in realism and point, if it was the custom of antiquity, as it is now, to bring a first-rate racer to the stud ( ἱπποφόρβιον, ἱπποτροφεῖον) when he ceased to run. This is probable in itself, and supported to some extent by a comparison of Plut. Lyc. 15. 12 ἵππους ὑπὸ τοῖς κρατίστοις τῶν ὀχείων βιβάζουσι, χάριτι πείθοντες ἢ μισθῷ τοὺς κυρίους with Virg. Georg. 3. 209—211. Just so Plato will not allow his guardians to marry until the fever in the blood has somewhat cooled: cf. Laws 775 B—776 B and J. B. Mayor in Cl. Rev. X p. 111. Stallbaum was the first to detect the poetical quotation. J. and C., though translating by “his swiftest prime of running,” follow Schleiermacher in understanding the phrase literally; but we may fairly doubt if Greek runners had passed their prime at 25, and, even if they had, “non hic erat tali designationi locus, nisi forte ob id ipsum, quod cursui minus idonei forent, ad nuptias idoneos visos credimus” (Schneider). παρῇ means ‘let go by,’ “hinter sich hat” (Schneider): cf. such expressions as παριέναι καιρόν (II 370 B al.), νύκτα μέσην παρέντες (Hdt. VIII 9), and especially Soph. O. C. 1230 εὖτ’ ἂν τὸ νέον παρῇ ‘when he hath seen youth go by’ (Jebb), and Bacchylides 3. 88 ed. Kenyon ἀνδρὶ δ) [ οὐ θ] έμις πολιὸν π[ αρ] έντα | γῆρας θάλ[ εια] ν αὖτις ἀγκομίσσαι | ἥβαν.

ἂν λάθῃ . “Si non latuerit foetus praeter legem susceptus, ne in lucem quidem edetur, sed antea opprimetur” (Schneider). Cf. C below.

γεννήσεται = ‘will be produced’ must, if right, refer to birth (“hervorkommen wird” Schneider), otherwise φύς is superfluous. Bekker and others are possibly right in reading γενήσεται with Ξ^{1} q, and some other MSS: cf. Hdt. VI 69, where Stein prints γεγενημένος in place of γεγεννημένος. See Introd. § 5.

φὺς ἅς . See cr. n. and Introd. § 5.

ἐξ ὠφελίμων κτλ. Cf. IV 424 A note

ὑπὸ σκότου . Cf. (with J. and C.) σκότιος=‘an unlawful child.’

γεγονώς =‘produced,’ ‘a product of,’ is cancelled by Hartman; but φύς is too far away, and γίγνεσθαι (‘to be produced’) is sufficiently accurate: cf. γένηται in 461 C.

ἀνέγγυον : ‘unauthorised,’ because the child of an irregular union. An ἀνέγγυος γάμος is a marriage without an ἐγγύη or contract between the parents of the betrothing parties (Blümner Privatalt. p. 262 note 2).

ᾧ. ᾗ is read by Ξ, Vind. E and Eusebius ( Praep. Ev. XIII 19. 18); but αὐτούς includes both sexes, and in such cases the masculine is preferred to the feminine. Hartman strangely thinks

ᾧ neuter.

θυγατρὶ κτλ. The cases enumerated are all in the direct line, and nothing is said forbidding unions between ‘brothers’ and ‘sisters.’ See however 461 E note Greek law permitted the marriage of uncles with nieces, aunts with nephews, and even half-brothers and halfsisters, provided they were not ὁμομήτριοι (Becker's Charicles E. T. p. 478, with the passages there cited). Some of Plato's contemporaries, notably the Cynics, entertained peculiarly revolting views on this subject, and the question was frequently agitated in his time: see Dümmler Proleg. zu Pl. St. pp. 52 ff. The Stoics agreed with the Cynics: see the authorities cited in Henkel Stud. zur Geschichte d. Gr. Lehre vom Staat p. 30.

καὶ ταῦτά γ’ ἤδη κτλ. : ‘and all this only after we have exhorted them’ etc. ἤδη goes with ἀφήσομεν (or the like) understood after πάντα. J. and C. wrongly connect πάντα with προθυμεῖσθαι (‘to use all diligence’). The voice should pause a little before διακελευσάμενοι.

μηδ’ εἰς φῶς κτλ. : ne in lucem quidem efferre. Much less shall we permit it to live if born: see App. IV. μηδέ prepares the way for ἐὰν δέ τι βιάσηται κτλ. Hartman strangely prefers μή, “cum post μάλιστα coniunctio μηδέ prorsus frigeat.” But μάλιστα μέν is, of course, ‘if possible.’

μηδὲ ἕν . See cr. n. μηδέ γ’ ἕν occurs in a few inferior MSS besides A, and is read by Baiter and others, but we do not find γε thus interposed between οὐδέ ( μηδέ) and εἷς.

γένηται κτλ. γένηται sc. κύημα. βιάσηται means ‘force its way’ sc. εἰς τὸ φῶς (J. and C.). The extreme emphasis shews what importance Plato attached to this provision. The procuring of abortion, though perhaps in certain cases punishable by law (Meier and Schömann Att. Process p. 381), was in practice common enough: see Blümner Privatalt. p. 76. Plato permits it also in the Laws (740 D). The general Greek sentiment on this matter is fairly represented by Aristotle when he says ( Pol. H 16. 1335^{b} 22 ff.) ὡρίσθαι γὰρ δεῖ τῆς τεκνοποιίας τὸ πλῆθος. ἐὰν δέ τισι γίνηται παρὰ ταῦτα συνδυασθέντων, πρὶν αἴσθησιν ἐγγενέσθαι καὶ ζωήν, ἐμποιεῖσθαι δεῖ τὴν ἄμβλωσιν· τὸ γὰρ ὅσιον καὶ τὸ μὴ διωρισμένον τῇ αἰσθήσει καὶ τῷ ζῆν ἔσται .

τιθέναι κτλ. : ‘so deal with them.’ τιθέναι is more delicate than ἐκτιθέναι, which was read before Bekker, although it has no MS authority. Herwerden suggests that τιθέναι means θάπτειν (as in 469 A), but Plato expresses himself with more refinement. τροφῆς does not mean, as some are fain to believe, merely the educational system reserved for the guardians: see on 459 E and App. IV.

πατέρας κτλ. : ‘how will they distinguish one another's fathers’ etc.? The Aristophanic parallel is here very close: Πῶς οὖν οὕτω ζώντων ἡμῶν τοὺς αὑτοῦ παῖδας ἕκαστος | ἔσται δυνατὸς διαγιγνώσκειν; Τί δὲ δεῖ; πατέρας γὰρ ἅπαντας | τοὺς πρεσβυτέρους αὑτῶν εἶναι τοῖσι χρόνοισιν νομιοῦσιν ( Eccl. 635—637). The question touches an obvious difficulty in any system of the community of children; but, as a link in the chain of evidence connecting the Ecclesiazusae and the Republic, the parallel deserves to carry weight, although it has sometimes been pressed too far. See on the one hand Teichmüller Lit. Fehd. 1 pp. 18—19 and Chiappelli Riv. di Filolog. XI p. 213, and on the other Zeller^{4} II 1. p. 551 note 2. Cf. also App. I.

δεκάτῳ κτλ. : ‘in the tenth month and also in the seventh month.’ δή (as J. and C. remark) draws attention to the more exceptional case: cf. II 367 C note The Greek cannot, I think, be taken as an inexact way of saying “from seven to ten months after” (J. and C.). In point of fact the majority of ancient writers on the subject denied that children were ever born in the eighth month of pregnancy: see Gellius Noct. Att. III 16 and Censorinus de die natali 7. 2.

ἐγέννων : ‘were engaged in begetting children’: cf. 460 E, and 461 B ( τῶν ἔτι γεννώντων). Richards has pointed out ( Cl. Rev. IV p. 7) that the imperfect refers “to the whole time of life during which father and mother were allowed, if the lot fell upon them, to take part in the regular unions.” Cf. Tim. 18 D νομιοῦσι δὲ πάντες πάντας αὐτοὺς ὁμογενεῖς, ἀδελφὰς μὲν καὶ ἀδελφοὺς ὅσοιπερ ἂν τῆς πρεπούσης ἐντὸς ἡλικίας γίγνωνται. Jowett's version—“all who were begotten at the time when their fathers and mothers came together”—mistakes both ἐν and ἐγέννων. Schneider translates the passage correctly.

ὥστε ‐‐ ἅπτεσθαι . I agree with Richards in understanding this of the “irregular unions which were last mentioned” (461 C). But in spite of the explicit reference in ὃ νῦν δὴ ἐλέγομεν, Plato has not as yet forbidden such unions between ‘brothers’ and ‘sisters’: see 461 C note The discrepancy is hard to explain, especially as the list in 461 C seems intended to be exhaustive. The effect of the prohibition (owing to the meaning now given to ‘brother’ and ‘sister’) would be greatly to restrict, but not to abolish, unauthorised liaisons.

ἀδελφοὺς κτλ. refers only to State-marriages, as ὁ κλῆρος shews. Without this exemption Plato's proposals would (according to Richards l.c.) “have rendered all unions whatever practically impossible.” Surely not; although they would have unduly favoured the τηλύγετος παῖς. A son, for example, who is born when his mother is 21 and his father 26, cannot marry till he is 49, because he is 29 before his bride can possibly be born, and she cannot marry under 20; whereas a son, whose father is 54 and mother 39 when he is born, can marry a girl only one year younger than himself, because his father and mother retire at 55 and 40 respectively. Did Plato intend the sons of elderly couples to marry young, and those of young couples to marry late? Such an inference is unlikely, although it is the logical outcome of his theories. In any case Plato did well to introduce a saving clause. The κομψὸς κλῆρος, obedient to the archons, would couple ‘brothers’ and ‘sisters,’ whenever it seemed desirable in the interests of the State, so long as they were not really blood-relations. (This the archons of course would know.) Apollo's priestess would platonize. We must suppose that her assent is given in advance, and once for all (although προσαναιρῇ is present and follows ξυμπίπτῃ), unless she had an accredited representative on the spot, which there is nothing to indicate. On Plato's attitude to Apollo see IV 427 C note

ἑπομένη ‐‐ πολιτείᾳ . This topic was not specified in the original distribution of the subject (458 B), but it is closely connected with ὡς μακρῷ βελτίστη. Plato does not deal with it till 464 B.

βεβαιώσασθαι κτλ. Hirschig cancelled παρὰ τοῦ λόγου: but cf. (with Stallbaum) Gorg. 489 A ἵνα—βεβαιώσωμαι ἤδη παρὰ σοῦ. ‘The argument’ is personified, as often.

ἆρα . <εἰ> ἄρα was suggested by Dobree; but cf. (with Stallbaum) Gorg. 475 B πρῶτον μὲν δὴ σκεψώμεθα, ἆρα—ὑπερβάλλει κτλ. The exact translation is ‘to enquire, Do the institutions we have described’ etc.

ἔχομεν οὖν κτλ. Cf. IV 422 E. στάσις was the greatest evil which a Greek city had to fear, and Athens had suffered from it grievously. Now individualism was the peculiar pride and glory of the Athenian State (Thuc. II 37), so that we need not wonder if Plato traced στάσις to individualism, and rushed to the opposite extreme. Cf. Krohn Pl. Fr. p. 4, Pöhlmann Gesch. d. ant. Kommunismus etc. pp. 146—184, and see on II 369 A.

ποιῇ μίαν . J. and C. assert that Plato “has no idea of a unity of opposites or differences— τὸ ἀντίξουν συμφέρον,” and Aristotle argues to the same effect in Pol. B 2. 1261^{a} 22 ff. But it is in fact on such a unity that the entire fabric of Plato's city rests: see IV 423 D note, and cf. also 432 A, 443 D. The perfect city is a ἕν with three πολλά—rulers, auxiliaries, farmers and artisans, or, if rulers and auxiliaries are classed together as guardians, then with two. Plato's object throughout this episode is to keep the whole city ‘one’ by preventing one of its constituent factors, viz. the guardians, from becoming ‘many.’ If the guardians are united—so he holds—no danger to the city's unity need be apprehended from the others (465 B). With the sentiment generally cf. Ar. Eccl. 594 and 674 ( μίαν οἴκησίν φημι ποιήσειν συρρήξασ’ εἰς ἓν ἅπαντα | ὥστε βαδίζειν εἰς ἀλλήλους). See also on 463 E and App. I.

οἱ μὲν ‐‐ τῆς πόλεως . As when a national disaster is made the occasion of a party victory. Plato may be thinking of scenes which he had witnessed in his native city. Bosanquet cites an excellent illustration from Dem. de Cor. 217.

καὶ ‐‐ ταὐτά : i.e. ὅταν μὴ ἅμα φθέγγωνται—τό τε ἀλλότριον καὶ τὸ οὐκ ἀλλότριον. Hartman ejects καὶ περὶ τοῦ ἀλλοτρίου as a “futile interpretamentum” on τὸ οὐκ ἐμόν. There is nothing to prove that καὶ—ταὐτά was read by Aristotle ( Pol. B 3. 1261^{b} 18), Plutarch (140 D, 484 B, 767 D), Iamblichus ( de vita Pythag. 167) or Proclus ( in remp. ed. Kroll II pp. 78. 28, 365. 11), though Iamblichus uses the word ἀλλότριον instead of Plato's οὐκ ἐμόν. But as none of these authors pretends to be quoting Plato's ipsissima verba, the omission proves nothing. Although the words add nothing to the sense, they approach the matter from another point of view, and are in my judgment certainly genuine.

ἐν ᾗτινι κτλ. ‘Thus in whatever city the largest number of men agree in applying these expressions, “mine” and “not mine,” to the same thing,’ etc. τοῦτο agrees with the nearest of the two objects, viz. τὸ ἐμόν. For the use of ἐπί cf. Parm. 147 D . The reading ἐπὶ τὸ αὐτό—see cr. n.—is as old as Iamblichus: see the passage referred to above, where Iamblichus has ἐπὶ τὸ αὐτὸ τὸ ἐμὸν φθέγγεσθαι καὶ τὸ ἀλλότριον. It is retained by the majority of editors; but no other instance of λέγειν ἐπί τι has yet been adduced, and the expression is certainly very strange. φέρειν ὄνομα ἐπί τι ( Soph. 237 C , D: cf. also Tim. 37 E ), of which Schneider reminds us, is a different thing from λέγειν ὄνομα ἐπί τι. Various emendations have been proposed. The choice seems to me to lie between ἐπὶ τῷ αὐτῷ and ἐπὶ τοῦ αὐτοῦ. The latter emendation—which I once adopted—was (as I learn from Schneider) proposed by Küster instead of ἐπὶ τὸ αὐτό in Iamblichus: cf. ὥσπερ ἐπὶ τοῦ δακτύλου ἐλέγομεν VII 524 E (‘in the case of the finger’) al. Although the genitive may be right, the dative now seems to me slightly more natural and easy. Hartman ejects ἐπὶ τὸ αὐτό altogether, but there is no occasion for the knife. Cf. IV 436 B note For the error see Introd. § 5.

καὶ ἥτις δὴ κτλ. δή is illative, and καί ‘also.’ καὶ—ἔχει (sc. ἄριστα διοικεῖται) is certainly interrogative, as Schneider pointed out: see in D below τοῦτο δ ἐρωτᾷς. Plato recurs to his favourite analogy between man and the State: cf. II 368 E f. notes

οἷον ὅταν κτλ. Poschenrieder ( Die Pl. Dial. in ihrem Verhältnisse zu d. Hippokratischen Schr. p. 67) cites a remarkable parallel from the author of the treatise de locis in homine (Littré VI p. 278 c. 1) εἴ τις βούλεται τοῦ σώματος ἀπολαβὼν μέρος κακῶς ποιέειν τὸ σμικρότατον, πᾶν τὸ σῶμα αἰσθήσεται τὴν πεῖσιν, ὁκοίη ἄν τις ᾗ, διὰ τόδε, ὅτι τοῦ σώματος τὸ σμικρότατον πάντα ἔχει, ὅσαπερ καὶ τὸ μέγιστον· τοῦτο δ’ ὁποῖον ἄν τις πάθῃ ἐπαναφέρει πρὸς τὴν ὁμοεθνίην ἕκαστον πρὸς τὴν ἑωυτοῦ, ἤν τε κακόν, ἤν τε ἀγαθόν ᾗ· καὶ διὰ ταῦτα καὶ ἀλγέει καὶ ἥδεται ὑπὸ ἔθνεος τοῦ σμικροτάτου τὸ σῶμα, ὅτι ἐν τῷ σμικροτάτῳ πάντ’ ἔνι τὰ μέρεα καὶ ταῦτα ἐπαναφέρουσι ἐς τὰ σφῶν αὐτῶν ἕκαστα καὶ διαγγέλλουσι πάντα. The ‘sympathy’ of the different parts of the human body was a Hippocratean tenet ( ξυμπαθέα πάντα de alimento IX c. 23 Littré). Cf. Shakespeare Othello III 4. 146—148, “For let our finger ache, and it indues Our other healthful members ev'n to that sense Of pain.” Plato goes farther, and represents the partnership as extending also to the soul: see next note.

πᾶσα ἡ κοινωνία κτλ. : ‘the entire partnership pervading the body with the soul, organized into a single composite organization, viz. that of the ruling power in the partnership’ etc. Plato's language is precise, but difficult. I take ἡ—ψυχήν as defining the κοινωνία. κατὰ τὸ σῶμα is written rather than τοῦ σώματος, because the partnership is not only a partnership of body with soul, but also a partnership of the different parts of body with one another. τεταγμένη—see cr. n. and App. V —appears to suit σύνταξις better than τεταμένη. A σύνταξις is the ordered combination of two or more elements: cf. Tim. 24 C and Laws 903 D ψυχὴ συντεταγμένη σώματι. The words τοῦ ἄρχοντος define the σύνταξις; although neuter in gender, they really refer, not to the soul, but to the whole σύνταξις or σύνολον, i.e. ὁ ἄνθρωπος. It is ὁ ἄνθρωπος who rules in the partnership, although he is himself a partner only in the sense in which the whole is partner with its parts. The expression ὁ ἄνθρωπος τὸν δάκτυλον ἀλγεῖ is thus seen to be as exact as possible. The confusion between τεταγμένος and τεταμένος is easy: συντεταμένως, for example, and συντεταγμένως are often confused in MSS: see Ast's Lex. Plat. s.v. ξυντεταμένως and my edition of the Apology p. 127. Cf. also infra 474 A note

ἐν αὐτῇ : i.e. ἐν τῇ κοινωνίᾳ (so also Schneider), not (as Stallbaum) ἐν τῇ ψυχῇ. Plato means that every single man ( ἑνὸς ἀνθρώπου above) is a single organized whole—a partnership in which the whole is partner with, and rules, the parts. See also App. V.

ᾔσθετό ‐‐ ξυνήλγησεν : ‘momentary’ aorists: cf. Theaet. 156 E .

ἄλλου ὁτουοῦν : sc. besides the finger.

ἑνὸς δὴ κτλ. We may compare the Stoic doctrine “incommoda autem et commoda (ita enim εὐχρηστήματα et δυσχρηστήματα appello) communia (sc. inter sapientes) esse voluerunt” (see Cicero de Fin. III 69, and Madvig's note). Not a few of Plato's regulations in Book V foreshadow the communistic theories of Stoicism: see Dyroff Ethik d. alten Stoa pp. 211 f., 226—231. Plato however contrives to make his communism live; whereas the Stoics seldom did.

αὕτη . See cr. n. Schneider says αὐτή is “ea potissimum,” referring to VII 516 B, where however we should (I believe) read οὗτος. See note ad loc. Here αὕτη is required by the contrast with εἴτε καὶ ἄλλη τις μᾶλλον. For the error cf. VIII 552 A, where q and several MSS wrongly read αὐτή. See also Introd. § 5.

ἔστι . For the syntax see on II 363 A. ἔστι is a privileged verb in Attic prose: cf. Kühner Gr. Gr. II p. 61.

δεσπότας . Demosthenes remarks that the subjects in an oligarchy are ‘cowards and slaves’ ( ἄνανδροι καὶ δοῦλοι). See in Timocr. 75 and Whibley Gk. Oligarchies p. 143.

ἄρχοντας . Plato is thinking of the Athenian Archons. The object of this chapter, which seems at first sight somewhat loosely constructed, is to prove that συμπάθεια prevails to a unique extent in the Platonic city. The appellations σωτῆρες and ἐπίκουροι, on the one hand, and μισθοδόται and τροφεῖς on the other, involve a greater degree of interdependence than is expressed by the corresponding names in other cities. The archons too are more than fellow-rulers: they are fellow guardians, their official designation among one another serving continually to remind them of their duty to the lower classes. Among themselves they use the terms of family relationship, and with these their actions correspond. Thus the distinction between meum and tuum is more nearly obliterated than in any other city. Everything is meum.

ἐπικούρους . The official designation of the second order is applied by the people to the ruling class as a whole. They are expected to look upon the ἐπίκουροι as ‘helpers of the people’ rather than as the rulers' auxiliaries, although it is the latter function which gave them their name (III 414 B). This is clear from σωτῆράς τε καὶ ἐπικούρους, both of which epithets are suggestive of protecting deities. See also on 464 B.

παντὶ γὰρ ‐‐ ἐντυγχάνειν . A slight exaggeration: see 461 D, E notes Cf. Hdt. IV 104 ἐπίκοινον δὲ τῶν γυναικῶν τὴν μῖξιν ποιεῦνται ( οἱ Ἀγάθυρσοι), ἵνα κασίγνητοί τε ἀλλήλων ἔωσι καὶ οἰκήιοι ἐόντες πάντες μήτε φθόνῳ μήτε ἔχθει χρέωνται ἐς ἀλλήλους and ib. 180 ad fin. Similar motives for domestic communism are mentioned by Diod. Sic. II 58. See also, for other traces, whether real or legendary, of community of wives and children in antiquity Xanthus Fr. 28, Ephorus Fr. 76 and Theopompus Fr. 222 (in Müller Frag. Hist. Gr. Vol. I), together with Arist. Pol. B 3. 1262^{a} 19.

περί τε κτλ. τε is ἀνακόλουθον: we should expect καὶ περὶ τοὺς ἄλλους ξυγγενεῖς to follow. Instead, we have a change of construction, and καὶ περὶ πατέρων—καὶ περὶ τῶν ἄλλων ξυγγενῶν (line 27). Cf. II 373 B note ἤ=‘alioquin,’ as often after a verb of obligation (here δεῖν): cf. VI 489 E, 503 A.

νόμος : sc. ἔστι πράττειν.

αὐτῷ : though αὐτοῖς in C: cf. I 347 A note

φῆμαι . See on III 415 D. φήμη is the half-personified vox populi, vox Dei: cf. Nägelsbach Nachhom. Theol. p. 165. It is the quasi-personification of φῆμαι which accounts for the active ὑμνήσουσιν (‘will sing in the ears of’ etc.): cf. IX 573 A περὶ αὐτὸν βομβοῦσαι.

ὃ ‐‐ ῥῆμα is the object of ξυμφωνήσουσιν (Schneider), just as in IV 432 A ταὐτόν depends upon ξυνᾴδοντας. Aristotle's criticism deserves to be quoted ( Pol. B 3. 1262^{a} 1 ff.) οὕτως ἕκαστος ‘ἐμὸσ’ λέγει τὸν εὖ πράττοντα τῶν πολιτῶν ἢ κακῶς, ὁπόστος τυγχάνει τὸν ἀριθμὸν ὤν, οἷον ‘ἐμὸς ἢ τοῦ δεῖνος,’ τοῦτον τὸν τρόπον λέγων καθ’ ἕκαστον τῶν χιλίων, ἢ ὅσων ἡ πόλις ἐστί, καὶ τοῦτο διστάζων· ἄδηλον γὰρ ᾧ συνέβη γενέσθαι τέκνον ἢ σωθῆναι γενόμενον. There is a far deeper truth in Plato's saying than in Aristotle's animadversions thereupon, and “das schöne Wort, dass alle dasselbe mein nennen sollen, hat es nicht verdient, von Aristoteles mit logischen Regeln gehetzt zu werden. Die Geschichte hat überall wo eine erhabene Idee eine Gruppe von Menschen so durchdringen sollte, dass der Wille jedes Einzelnen nur auf dieses Gemeinsame gerichtet sei, dem platonischen Gedanken im Prinzipe Recht gegeben” (Nohle Die Statslehre Platos etc. p. 133). See also on 457 B ff.

ἔφαμεν κτλ. 462 B, C. οὐκοῦν—ἕξουσιν sums up. ὃ δὴ—ὀνομάσουσιν is parenthetical, ‘to which, as we have seen, they will apply the name “mine.”’

ἀπεικάζοντες κτλ. See on 462 C and App. V.

ἐπικούροις . Why not φύλαξι? The word φύλακες regularly includes both the ἄρχοντες and the ἐπίκουροι, but it is strange to find ἐπίκουροι including the τέλεοι φύλακες or rulers (see on II 374 D), as it appears to do here and in 466 A. The following explanations may be suggested. (1) Plato intends the community of wives and children to extend only to the Auxiliaries, and not also to the Guardians. This view is taken by Blaschke ( Familienu. Gütergem. d. Pl. St. p. 10), who asserts that the Rulers proper have already past the limits of age prescribed for matrimony. In point of fact, however, a man may become a τέλεος φύλαξ at 50 (VII 540 A, B), whereas he can marry till he is 55 (460 E). (2) As by far the largest number of husbands would be only Auxiliaries, Plato speaks somewhat loosely, as if matrimonial community were confined to them. This explanation is possible enough in itself, but fails to explain the usage in 466 A. (3) ἐπίκουροι is used with the new and deeper meaning given to it in 463 B (where see note), ‘helpers of the people,’ rather than in its original and technical sense of the rulers' auxiliaries. This suits all the passages, and is in my judgment what Plato intended. ἐπίκουρος is not the only term whose connotation deepens as the Republic proceeds: cf. II 376 B, III 392 C notes

ὡμολογοῦμεν . I formerly read <*>μολογοῦμεν with Ξ q^{2}, Stallbaum, and others; but Schneider, as I now think, is right in retaining the imperfect and referring it to the original mention of domestic communism in Book IV. The whole of this discussion may in fact be regarded as a defence in the form of an explanation of the sentence IV 423 E— 424 A. See also App. I.

τροφὴν λαμβάνοντας κτλ. summarises III 416 D, E.

ἑτέρους = μὴ τοὺς αὐτούς depends on ὀνομάζοντας understood. D. and V. make ἕλκοντα govern γυναῖκα— ἑτέρους, as Stallbaum formerly did, but Plato could not have said anything so ludicrous.

ὁμοπαθεῖς : ‘simultaneously affected by’ D. and V. ὁμοιοπαθής (Ast) would mean ‘of like passions with.’

ὡς ἔπος εἰπεῖν with οἰχήσεται=‘almost have disappeared,’ “so gut wie verschwunden sein” (Schneider): see on I 341 B. The English translators either omit or misinterpret the phrase. Aristophanes furnishes several pretty close parallels to Plato's reasoning here: cf. Eccl. 560—610 and especially 657 ( ἀλλ’ οὐδὲ δίκαι πρῶτον ἔσονται)—672. See Chiappelli Riv. di Filol. XI pp. 212 ff. and on the whole subject App. I.

οὐδὲ ‐‐ αὐτοῖς . The first οὐδέ is of course ne—quidem. Hoefer should not have conjectured οὔτε—οὔτε ( de part. Pl. p. 41).

δικαίως is ejected by Cobet and Herwerden, but δίκαιον just below supports it. There cannot justly be any lawsuits for outrages on the person, if we declare it just and honourable for a man to take the law into his own hands. This explanation is perhaps better than to translate ‘we may fairly suppose that there will not be’ etc.

ἥλιξι κτλ. Cf. (with J. and C.) Laws 879 E ἧλιξ δὲ ἥλικα—ἀμυνέσθω κατὰ φύσιν ἄνευ βέλους ψιλαῖς ταῖς χερσίν. It should be remembered that in cases of αἰκεία the guilty party was the one ὃς ἂν ἄρξῃ χει<*>ῶν ἀδίκων πρότερος (Meier u. Schömann Att. Process p. 648).

ἀνάγκην ‐‐ τιθέντες =‘curae corporum necessitatem imponentes,’ ‘compelling them to keep themselves in condition.’ Cf. Xen. Rep. Lac. 4. 6 ἀνάγκη δ’ αὐτοῖς εὐ εξίας ἐπιμελεῖσθαι· καὶ γὰρ πυκτεύουσι διὰ τὴν ἔριν, ὅπου ἂν ξυμβάλωσι. It is probably of Sparta that Plato is thinking. I have now reverted to the best supported reading, although the use of τιθέντες as virtually equivalent to ἐπιτιθέντες is not free from difficulty. There is considerable MS authority (including II) for ἀνάγκῃ, and as ἐπιμέλειαν was read by A^{1} (see cr. n.) and several other MSS, I once conjectured <ἐν> ἀνάγκῃ σωμάτων ἐπιμέλειαν τιθέντες, taking ἐν ἀνάγκῃ as meaning ἀναγκαίαν; but this idiom is very rare except with ἐστί, ἦν and the like. Stobaeus ( Flor. 43. 102) and Stallbaum read ἀνάγκην σωμάτων ἐπιμελείας. In q and two other MSS the text runs ἀνάγκῃ (or ἀνάγκη) σωμάτων ἐπιμελείας τιθέντες. Does this mean ‘requiring them to guard against violence to the person’ ( ἀνάγκῃ σωμάτων)? If Plato meant to convey this meaning, it would be preferable to read ἀνάγκῃ σωμάτων ἐπιμέλειαν τιθέντες (for which there is also better MS authority), or possibly ἀνάγκῃ σωμάτων <*>πιμέλειαν <ἀντι> τιθέντες, but ἀνάγκῃ σωμάτων would be a fantastic expression, though perhaps intelligible after βιαίων and αἰκείας. On the whole, I think the reading printed above has most in its favour.

ἐν τῷ τοιούτῳ : i.e. by a personal encounter.

πρεσβυτέρῳ κτλ. This too is probably Spartan: cf. Xen. Rep. Lac. 2. 10. Patriarchal discipline is in perfect harmony with Plato's conception of the State as a single family.

καὶ μὴν ὅτι γε κτλ. An anacoluthon, the construction being broken by οἶμαι δ’ οὐδὲ κτλ.: see I 352 B note and infra 471 C. Here the apodosis would have been δῆλον ἂν εἴη or the like: cf. Stallbaum on Laws 677 B. Schneider and others suppose that ὡς τὸ εἰκός is substituted ἀνακολούθως for εἰκός ἐστι— a tolerably common form of anacoluthon in Plato and elsewhere (I 347 A note): but such an idiom is awkward here. It is difficult again to supply δῆλον from Glauco's answer, though the presence of δῆλον may render the anacoluthon a trifle easier; nor can a governing verb be elicited from προστετάξεται. Others propose to abolish the anacoluthon: Ast by reading ὅ γε νεώτερος, Hartman by emending to προστετάξεσθαι <δῆλον>. Δῆλον. Καὶ μὴν κτλ. Neither alternative is satisfactory: and Hartman's is not even Greek. It should be noted that Aristophanes deals with the same subject in Eccl. 638 ff. See App. I.

ἄρχοντες . Stallbaum reads οἱ ἄρχοντες with q. “At varii sunt in civitate magistratus, neque semper eorundem nedum omnium est, tale quid mandare iunioribus” (Schneider).

τὸ ‐‐ βοηθεῖν . τό belongs to δέος, “ut sensus idem sit, ac si dictum esset δέος δὲ τὸ τῆς τῶν ἄλλων βοηθείας τῷ πάσχοντι. δέος ἐστὶ τοὺς ἄλλους βοηθεῖν quin recte dicatur, nemo ambigit: quidni etiam τὸ τοὺς ἄλλους βοηθεῖν δέος dicere liceat” (Schneider)? Cf. οὐ παρὰ φύσιν τὴν τοῦ θήλεος πρὸς τὸ ἄῤῥεν (466 D). Madvig's change of τό to τοῦ has met with much favour, and is accepted even by J. and C. To my mind it destroys the balance of the two clauses, by dropping the personification of δέος, while retaining that of αἰδώς. For the sense cf. Ar. Eccl. 641 —643 ἀλλ) <*> παρεστὼς οὐκ ἐπιτρέψει· τότε δ) αὐτοῖς οὐκ ἔμελ) οὐδὲν | τῶν ἀλλοτρίων (sc. πατέρων) ὅστις τύπτοι· νῦν δ’ ἢν πληγέντος ἀκούσῃ, | μὴ αὐτὸν ἐκεῖνον (illum ipsum sc. suum ipsius parentem, as Blaydes explains) τύπτῃ δεδιὼς τοῖς δρῶσιν τοῦτο μαχεῖται. Aristophanes' verses illustrate τοὺς μὲν ὡς ὑεῖς exactly; the parallel could scarcely be closer. Cf. App. I.

οὐδὲν δεινὸν μή . This construction occurs only four times in the Platonic corpus: viz. in Ap. 28 A , Phaed. 84 B , Gorg. 520 D , and Epp. 7. 344 E (Weber in Schanz's Beiträge II 2, p. 50).

κολακείας κτλ. πένητες has been variously explained as (1) for <αἷς ἔνοχοι ἂν εἶεν> πένητες or the like (Schneider), (2) in partitive apposition with the subject of ἀπηλλαγμένοι ἂν εἶεν (one of J. and C.'s alternatives), (3) nominative to ἴσχουσι (Shorey in A. J. Ph. XVI p. 237). J. and C. also suggest that κολακείας is “genitive singular in the same case as ὧν.” If so, we should read ἀλγηδόνος with q: but there is no room for doubt that κολακείας is the accusative plural. Of these interpretations (1) is too difficult, while (3) is hardly possible, unless πένητες is placed after ἴσχουσι, as was once proposed by Ast, who afterwards preferred to read ἀπηλλαγμένοι ἂν εἶεν <πένητες>, and finally wished to excise the word altogether. (2) is, I think, defensible, if we remember the Greek partiality for this kind of construction (IV 431 A note), and the occasional irregularities of Platonic style. See also on VIII 556 C, D. Jackson conjectures πένητος (‘the poor man's flatteries of the rich’), Stallbaum πενίας in the sense of πενήτων. I think πένητες is probably due to Plato: but if not, the word may be a gloss on κολακείας τε πλουσίων or on ἴσχουσι .

οἰκετῶν : not= οἰκείων as the Scholiast says, but domestici, ‘those of the household’ ( οἱ κατὰ τὸν οἷκον πάντες Hesychius), including, of course, slaves. Where there is no οἰκία, as in Plato's eity, there can be no οἰκέται. Plato's communism involves the abolition of domestic slavery as well as of family ties. See also on 469 B, C.

τὰ μὲν ‐‐ παραδόντες : an interesting glimpse of the economic condition of the Athenian poor. Cf. Ar. Clouds 1172 ff. The agreement in tense makes it probable that πορισάμενοι, θέμενοι, and παραδόντες are grammatically coordinate; although the money must of course be procured before it is deposited. The asyndeton has a rhetorical effect: cf. II 362 B note Hartman would omit παραδόντες; but παραδιδόναι takes an infinitive more easily than τίθεσθαι .

ὅσα τε κτλ. : ‘and the various and manifold troubles which men suffer in connexion with such matters, all of them obvious enough and ignoble, and not worth spending words upon.’ δειλά τε for δῆλά τε δή has slight MS authority, but is only an absurd attempt to represent δἰ ἀπρέπειαν in C above. Still worse is the conjecture δοῦλά, which Herwerden approves.

ἀπαλλάξονται . I formerly adopted Cobet's conjecture ἀπηλλάξονται ( N. L. p. 243), which is attractive in itself, and also because of its correspondence with ἀπηλλαγμένοι ἂν εἷεν in C. But even on the score of meaning the change can hardly be called a necessary one, and there is no MS authority for the form ἀπηλλάξονται either here or (so far as I can discover) elsewhere.

ὀλυμπιονῖκαι κτλ. ‘To him that overcometh’ etc. Plato frequently borrows similitudes and phrases from the national games. Cf. VI 503 A, 504 A, IX 583 B note, X 613 B, C, 621 D, and Phaedr. 256 B . Here he sings a sort of paean in honour of his more than Olympic conquerors. νίκη, ἡ <*>κ τοῦ δημοσίου τροφή (cf. Ap. 36 D ), ἀναδοῦνται, γέρα (such as προεδρία Xenophanes Fr. 2. 7) and ταφῆς ἀξίας μετέχουσιν are each of them significant points in the comparison.

ὧν ‐‐ ὑπάρχει . The nominative of a relative pronoun is very rarely attracted into the genitive. Van Cleef ( de attract. in enunt. rel. usu Plat. p. 42) cites only two other certain instances in Plato, viz. Theaet. 158 A and Alc. II 148 A. περὶ πάντων ὧν γέγονε is found in an Attic inscription about the end of the fourth century B.C. (Meisterhans^{3} p. 238). In Phaed. 69 A the nominative passes into a dative: cf. also οἷς ἐξόν in 466 A and Gorg. 492 B .

ζῶντές τε . We should expect τε to follow γέρα, but cf. 452 A. Here, as there, one or two MSS (with Stobaeus Flor. 43. 102 ad fin.) omit τε. Hartman is suspicious of ταφῆς ἀξίας μετέχουσιν, especially as καὶ μάλα—καλά refers to γέρα. καλά might conceivably be the marginal comment of an approving reader; but this kind of looseness is not uncommon in replies (cf. II 372 A, III 405 D, IV 436 E, 468 A, VI 500 B, VII 535 C, VIII 558 A, B, Gorg. 467 E and elsewhere, with Riddell Digest of Platonic Idioms § 306), and the expression ταφῆς ἀξίας μετέχουσιν is much too quiet and refined for the ordinary scribe.

οὐκ οἶδα ὅτου : said with a glance at Adimantus, who had been the spokesman of these views (IV 419 A ff.). Cf. the use of τισίν in II 372 E.

ποιοῖμεν ‐‐ σκεψοίμεθα . See cr. nn. I agree with most of the recent editors in writing the optative. σκεψόμεθα is perhaps defensible, for we may regard τοῦτο μὲν—σκεψόμεθα as oratio recta; but ποιοῦμεν would be very awkward, if not positively wrong, in view of the optative ὡς οἷοί τ’ εἶμεν. It is noticeable that Plato did not expressly promise to examine this point; although the solution is already hinted at in IV 420 B.

οἷς ἐξόν . Hirschig would write οἵ for οἷς, but see 465 D note The same attraction is found in other authors besides Plato: see Kühner Gr. Gr. II p. 925.

ἐπικούρων has now a more exalted sense than formerly (see 463 B, 464 B notes), and includes the Rulers. Aristotle perversely misrepresents Plato's position in regard to the happiness of the guardians when he remarks ἔτι δὲ καὶ τὴν εὐδαιμονίαν ἀφαιρούμενος τῶν φυλάκων, ὅλην φησὶ δεῖν εὐδαίμονα ποιεῖν τὴν πόλιν τὸν νομοθέτην ( Pol. B 5. 1264^{b} 15 ff.): see Susemihl ad loc.

ἐκεῖ . IV 420 ff.

διὰ δύναμιν : ‘because he has the power,’ “weil er kann” (Schneider). The possession of the power to do wrong is itself a temptation, according to Plato: cf. Gorg. 525 D οὗτοι (tyrants etc.) γὰρ διὰ τὴν ἐξουσίαν μέγιστα καὶ ἀνοσιώτατα ἁμαρτήματα ἁμαρτάνουσι, and ib. 526 A. Whibley points out that in the language of Greek politics and political science δύναμις was often used in a quasitechnical sense, denoting ‘power due to wealth, connexions,’ etc. ( Gk. Olig. p. 125 note 7), but it can hardly have such a meaning here. Madvig conjectures, absurdly enough, διαδῦναι.

Ἡσίοδον . OD. 40.

μενεῖ ἐπί : ‘will remain true to,’ as in VI 496 B.

συγχωρεῖς is followed first by the accusative κοινωνίαν and afterwards by the accusative with infinitive κατά τε πόλιν—ἄῤῥεν (J. and C.). Ast desired to cancel καί before παίδων, and is commended for this by Hartman, who remarks “quasi unquam παῖδες gigni possint sine mulieris et viri κοινωνίᾳ!” “Nodum in scirpo,” as Schneider caustically observes. Plato is speaking of κοινωνία περὶ παίδων not between one woman and one man, but between several women and several men ( τῶν γυναικῶν τοῖς ἀνδράσι). The children are common to all the guardians of either sex.

οὐ παρὰ φύσιν . Before taking leave of the subject, Plato reiterates the principle on which his communism rests. “Equal companionship in the work and interests of life is the natural relation of the sexes, whereas it is the existing relation which is unnatural” (Bosanquet). Cf. 456 C note

ὥσπερ ‐‐ ζῴοις . Cf. 451 D.

περὶ μὲν γὰρ κτλ. = ‘for as to war’ etc. is a dexterous way of making room for the episode on war, and at the same time postponing ‘the great peripeteia, the on-rushing of the third wave,’ which “is made more impressive by being delayed” (J. and C.). For μὲν γάρ cf. VIII 562 A note

ὥσπερ κτλ. Handicrafts were usually hereditary among the Greeks: cf. Prot. 328 A and Blümner Privatalt. p. 395 notes διακονεῖν should be taken with ἄξουσι. The change of construction is illustrated by Schneider ( Addit. p. 41) from Tim. 74 B ἐμηχανᾶτο, ἵνα—παρέχοι, τὴν δὲ σάρκα—ἔσεσθαι κτλ. Herwerden inserts δεῖ, and Richards διδάσκωνται, after θέᾳ, but the text is probably sound.

μαχεῖται ‐‐ τέκῃ . Cf. Xen. Cyr. IV 3. 2 and Tac. Germ. 7 quodque praecipuum fortitudinis incitamentum est —in proximo pignora, unde feminarum ululatus audiri, unde vagitus infantium.

ἀναλαβεῖν =‘to recover.’ This intransitive use of ἀναλαμβάνειν is especially common in medical writers: see Stephanus-Hase Lex. s.v. It arises from the omission of the reflexive pronoun, which is a common way of making transitive verbs into intransitive: see on I 336 B.

παῖδας τοὺς ἄνδρας. Ξ with several other MSS reads τοὺς παῖδας instead of παῖδας. But παῖδας is predicative, and goes with θεωρεῖν. “Socrates plurimum referre dicit, ut qui adulti bellicosi futuri sint, iam pueri res bellicas spectent” (Schneider). Hartman seriously weakens the contrast between παῖδας and ἄνδρας by reading ἄνδρας <τοὺς> πολεμικούς.

διαφέρει . We should at first sight expect <πολὺ> διαφέρει, and so Richards suggests. But (as Hartman points out) the introduction of οὐκ ἄξιον κινδύνου breaks the continuity between the original question and the reply. Hence, too, the reply has διαφέρει, not διαφέρειν (the reading of Ξ and a few other MSS, wrongly preferred by Hartman).

τοῦτο μὲν κτλ. ὑπαρκτέον= δεῖ <*>πάρχειν (intransitive), not ‘we must begin with,’ as J. and C. suppose. Cf. ἑκτέον 468 A. τοῦτο (accusative: see on III 400 D) is explained by θεωροὺς— ποιεῖν. With προσμηχανᾶσθαι, δεῖ or the like is understood out of ὑπαρκτέον: cf. Gorg. 492 D τὰς μὲν ἐπιθυμίας φὴ|ς οὐ κολαστέον—<*>ῶντα δὲ αὐτὰς ὡς μεγίστας πλήρωσιν—ἑτοιμάζειν and Crito 51 C. Richards needlessly proposes to read προσμηχανητέον or to insert “something like δεήσει.”

εὐλαβήσονται : sc. ἄγειν.

παιδαγωγούς . The tutorial office in Athens was assigned to slaves. In Plato it is exercised by the very best of the citizens. Bosanquet justly emphasizes the revolution which Plato's arrangement would involve in the education of the young.

ἀλλὰ γάρ . II 365 C note

διδαξαμένους . Schneider reads διδαχθέντας, while preferring his own conjecture δεδιδαξομένους. The future διδαξομένους cannot be right: for the children would certainly be taught to ride, before going on such expeditions (J. and C.). It would be too hazardous in such a case ἐν πίθῳ κεραμεύειν. Against Schneider's conjecture it may be urged that the future perfect participle should not be used where the aorist participle is enough. διδαχθέντας is an obvious ‘correction.’ With διδαξαμένους the meaning is simply ‘when they have taught them to ride.’ The middle expresses personal interest; and does not imply that the ἐπίκουροι get them taught by others. See on this point IV 421 E note It may be noted that in Sparta great importance was attached to learning the accomplishment of riding (Müller Dorians II p. 316).

τί δὲ δὴ κτλ. This punctuation is better than to place the mark of interrogation after δή, and take τὰ περὶ τὸν πόλεμον as an internal accusative with πῶς ἑκτέον κτλ., because τὰ περὶ τὸν πόλεμον is already practically involved in the word στρατιώτας. I agree with Hartman that Richards' proposal— τί δὲ δή; εἶπον· τὰ περὶ τὸν πόλεμον, πῶς— πολεμίους, ἆρα κτλ.;—is far from elegant.

ποῖα . See cr. n. ποἶ ἄν, which is generally read, surely cannot be right. Schneider remarks “ ποἶ ἄν breviter dictum accipio pro ποῖα ἂν ὄντα τὰ περὶ τὸν πόλεμον ὀρθῶς ἔχειν λέγεις.” J. and C. are content with supplying εἴη τὰ σοὶ καταφαινόμενα. But ellipses of this kind are too severe a strain upon the imagination. ποῖα δή is suggested by Richards, πῇ δή by Hartman: but is δή in place here? I think not. I take ποῖα sc. ἐστι to refer to τὰ περὶ τὸν πόλεμον. Glauco addresses himself to the first of Socrates' questions: cf. 465 E note and Soph. Trach. 421 —423. The corruption is common enough: see Introd. § 5.

αὐτῶν = ‘ipsorum’ contrasts Plato's soldiers with their enemies (cf. πρὸς α<*>τούς τε καὶ τοὺς πολεμίους just before). μέν prepares us for the second part of this topic, beginning at 469 B. We certainly should not read μήν (with Hartman). Plato's treatment of cowardice in battle may be compared with the punishment of τρέσαντες in Sparta: see Gilbert Gk. Constit. Ant. E.T. p. 77. Cf. also Laws 943 D ff.

ἑλοῦσι . Van Leeuwen's emendation—see cr. n.—seems to me admirable. The contrast between ἁλόντα and ἑλοῦσι is precisely what is wanted: cf. Xen. Cyr. VII 5. 73 νόμος γὰρ ἐν πᾶσιν ἀνθρώποις ἀΐδιός ἐστιν, ὅταν πολεμούντων πόλις ἁλῷ, τῶν ἑλόντων ἐ=ναι καὶ τὰ σώματα— καὶ τὰ χρήματα. With the infinitive van Leeuwen compares Laws 879 A παραδότω τὸν δοῦλον—χρῆσθαι ὅ τι ἂν ἐθέλῃ. θέλουσι is not free from objection. Paris A generally has ἐθέλω, the usual Attic form; moreover, the word itself, if taken with χρῆσθαι, is too weak; nor can we (with J. and C.) readily understand ἔχειν. Plato's ordinances on this matter are far more drastic than anything known even in Sparta: see Müller Dorians II p. 238.

οὐκέτι σοι δοκεῖ : said with playful irony, for Glauco is an ἀνὴρ ἐρωτικός (474 D). A vein of irony runs through all this passage, as Dugas has pointed out ( L'Amitié Antique p. 121); but it is not wholly ironical. Plato may have been willing to allow more latitude to soldiers on a campaign than he would permit to others, without sanctioning the usual abuses of camp life (see Dugas l.c. p. 87). There is nothing in this passage which is necessarily inconsistent with the self-restraint enjoined in III 403 B, although in practice abuses might have arisen. See also Laws 636 C ff.

καὶ προστίθημί γε κτλ. Glauco's enthusiasm is in keeping with his character: see last note.

βούληται : sc. ὁ ἀριστεὑσας τε καὶ εὐδοκιμήσας.

ἐάν τις ‐‐ φέρειν . See Symp. 178 E — 179 B. The principle underlying Glauco's remark was widely accepted by Greek military authorities (see Hug on Symp. l.c. and Dugas l.c. pp. 90—104). The Theban Sacred Band, composed of ἐρασταί and ἐρώμενοι, is the best-known instance of its application in actual warfare (Athen. XIII 561 F).

αἱρέσεις means selections by the rulers (so also Schneider): cf. 460 B, to which εἴρηται ἤδη refers. J. and C.'s alternative rendering “success in winning such prizes” cannot stand: still less the translation of D. and V. “to exercise more than the usual liberty of choice in such matters.”

τῶν τοιούτων : i.e. τῶν ἀγαθῶν.

Ὅμηρος κτλ. Il. 7. 321 f. νώτοισιν δ’ Αἴαντα διηνεκέεσσι γέραιρεν | ἥρως Ἀτρείδης. In Plato, Αἴαντα is omitted by q, and three other MSS: one MS places it before νώτοισιν, and four after ἔφη. The word may be a gloss; but as it is present in ΑΠΞ, in the same position as in Homer, it is safer to retain it. Plato often makes his Homeric quotations complete, even at the cost of a little awkwardness: cf. II 363 B. Aristophanes, it may be noted, has the converse of Plato's proposal in Eccl. 680.

γε reminds us that Homer is not in other respects a persona grata in our city.

ἕδραις κτλ. ἕδρῃ τε κρέασίν τε ἰδὲ πλείοις δεπάεσσιν in Il. VIII 162 al.

στρατείας : not of course στρατιᾶς (Herwerden), for στρατιά is ‘army,’ στρατεία ‘campaign.’

τοῦ χρυσοῦ γένους . III 415 A. Cf. Heracl. Fr. 102 ed. Bywater ἀρηιφάτους θεοὶ τιμῶσι καὶ ἄνθρωποι.

τοῦ τοιούτου γένους . Plato compares his ‘golden citizens’ with the heroes of the Hesiodic golden age. He would fain surround them with some of the romantic and religious sentiment that clung around the golden age of Greek poetry and legend.

οἱ μὲν ‐‐ ἀνθρώπων . Cf. Crat. 397 E . The nearest approach to these lines in our Hesiod is to be found in OD. 122 f. τοὶ μὲν—the departed children of the golden age— δαίμονές εἰσι Διὸς μεγάλου διὰ βουλὰς | ἐσθλοί, ἐπιχθόνιοι, φύλακες θνητῶν ἀνθρώπων.

τοῦ θεοῦ . Apollo, our πάτριος ἐξηγητής: see IV 427 B note

τιθέναι : ‘to bury.’

τίνι διαφόρῳ : ‘with what distinction’ (“mit welcher Auszeichnung” Schneider). The occurrence of θήκας διαφόρους in Laws 947 B is no ground for reading <θήκῃ> τίνι διαφόρῳ here, as Richards bids us read.

ὡς δαιμόνων ‐‐ θήκας is another link with Greek religion. Cf. Eur. Alc. 1000 ff. καί τις δοχμίαν κέλευθον | ἐμβαίνων τόδ’ ἐρεῖ | “ αὕτα ποτὲ προὔθαν’ ἀνδρός, | νῦν δ’ ἐστὶ μάκαιρα δαίμων. | χαῖρ’ ὦ πότνἰ, εὖ δὲ δοίης.” | τοῖαί νιν προσεροῦσι φᾶμαι, and other passages cited by Nägelsbach Nachhom. Theol. pp. 108—110.

Ἕλληνας ‐‐ ἄλλῃ . Ἕλληνας is the object, not, as is sometimes held, the subject, of ἀνδραποδίζεσθαι. It rightly occupies the emphatic place, because the point is that Greek cities should not enslave Greeks—no one objects to their enslaving barbarians,—and not that Greeks (as opposed to barbarians) should not enslave Greek cities. Cf. the order in 471 A οὐδ’ ἄρα τὴν Ἑλλάδα Ἕλληνες ὄντες κεροῦσιν. A further reason for taking this view is that Ἑλληνίδας πόλεις points the allusion to Plato's city, which is a Ἑλληνὶς πόλις (470 E), and therefore will not reduce Greeks to slavery. Finally, μηδ’ ἄλλῃ (sc. Ἑλληνίδι πόλει) is easy and natural only if Ἑλληνίδας πόλεις is treated as the subject. The difficulty of μηδ’ ἄλλῃ (on the usual interpretation) led to the correction μηδ’ ἄλλοις (Stallbaum with v and Flor. RT), and has recently caused Hartman to propose μηδαμῇ, on the ground that ἄλλῃ after Ἕλληνας could only mean βαρβάρῳ. In so saying, he goes. I think, too far; but my explanation removes the difficulty.

ἐθίζειν : sc. τοὺς Ἕλληνας.

ες)λαβουμένους agrees with the subject of φείδεσθαι rather than with that of ἐθίζειν. The Spartan Callicratidas agreed with Plato here: οὐκ ἔφη ἑαυτοῦ γε ἄρχοντος οὐδένα Ἑλλήνων εἰς τοὐκείνου δυνατὸν ἀνδραποδισθῆναι ( Xen. Hell. I 6. 14 ). To enslave barbarians, on the other hand, is just: for the barbarian is φύσει δοῦλος ( Eur. Iph. Aul. 1401 and elsewhere: Arist. Pol. A 2. 1252^{b} 9). See also on 470 C.

ὅλῳ καὶ παντί . So in Phaed. 79 E , Crat. 433 E . In VII 527 C we have τῷ ὅλῳ καὶ παντί, and even τῷ παντὶ καὶ ὅλῳ in Laws 734 E.

μηδέ : with ἐκτῆσθαι. They must neither enslave their countrymen ( ἀνδραποδίζεσθαι above), nor hold a Greek in slavery: cf. I 351 B. J. and C. wrongly translate μηδέ as ‘not even,’ and Hartman needlessly proposes μηδέν). Greek slaves were of foreign nationality, except such as had been sold into slavery on the destruction of their city by war (Blümner Privatalt. p. 87 note 1). Plato disapproves of the exception: does he mean to approve the rule, so far as his own city is concerned? Steinhart ( Einleitung p. 202) asserts that Plato expressly recognises slavery in his State. It is clear from the present section that Plato does not impugn the principle of slavery, so long as the slaves are of barbarian origin; but he nowhere says that his perfect city is actually to contain slaves, nor is it easy to see what there would be for them to do, unless they were employed to work under the farmers and artizans, or as personal attendants at the συσσίτια and the like. Slaves are present, of course, in the city of the Laws (776 C ff.).

σκυλεύειν ‐‐ καλῶς ἔχει . Cf. Xen. Hell. II 4. 19 (quoted by J. and C.) καὶ τὰ μὲν ὅπλ<*> ἔλαβον, τοὺς δὲ χιτῶνας οὐδενὸς τῶν πολιτῶν ἐσκύλευσαν. Such moderation was unusual.

ἀποπταμένου is (as Schulze pointed out in Fl. Jahrb. 1887 pp. 226 ff.) a reminiscence of Homer's ἀπὸ δ’ ἔπτατο θυμός ( Il. 16. 469 and elsewhere). Hence the poetic form, as in οἴχεται ἀποπτάμενος ( Symp. 183 E from Il. II 71). The ordinary aorist in prose is ‐επτόμην, as in II 365 A. Compare Phaed. 115 C f. and Eur. Fr. 176 . 3—6 τίς γὰρ πετραῖον σκόπελον οὐτάζων δορὶ | ὀδύναισι δώσει; τίς δ) ἀτιμάζων νέκυς, | εἰ μηδὲν αἰσθάνοιτο τῶν παθημάτων; and Plut. Apophtheg. Lac. 228 F.

αἳ ‐‐ ἁπτόμεναι . Aristotle read βάλλοντος, and not βαλόντος (see cr. n.), as appears from Rhet. III 4. 1406^{b} 33, where he refers to Plato's illustration as follows: καὶ τὸ ἐν τῇ πολιτείᾳ τῇ Πλάτωνος, ὅτι οἱ τοὺς τεθνεῶτας σκυλεύοντες ἐοίκασι τοῖς κυνιδίοις, ἃ τοὺς λίθους δάκνει τοῦ βάλλοντος οὐχ ἁπτόμενα. The present is more picturesque and true to nature: the dog worries the stones, while his tormentor amuses himself by throwing more. It is true that the simile is not quite accurate, because a ‘flown antagonist’ cannot continue to do mischief; but βαλόντος, which is generally read, though not by Schneider, is also inexact, because you cannot attack a vanished foe. In either case, the analogy is near enough. Moreover the consensus of all the other MSS, coupled with Aristotle, outweighs the authority of A where lipography is possible. See Introd. § 5.

ἀναιρέσεων . The laws of Greek warfare permitted ἀναίρεσις of the dead, unless the petitioning parties had forfeited their rights by robbing a temple or desecrating a shrine (Busolt Gr. Alterth. p. 55, where the authorities are cited).

οὐδὲ μὴν ‐‐ Ἑλλήνων : as was usual in Greece: see for example Thuc. III 114. 1. Plutarch however implies that the Spartans were an honourable exception to this rule ( Apophtheg. Lac. 224 B). With Plato's sentiment cf. “aeternum inimicitiarum monumentum Graios de Graiis statuere non oportet” (Cic. de Inv. II 70. Cicero is referring to an incident arising out of a war between Sparta and Thebes).

Stable identifiers

Citation URI: http://catalog.perseus.org/citations/urn:cts:greekLit:tlg0059.tlg030.perseus-grc2:5.469

Text URI: http://catalog.perseus.org/texts/urn:cts:greekLit:tlg0059.tlg030.perseus-grc2

Work URI: http://catalog.perseus.org/texts/urn:cts:greekLit:tlg0059.tlg030

Catalog Record URI: http://catalog.perseus.org/catalog/urn:cts:greekLit:tlg0059.tlg030.perseus-grc2

Plato Republic 5.469

ed. John Burnet

Search the Perseus Catalog for:

Current edition Perseus

Commentary

Stable identifiers