Πρωταγόρας ἢ σοφισταί· ἐνδεικτικός . This is how the title appears in B, except that ἐνδεικτικός (which is absent in T) is apparently from a later hand. B denotes the Bodleian MS. = Bekker's U: T a manuscript in the library of St Mark's at Venice (append. class. 4 cod. 1) = Bekker's t. The former is the best representative of the MSS. of the first class: the latter is supposed by Schanz to be the source of all the MSS. of the second class. Plato probably called the dialogue Πρωταγόρας (compare the titles , etc.— the majority of the dialogues are named after an interlocutor): the words ἢ σοφισταί and ἐνδεικτικός were added by Thrasylus, whose date is about the Christian era. The is the sole example of an ‘endeictic’ dialogue: it is so called because it was thought to be an arraignment ( ἔνδειξις ) of the Sophists. That this is the meaning of ἐνδεικτικός , and not ‘probative’, can be inferred from the fact that Albinus used ἐλεγκτικός in its place ( εἰσαγωγὴ εἰς τοὺς Πλάτωνος διαλόγους , ch. v. Cobet takes the right view: see his edition of Diog. Laert. III. 51). The arrangement of the Platonic dialogues into tetralogies is also probably due to Thrasylus. The is the second dialogue of the sixth tetralogy, the others being (1) (3) (4) See Grote's 1, 158 ff.

Socrates meets a friend, who asks him to describe his interview with Protagoras.

1. πόθεν—ὥραν . The opening words of Cicero's translation of the are preserved by Priscian (VI. 63): quid tu? unde tandem appares, o Socrate? an id quidem dubium non est, quin ab Alcibiade? ἢ = Latin an? introduces a second question intended to forestall Socrates' answer to the first: cf. 26B πῶς με φὴ|ς διαφθείρειν, ὦ Μέλητε, τοὺς νεωτέρους; ἢ δῆλον δὴ ὅτι κτλ.;

7. ὃς ἔφη χαριεστάτην ἥβην κτλ. Homer, XXIV. 348 and X. 279 πρῶτον ὑπηνήτῃ τοῦπερ χαριεστάτη ἥβη. To insert τήν (with Hirschig) before ἥβην would make the reference to Homer less precise. The line in Homer refers to Hermes, and Sauppe quotes Clement to show that sculptors modelled their busts of Hermes after Alcibiades.

11. εὖ ἔμοιγε ἔδοξεν , sc. διακεῖσθαι . Socrates replies to his friend's second question ( καὶ πῶς πρὸς σὲ ὁ νεανίας διάκειται ;) first, and to his first question ( ἦ παρ’ ἐκείνου φαίνει ; ) second, in the words καὶ οὖν καὶ ἄρτι ἀπ’ ἐκείνου ἔρχομαι , where οὖν marks the regression to the earlier inquiry. Both B and T read ἄρχομαι by mistake for ἔρχομαι : ἔρχομαι is found in a Vienna codex (suppl. phil. gr. 7) which Kral and Wohlrab place along with B in the first class of MSS.

14. οὔτε προσεῖχον—ἐπελανθανόμηντε. τεfollowing οὔτε throws emphasis on the second clause: e.g. 26C παντάπασί με φὴ|ς οὔτε αὐτὸν νομίζειν θεοὺς τούς τε ἄλλους ταῦτα διδάσκειν. The idiom is very common in Plato (e.g. below 347E 360D 361E and corresponds to neque—que or (more frequently) neque—et in Latin. For the interchange of pronouns ἐκείνου— αὐτοῦ see on 310D

19. καὶ πολύ γε , i.e. καλλίονι ἐνέτυχον .

23. Ἀβδηρίτῃ . Abdera, on the coast of Thrace, was the birthplace of Democritus and of Protagoras. The reputation of the city for heaviness and stupidity seems not to be earlier than the age of Demosthenes: see pseudo-Dem. περὶ τῶν πρὸς Ἀλέξανδρον , 23 ὥσπερ ἐν Ἀβδηρίταις ἢ Μαρωνείταις ἀλλ’ οὐκ ἐν Ἀθηναίοις πολιτευόμενοι and Cic. ad Atticum, VII. 7. 4.

25. υἱέος . Schanz writes ὑέος in conformity with the general usage of inscriptions about Plato's time, but MSS. upon the whole favour υἱέος : see the Editor's note on 45C.

26. τὸ σοφώτατον . So the MSS.: Schanz and others read σοφώτερον , apparently the reading of Ficinus, who translates the word by sapientius. Socrates, however, is thinking of Protagoras, who is not σοφός , but σοφώτατος (l. 31): the effect of the neuter τὸ σοφώτατον is to generalise the statement into a kind of adage. σοφώτερον would introduce a somewhat frigid comparison between Alcibiades and Protagoras in respect of wisdom; and it should also be noted that the MSS. reading σοφώτατον was more likely to be changed to σοφώτερον by mistake than vice versa. There may be an allusion to some proverbial form of speech resembling that in Theognis, 255 κάλλιστον τὸ δικαιότατον· λῷστον δ’ ὑγιαίνειν κτλ. : cf. also the scolion referred to in 451E. The sentiment is an interesting anticipation of the Stoic paradoxes as to the beauty of the wise man.

28. ἀλλ’ ἦ expresses surprise and interest: ‘What! have you just left etc.’ So in 447A ἀλλ’ ἦ τὸ λεγόμενον κατόπιν ἑορτῆς ἥκομεν ;Presently μὲν οὖν is as usual corrective: see on 26B.

31. σοφώτατος εἶναι Πρωταγόρας . The interest is sustained by reserving the name of Protagoras to the end.

35. πάνυ γε πολλὰ καὶ εἰπὼν κτλ. Sauppe places a comma after πάνυ γε , but it suits the rapid movement of the dialogue better to take πάνυ with πολλά .

36. τί οὖν οὐ διηγήσω . Literally ‘why didn't you relate’, i.e. ‘tell us at once’. So in 317D τί οὖν—οὐ καὶ Πρόδικον καὶ Ἱππίαν ἐκαλέσαμεν ; This construction of τί οὐ and τί οὖν οὐ is common in animated conversational style, especially with the second person: e.g. 503B τί οὐχὶ καὶ ἐμοὶ αὐτὸν ἔφρασας τίς ἐστιν ;= φράσον ὅτι τάχιστα—οὐκ ἂν φθάνοις φράζων as Thompson remarks.

τὴν ξυνουσίαν recalls συγγεγονώς in l. 34. The continual use of the words συνεῖναι, συγγίγνεσθαι, πλησιάζειν, ἰέναι ἐπί, ἐρᾶν and the like to denote the relation between learner and teacher in Plato's dialogues depends upon the conception of the philosophical impulse as ἔρως : see 210.

37. ἐξαναστήσας τὸν παῖδα τουτονί . The slave was doubtless in attendance on the Friend.

40. ἀκούητε . From this, as well as from ἡμεῖς and ἀκούετε , it appears that the Friend was not the only listener.

42. διπλῆ ἂν εἴη ἡ χάρις . The expression is almost proverbial: cf. (with Schneidewin on Soph. 1370) Eur. 162-3 παντὶ γὰρ προσκείμενον κέρδος πρὸς ἔργω? τὴν χάριν τίκτει διπλῆν, and Eur. 333-4 τῷδέ τ’ εἴρηκας καλῶς κἀμοί· διπλοῦν δὲ χάρμα γίγνεται τόδε.

2. ἀποπειρώμενος—διεσκόπουν. ῥώμηof strength of will and resolution (cf. ἀνδρεία above 310D as in 259C πρὸς τὴν τῆς ψυχῆς σύνεσιν καὶ ῥώμην. γνώμης , the suggestion of Hoenebeek, would be much less forcible and exact.

6. ὥσπερ ἂν εἰ κτλ. ἄνgoes with ἀπεκρίνω in l. II. The application of the similes follows in D, and each illustration contains two subordinate protases, viz. (in B) (1) εἰ ἐπενόεις— ἐκείνῳ , (2) εἴ τίς σε ἤρετο , and (in C) (1) εἰ δὲ—ἐκείνοις , (2) εἴ τίς σε ἤρετο . With this multiplication of protases Heindorf compares 74B μανθάνεις γάρ που ὅτι οὑτωσὶ ἔχει περὶ παντός· εἴ τίς σε ἀνέποιτο τοῦτο ὃ νῦν δὴ ἐγὼ ἔλεγον, τί ἐστι σχῆμα, ὦ Μένων; εἰ αὐτῷ εἶρες ὅτι στπογγυλότης, εἴ σοι εἶρεν ἅρεπ ἐγώ, ρότεπον σχῆμα ἡ στπογγυλότης ἐστὶν ἢ σχῆμά τι; εἶρες δή ρου ἂν ὅτι σχῆμά τι. The reasoning from analogy is quite in Socrates' style: see for example 448Bff.

7. ‘Ιπποκράτη—τὸν τῶν’ Ασκληπιαδῶν . It is unnecessary (with Naber) to bracket Ἱπποκράτη . Hippocrates, the founder of medical science, born about 460 B.C. in Cos, was at the height of his renown about the year 400. He is referred to also in 270C ( εἰ μὲν οὖν Ἱπποκράτει γε τῷ τῶν Ἀσκληπιαδῶν δεῖ τι πείθεσθαι ) as holding the view that the human body could not be Compare Ar. 383 εἰπέ μοι τί πάσχετ’ ὦνδρες; Pl. 283B εἰπέ μοι, ἔφη, ὦ Σώκρατές τε καὶ ὑμεῖς οἱ ἄλλοι . The exhortation or command is in general primarily addressed to one man: contrast, however, below, 330B and 186E σὺ δ’, ὦ Λάχης καὶ Νικία, εἴπετον ἡμῖν ἑκάτερος.

28. ἄλλο γε , i.e. other than the name ‘Protagoras’. It is more in accordance with Plato's usage to make the first question end with the first ἀκούομεν , and regard the ὥσπερ clause as introducing the second question τί τοιοῦτον περὶ Πρωταγόρου ἀκούομεν ; This second question is rejected by Cobet as spurious, but the punctuation which we have adopted seems to remove the difficulty. The ὥσπερ clause defines in advance the meaning of τοιοῦτον : for this and for the asyndeton compare 258B-C πότερον οὖν—δεῖ θαρροῦντα ἤδη λέγειν ὅτι τὸ μὴ ὂν βεβαίως ἐστὶ τὴν αὑτοῦ φύσιν ἔχον; ὥσρεπ τὸ μέγα ἦν μέγα καὶ τὸ καλὸν ἦν καλὸν καὶ τὸ μὴ μέγα μὴ μέγα καὶ τὸ μὴ καλὸν μὴ καλόν, οὕτω δὲ καὶ τὸ μὴ ὂν κατὰ ταὐτὸν ἦν τε καὶ ἔστι μὴ ὄν, ἐνάπιθμον τῶν πολλῶν ὄντων εἶδος ἕν ; Crat. 394A B; 172Dwith Heindorf's note. See also 330A ὥσπερ τὰ τοῦ προσώπου , where the same punctuation should be adopted. In l. 31 the MSS. read ἠκούομεν by mistake for ἀκούομεν .

31. σοφιστήν—εἶναι . On σοφιστής see below, note on 312C γε implies that Protagoras may be a sophist only in name. εἶναι is frequently used with verbs of naming: cf. 192A ὃ ἐν πᾶσιν ὀνομάζεις ταχυτῆτα εἶναι. Cobet's suggestion to read εἶεν, ἔφην, ὡς for εἶναι, ἔφη. ὡς is ingenious but needless.

34. αὐτὸς δὲ δὴ—παρὰ τὸν Πρωταγόραν , sc. τί ἂν ἀποκρίναιο ;

36. ὑπέφαινέν τι ἡμέρας. ὑποφαίνει ἡμέραis used, as Heindorf says, de die illucescente. Here τι ἡμέρας = ‘something of day’, ‘some daylight’.

37. εἰ μέν τι—ἔοικεν . Heindorf remarks that Stephanus' conjecture τοῦτο ἔοικεν is needless, the vague indefinite subject being frequently omitted. Cf. 387D εἴπερ τι τοῖς ἔμπροσθεν μέλλει ὁμολογούμενον εἶναι; 99E ἴσως μὲν οὖν ὧ? εἰκάζω τρόπον τινὰ οὐκ ἔοικεν.

39. εἰς τοὑς Ἕλληνας σαυτόν. εἰς τοὺς Ἕλληναςgoes with παρέχων : cf. 179B ἱκανὴν μαρτυρίαν παρέχεται—εἰς τοὺς Ἕλληνας. Young men of fashion were fond of looking to Greek, as opposed to Athenian, public opinion. The MSS. have αὑτόν , but the use of the third personal reflexive pronoun for the first and second in the singular number does not seem to be certain in Plato; and it is simplest to suppose that ς fell out after Ἕλληνας : Schanz, VII, xii.

41. ἀλλ’ ἄρα—μὴ οὐ τοιαύτην. οὐgoes closely with τοιαύτην , and μή is virtually ‘perhaps’ (originally ‘lest’ as in ὅρα μή 145B): ‘but perhaps after all ( ἄρα ) this is not the kind of learning which etc.’ Cf. 290E ἀλλ’ ἄρα—μὴ ὁ Κτήσιππος ἦν ὁ ταῦτ’ εἰπώνand 25A ἀλλ’ ἄρα—μὴ οἱ ἐν τῇ ἐκκλησία?, οἱ ἐκκλησιασταί, διαφθείρουσι τοὺς νεωτέρους ; 89C μὴ τοῦτο οὐ καλῶς ὡμολογήσαμεν . In all of these passages it is better to take μή in this way than as equivalent to Latin num. From this use of μή grew up the use of μήποτε = ‘perhaps’, frequent in Aristotle and later, e.g. X. 2. 1173 a . 22 μή ποτ’ οὐ λέγουσιν το αἴτιον .

43. οἵαπερ ἡ παρἀ is the reading of T: B has οἵα περὶ . The γραμματισταί of Athens were schoolmasters, who besides teaching reading and writing (cf. below, 326D translated ( ἑρμηνεύειν ) Homer and interpreted his γλῶτται . They were distinct from the γραμματικοί or κριτικοί who pursued more scientific literary and grammatical studies. The κιθαριστής and γραμματιστής between them taught μουσική , and the παιδοτρίβης γυμναστική ; and μουσική and γυμναστική were the two parts of παιδεία or liberal education: II. 376E.

45. ἐπὶ τέχνῃ—ἀλλ’ ἐπὶ παιδείᾳ : cf. 315A ἐπὶ τέχνῃ μανθάνει, ὡς σοφιστὴς ἐσόμενος . The distinction between professional and liberal ( ὡς τὸν ἰδιώτην καὶ τὸν ἐλεύθερον πρέπει ) education is frequently emphasised by Plato (see especially I. 643D): his word for the latter is always παιδεία , cf. 485A, VI. 492C. The ‘arts’ are throughout the looked on as βάναυσοι , and unfit for men whose souls are free.

3. παρασχεῖν θεραπεῦσαι ἀνδρἱ σοφιστῇ . The word παρέχω is often used of putting oneself in the hands of a doctor: cf. 456B φάρμακον πιεῖν ἢ τεμεῖν ἢ καῦσαι παρασχεῖν τῶ? ἰατρῷ. ἀνήρis regularly used (mostly in a complimentary sense, real or feigned) with words which denote one's profession, standing or the like, e.g. ἀνὴρ μάντις, ἀνὴρ νομεύς, ἄνδρες δικασταί ; cf. 15D. For παρασχεῖν Cobet reads παρέχειν , but if MSS. can be trusted, Plato used the aorist infinitive with μέλλω tolerably often: see Schanz, Preface to p. vii.

6. οὔτ’ εἰ ἀγαθῷ οὔτ’ εἰ κακῷ πράγματι . So in 520B οὐκ ἐγχωρεῖν μέμφεσθαι τούτῳ τῷ πράγματι (their pupils) ὃ αὐτοὶ παιδεύουσιν . Here Socrates uses the most general form of expression because ex hypothesi nothing is yet known as to the sophist: cf. below, 330C ἡ δικαιοσύνη πρᾶγμά τί ἐστιν ἢ οὐδὲν πρᾶγμα ; Cases like 53D οὐκ οἴει ἄσχημον ἂν φανεῖσθαι τὸ τοῦ Σωκράτους πρᾶγμα ;are somewhat different and contain a slight admixture of contempt: see the Editor's note in loc.

7. οἶμαί γ’ εἰδέναι . Hippocrates οἴεται εἰδέναι μὴ εἰδώς : he is thus, according to Socrates, in the worst of all states: cf. ch. VI ff. Socrates now proceeds to convict him of ignorance.

8. ὥσπερ τοὔνομα λέγει—τῶν σοφῶν ἐπιστήμονα . Hippocrates derives σοφιστής from σοφός and ( ἐπ ) ίστ ( αμαι ) ‘quasi sit ὁ τῶν σοφῶν ἴστης ’ (Heindorf, comparing the derivation of Ἥφαιστος in 407Cfrom Φάεος ἵστωρ ). The correct derivation is perhaps given by Suidas s.v.: σοφιστὴς καὶ ὁ διδάσκαλος ὡς σοφίζων (cf. σωφρονίζω =make σώφρων ), but ‐ίζω is very elastic in meaning, and σοφίζω may very well mean ‘play the σοφός ’.

12. τῶν τί σοφῶν. τῶν πρὸς τί σοφῶνwould be more precise, but the accusative of reference is preferred for brevity. For the construction Kroschel compares 125C τῶν τί σοφῶν συνουσίᾳ φὴ|ς σοφοὺς εἶναι τοὺς τυράννους , where, however, σοφῶν is masculine. The neuter of σοφός is not here used ironically as in 157C παρατίθημι ἑκάστων τῶν σοφῶν ἀπογεύσασθαι.

15. ὁ δὲ σοφιστὴς τῶν τί σοφῶν ἐστιν . Heusde suggested ἐπιστήμων after ἐστιν , but the sense can be supplied out of the ‐ιστής of σοφιστής , according to the derivation of the word just given: the full sense is as it were ὁ δὲ σοφιστὴς τῶν τί σοφῶν ἐστιν ( σοφ ) ιστής .

16. τί ἂν ἀποκρινοίμεθα αὐτῷ . There is no need to change the verb to ἀποκριναίμεθα , as was done by Bekker to suit ἔροιτο : cf. below, 354A εἰ ἐροίμεθα—φαῖεν ἄν and 259A εἰ οὖν ἴδοιεν—δικαίως ἂν καταγελῷεν.

17. ποίας ἐργασίας ἐπιστάτης . These words are not of course part of the imaginary questioner's interrogation, but are spoken by Socrates to help out Hippocrates' answer. Socrates suggests that Hippocrates should say that the sophist is ἐπιστάτης of some sort of ἐργασία , but in order that the answer should come from Hippocrates himself, he substitutes for the desired answer: ‘(The sophist is) ἐπιστάτης —of what kind of ἐργασία ?’ Hippocrates then replies by explaining the ἐργασία , viz. τὸ ποιῆσαι δεινὸν λέγειν . The full grammatical construction would be ποίας ἐργασίας ἐπιστάτης ( ὅτι ἐστὶν ἀποκριναίμεθα ἂν αὐτῷ ) ; If we take this view, it is not necessary to insert ἐστίν after ἐπιστάτης (with Hirschig).

Note that ἐπιστάτης is substituted here for ἐπιστήμων : it is clear from 47B, that Plato connected the two words— probably because both contain the syllable ‐ιστ‐ as in ἵστωρ : he frequently plays on the similarity of form between ἐπίσταμαι ἐπιστήμη and ἐπιστατεῖν : see the Editor's note on ἐπιστάτῃ καὶ ἐπαΐοντι in the

17. τί ἂν εἴποιμεν—ὦ Σώκρατες ; Hippocrates is on the verge of ἀπορία , and merely throws out his suggestion ἐπιστάτην τοῦ ποιῆσαι δεινὸν λέγειν for what it is worth. The words ἐπιστάτην τοῦ ποιῆσαι δεινὸν λέγειν are strictly speaking a reply to the question of Socrates ποίας ἐργασίας ἐπιστάτης ; for τοῦ ποιῆσαι δεινὸν λέγειν answers ποίας ἐργασίας , and but for the intervening clause ( τί ἂν εἴποιμεν αὐτὸν εἶναι ; ) the word ἐπιστάτην would have been in the nominative.

Thus explained, the MSS. reading need not be changed. The next best view is to read (with Schanz) τί ἂν <εἰ> εἴποιμεν αὐτὸν εἶναι, ὦ Σώκρατες, ἐπιστάτην τοῦ ποιῆσαι δεινὸν λέγειν ;

22. ὥσπερ ὁ κιθαριστὴς κτλ. Cf. 449Eand 451A ff., where much the same reasoning is employed to discover τί ἐστι τοῦτο τῶν ὄντων, περὶ οὗ οὗτοι οἱ λόγοι εἰσίν, οἷς ἡ ῥητορικὴ χρῆται ( 451D). Rhetoric and Sophistic were regarded by Plato as sisters: cf. 464Bff. The clause ὥσπερ ὁ κιθαριστὴς κτλ. is logically the protasis to εἶεν· ὁ δὲ δὴ σοφιστὴς κτλ. : see on 311E above.

24. εἶεν· ὁ δὲ δὴ σοφιστής . For εἶεν and δὲ δή see on 311C above.

25. δῆλον ὅτι περὶ οὗπερ καὶ ἐπίστασθαι . Stahl's emendation ( ἐπίστασθαι for ἐπίσταται ), which had occurred to us independently, seems to be certain. Most editions read ἢ δῆλον ὅτι περὶ οὗπερ καὶ ἐπίσταται ; inserting ἤ before δῆλον (with Heindorf) and giving the words to Socrates, but it is surely more natural to regard them as giving Hippocrates' reply to Socrates' question, in which case δῆλον ὅτι is right. The MS. reading ἐπίσταται gives a non sequitur ; for the harpist makes one δεινὸς λέγειν περὶ οὖπερ καὶ ἐπιστήμονα i.e. περὶ οὖπερ καὶ ἐπίστασθαι , not περὶ οὗπερ καὶ ἐπιστήμων‐ἐστὶν i.q. ἐπίσταται . The next sentence τί δή ἐστι τοῦτο, περὶ οὗ αὐτός τε ἐπιστήμων ἐστὶν ὁ σοφιστὴς καὶ τὸν μαθητὴν ποιεῖ (sc. ἐπίστασθαι ); in no way invalidates the reading ἐπίστασθαι : it is everywhere assumed in the Platonic writings that he who makes others know has knowledge himself: see for example 111B οὐκ οἶσθ’ ὅτι χρὴ τοὺς μέλλοντας διδάσκειν ὁτιοῦν αὐτοὺς πρῶτον εἰδέναι; ἢ οὔ; πῶς γὰρ οὔ ; and ibid. 113C, 118C.

27. οὐκέτι . See below on οὐκέτι in 321D

2. ἢ εἰ μὲν τὸ σῶμα κτλ. For ἤ cf. 50E ἢ πρὸς μὲν ἄπα σοι τὸν ρατέπα—ρπὸς δὲ τὴν ρατπίδα ἄπα καὶ τοὺς νόμους ἔσται σοι; and for the general form of the sentence 28E εἰ ὅτε μέν—τότε μέν—τοῦ δὲ θεοῦ τάττοντος—ἐνταῦθα δέ, 94C-D, 512A: see the Editor's note on

4. πολλὰ ἂν περιεσκέψω : but ἔδει in l. 3 and παρεκάλεις in 6. The effect is to represent the process of reflection by oneself as prior to consultation with friends. Heindorf on 514Dquotes a parallel from the (144E): ἀτὰρ εἰ, νῷν ἐχόντοιν ἑκατέπου λύπαν, ἔφη αὐτὰς ἡπμόσθαι ὁμοίως, ρότεπον εὐθὺς ἂν ἐπιστεύομεν ἢ ἐπεσκεψάμεθα ἂν (i.e. should have inquired first ) εἰ μουσικὸς ὢν λέγοι ;

8. τὴν ψυχήν . Cf. (with Heindorf) IX. 583E ὃ μεταξὺ ἄπα νῦν δὴ ἀμφοτέπων ἔφαμεν εἶναι, τὴν ἡσυχίαν, τοῦτό ροτε ἀμφότερα ἔσται. ἐν ᾧpresently is like ἐν τούτῳ in 310D where see note: Heindorf cites Eur. 1057 καὶ τἄμ’ ἐν ὑμῖν ἐστιν ἢ καλῶς ἔχειν ἢ μηδὲν εἶναι . For καὶ ἐν ᾧ we should at first sight expect καὶ ἐν αὐτῷ : for the Greek idiom is relative+anaphoric pronoun, not relative+relative, when the two pronouns have the same antecedent and are connected by a conjunction: e.g. 452D τί ἐστι τοῦτο ὃ φὴ|ς σὺ μέγιστον ἀγαθὸν εἶναι—καὶ σὲ δημιουργὸν εἶναι αὐτοῦ: see the Editor's note on 40A. Here, however, the fact that the relative precedes its antecedent ( ὃ δὲ περὶ πλείονος—περὶ δὲ τούτου ) makes the rule inoperative; and there are other exceptions, e.g. II. 374B καὶ τῶν ἄλλων ἑνὶ ἑκάστῳ ὡσαύτως ἓν ἀρεδίδομεν, ρπὸς ὃ ρεφύκει ἕκαστος καὶ ἐφ’ ᾧ ἔμελλε τῶν ἄλλων σχολὴν ἄγων διὰ βίου αὐτὸ ἐπγαζόμενος—καλῶς ἀπεργάζεσθαι, and 192B καὶ ὃ οἶδεν καὶ ὃ αἰσθάνεται(where, however, Bonitz rejects the second ὅ , perhaps rightly).

12. τῷ ἀφικομένῳ τούτῳ ξένῳ . Heindorf's suggestion τῷ ἀφικομένῳ τούτῳ τῷ ξένῳ would convey a somewhat different meaning, viz. ‘this arrival, the foreigner’. The presence of ἀφικομένῳ renders the article after τούτῳ unnecessary: cf. below, 337E τὸ ἀκριβὲς τοῦτο εἶδος . There is some contempt in τούτῳ = isti (see note on 45A), and much scorn in τὴν σὴν ψυχήν , repeated slowly at the end of the clause.

13. ὄρθριος : the MSS. have ὄρθριον by mistake. The adjectival construction of this word is found in XII. 961B δεῖν δὲ ὄρθριον εἶναι τὸν σύλλογον .

18. γιγνώσκεις . The present is regularly used of being acquainted with a person, e.g. 60A Ξανθίππην, γιγνώσκεις γάρ; 144C ἀλλὰ σκόπει εἰ γιγνώσκεις αὐτόν. γιγνώσκω. In Plato the perfect is generally used of knowing things : e.g. 23B ὅστις—ἔγνωκεν ὅτι κτλ., 2C τοσοῦτον πρᾶγμα ἐγνωκέναι.

20. ᾧ μέλλεις σαυτὸν ἐπιτρέπειν is wrongly rejected by Cobet. The words are to be taken closely with τὸν δὲ σοφιστὴν ὅ τί ποτ’ ἔστιν φαίνει ἀγνοῶν : that Hippocrates should entrust himself to that of which he knows nothing is the climax of Socrates' rebuke.

21. ἔοικεν : sc. εἶναι , i.e. be true (not ἐμὲ ἀγνοεῖν , which is doubtful Greek for ἔοικα ἀγνοεῖν ). The subject is simply ‘it’, as in I. 333C, where ἔοικεν is similarly for ἔοικεν εἶναι .

23. τυγχάνει ὢν—ψυχὴ τρέφεται. τυγχάνει ὤνis virtually equivalent to ‘really is’: cf. 468D οἰόμενος ἄμεινον εἶναι αὐτῷ, τυγχάνει δὲ ὂν κάκιον, and note on 4E. The ἔμπορος is a travelling merchant who trades on a larger scale than the retail dealer or κάπηλος : see II. 371D ἢ οὐ καπήλους καλοῦμεν τοὺς ρπὸς ὠνήν τε καὶ ρπᾶσιν διακονοῦντας ἱδπυμένους ἐν ἀγορᾷ, τοὺς δὲ πλανήτας ἐπὶ τὰς πόλεις ἐμπόρους ;The same account of the Sophist as ἔμπορός τις περὶ τὰ τῆς ψυχῆς μαθήματα ( 231E) is given in 223C-224E.

24. φαίνεται γὰρ ἔμοιγε τοιοῦτός τις . We follow Schleiermacher in giving these words to Socrates. Turner judiciously points out that γε in ἔμοιγε is only appropriate if Socrates speaks the words, and that Hippocrates could hardly assent till he knew what τροφὴ ψυχῆς meant.

26. ὅπως γε μὴ—ἐπαινῶν ἃ πωλεῖ ἐξαπατήσει. πωλεῖνis ‘to have on sale’: ‘to sell’ is ἀποδίδοσθαι . Cobet, Novae Lectiones, p. 159. For ἐξαπατήσει the MSS. have ἐξαπατήσῃ , but the 1st aor. conj. is very doubtful in Plato after ὅπως μή (see on the whole question Kühner's II, 899), and final ‐ῃ and ‐ει are frequently confused in the MSS.

28. ὁ ἔμπορός τε καὶ κάπηλος. ἔμπορός τε καὶ κάπηλοςtogether make a plural, and according to strict logic would require a plural article; but ὁ is written by attraction instead of the grammatically impossible οἱ . Cf. 186C ὁ διαγιγνώσκων —τὸν καλόν τε καὶ αἰσχρὸν ἔρωτα; below, 355E τά ὀνόματα—τὸ ἡδύ τε καὶ ἀνιαρόν .

40. τυγχάνεις ἐπιστήμων. τυγχάνειςhas sunk to a mere copula. It is not necessary to insert ὤν after ἐπιστήμων , although it is only in a few cases that τυγχάνω (in this sense) occurs in Plato's MSS. without the participle expressed, viz. 263C μέγιστον τῶν ἀγαθῶν τυγχάνει ; 502B εἰ δέ τι τυγχάνει ἀηδὲς καὶ ὠφέλιμον; II. 369B τυγχάνει ἡμῶν ἕκαστος οὐκ αὐτάρκης ἀλλὰ πολλῶν ἐνδεής; 129A ῥᾴδιον τυγχάνει τὸ γνῶναι ἑαυτόν ; ibid. 133A ἐκεῖνο ᾧ τοῦτο τυγχάνει ὅμοιον . In these cases (as here in the ) it is easy to suppose that the participle has fallen out from ‘lipography’: but in (perhaps pseudo-Platonic), 300A and in XI. 918C, and 61C, the participle cannot be so easily supplied: and that the construction without the participle was used in everyday speech appears from Ar. 1141 καὶ τῶν θεατῶν εἴ τις εὔνους τυγχάνει. See Rutherford's p. 342.

43. περὶ τοῖς φιλτάτοις. τὰ φίλταταis used here of the soul's health as in 513A σὺν τοῖς φιλτάτοις ἡ αἵρεσις ἡμῖν ἔσται ταύτης τῆς δυνάμεως τῆς ἐν τῇ πόλει.

44. κυβεύῃς τε καὶ κινδυνεύῃς . Cobet rejects τε καὶ κινδυνεύῃς , but Plato often puts metaphor and interpretation side by side. Cf. below 314B ἐν αὐτῇ τῇ ψυχῇ λαβόντα καὶ μαθόντα (where Deuschle wrongly rejected καὶ μαθόντα ); 334D σύντεμνέ μοι τὰς ἀποκρίσεις καὶ βραχυτέρας ποίει ; 297C πολὺ γάρ πού εἰμι φαυλότερος τοῦ Ἡρακλέους, ὃς οὐχ οἶός τε ἦν τῇ τε ὕδπᾳ διαμάχεσθαι, σοφιστπίᾳ οὔσῃ καὶ διὰ τὴν σοφίαν ἀνιείσῃ, εἰ μίαν κεφαλὴν ἀροτμηθείη τοῦ λόγου, ρολλὰς ἀντὶ τῆς μιᾶς, καὶ καπκίνῳ τινὶ ἑτέπῳ σοφιστῇ . . . ὃς ἐρειδὴ αὐτὸν ἐλύρει οὕτως ἐκ τοῦ ἐπ’ ἀριστερὰ λέγων καὶ δάκνων κτλ.; 194C χειμαζομένοις ὲν λόγῳ καὶ ἀποροῦσιν; 174C εἰς φρέατά τε καὶ πᾶσαν ἀπορίαν ἐμπίπτων. See also note on τοὺς τῶν νέων τὰς βλάστας διαφθείροντας in 3A.

47. παρὰ τοῦ καπήλου καὶ ἐμπόρου : so B. The κάπηλος is put in the foreground as the most usual seller of σιτία and ποτά : the article is expressed only once, because the κάπηλος and ἔμπορος both belong to the same genus ‘merchants’: cf. Hdt. IV. 71 τὸν οἰνοχόον καὶ μάγειρον καὶ ἱπποκόμον καὶ διήκονον καὶ ἀγγελιηφόρον κτλ. There is no sufficient ground for bracketing the words παρὰ τοῦ καπήλου καὶ ἐμπόρου as is done by Schanz, nor for reading παρά του (with T) and omitting καπήλου καὶ ἐμπόρου , as Hermann did.

ἐν ἄλλοις ἀγγείοις : i.e. other than our own bodies. Cf. 311E τί ὄνομα ἄλλο γε λεγόμενον περὶ Πρωταγόρου ἀκούομεν ; It need not be implied that the body is itself an ἀγγεῖον (viz. of soul), though the notion is not unplatonic, and kindred expressions are found in later philosophy, e.g. Marcus Aurel. X. 38 τὸ περικείμενον ἀγγειῶδες καὶ τὰ ὀργάνια ταῦτα τὰ περιπεπλασμένα (said of the body).

54. ἐν αὐτῇ τῇ ψυχῇ λαβόντα καὶ μαθόντα. λαμβάνειν ἐν(not εἰς ) as in VII. 517A εἴ πως ἐν ταῖς χερσὶ δύναιντο λαβεῖν; 243C ταὐτὸν τοῦτο πάθος εἰληφότες ἐν τῇ ψυχῇ. For καὶ μαθόντα , which is explanatory of ἐν αὐτῇ τῇ ψυχῇ λαβόντα , see note on κυβεύῃς τε καὶ κινδυνεύῃς .

57. νέοι ὥστε τοσοῦτον . Heindorf quotes Eur. 80 γέρων ἐκεῖνος ὥστε σ’ ὠφελεῖν παρών , and points out that whereas νεώτεροι ἢ ὥστε would deny altogether ἡ τοῦ διελέσθαι δύναμις , the words νέοι ὥστε are less strong; nobis nonnisi iuvenilis quaedam facultas suppetit ad tantam rem diiudicandam . The best MSS. of Plato read ταὐτόν, τοιοῦτον, τοσοῦτον , etc., in the great majority of cases rather than ταὐτό , etc. Schanz (Preface to p. vi) thinks it probable that Plato always used the forms in ‐ν . In inscriptions of Plato's time τὸ αὐτό and τὸ αὐτόν occur side by side, but apparently only τοιοῦτον, τοσοῦτον . See Meisterhans, 2 , p. 122.

61. Ἱππίας . Hippias of Elis was one of the most accomplished and—if we may trust the Platonic writings—ostentatious of the Sophists. According to the (285B ff.) he claimed to be at home in all the learning of the day—in Astronomy, Geometry, Arithmetic, Philology, Music, Mythology, History and Archaeology. See Zeller's I 4 , 956 ff.

62. οἶμαι δὲ καὶ Πρόδικον τὸν Κεῖον . Contrast 402B οἶμαι δὲ καὶ Ἡσίοδος. Either construction is admissible. Prodicus of Ceos is repeatedly mentioned in the Platonic writings. A fellow-citizen of the poet Simonides (below, 339E , he professed like Gorgias and Hippias to educate young men ( 19E, 127E, X. 600C) and received very large sums in return for his instruction together with the gratitude of his pupils. On one occasion, when in charge of a political mission from Ceos, he is said to have won great reputation in the βουλή at Athens for his conduct of public business, and to have given at the same time private lectures, which were popular and well paid ( 282C). He laid great stress on the importance of using words in their correct sense ( ὀρθότης ὀνομάτων ): see below, 337A 358A and 277E, 163D, 197D; cf. also 267B; but this was only taught (we are told) in his 50-drachma lecture; the impecunious Socrates had only paid one drachma and was not quite master of this subject ( 384B). Socrates is fond of professing himself a pupil of Prodicus, e.g. below, 341A 96D, 163D. Prodicus wrote eulogies of Heracles and others ( 177B): the substance, if not the actual words, of his Apologue of Heracles at the cross-roads is given by Xenophon, II. 1. 21. A scholiast on X. 600Csays the Athenians put him to death by hemlock for corrupting the youth, but there is no other authority for this unlikely story. Cf. Zeller, I 4 , 952 ff.

7. κινδυνεύει—ἄχθεσθαι . Present, because the dialogue is narrated just after it took place (see 309B , and as Protagoras is still with Callias, the stream of visitors probably continues. It is not quite clear why the number of the sophists should have made the porter annoyed with the visitors, unless the visitors themselves were sophists. Probably therefore τοῖς φοιτῶσιν means τοῖς φοιτῶσιν σοφισταῖς : in the porter's judgement the house was already too full of sophists. φοιτᾶν implies frequent visitation: cf. 181C χρῆν μὲν οὖν καὶ πρότερόν σε φοιτᾶν αὐτὸν παρ’ ἡμᾶς καὶ οἰκείους ἡγεῖσθαι.

9. ἐκρούσαμεν τὴν θύραν . There was a metal knocker ( ῥόπτρον, κόραξ, ἐπίσπαστρον ) on the αὔλειος θύρα . Blümner, Gr. Privatalt. p. 149.

10. οὐ σχολὴ αὐτῷ . For αὐτῷ ‘the master’ cf. below, 315B I. 327B ἠρόμην ὅπου αὐτὸς εἴη, and the Pythagorean αὐτὸς ἔφα ipse dixit .

11. ἀμφοῖν τοῖν χεροῖν . So B: T has ταῖν for τοῖν . Cobet has shown that the dual feminine of the article is the same as the dual masculine: p. 70.

13. ἐκρούομεν , but ἐκρούσαμεν in l. 9. Socrates pretends to be somewhat staggered by the rebuff.

ἐγκεκλῃμένης . So Bekker for ἐγκεκλειμένης of B: the Attic form is κλῄω not κλείω : Cobet, p. 159.

18. ἅνθρωπος . According to Ast, the Greeks sometimes used ἀνήρ and ἄνθρωπος without the article of a definite person sed ita ut vel contemptionem vel reprehensionem aliquam simul indicarent . But in the cases quoted in support of this usage where the word is in an oblique case, the meaning is quite general like the English ‘a man’, e.g. X. 596C δεινόν τινα λέγεις καὶ θαυμαστὸν ἄνδρα; 98B ὁρῶ ἄνδρα τῷ μὲν νῷ οὐδὲν χρώμενον; Soph. 1228 ἀπάταισιν αἰσχραῖς ἄνδρα καὶ δόλοις ἑλών ; 155D ἐάν σοι ἀνδρός, μᾶλλον δὲ ἀνδρῶν ὀνομαστῶντῆς διανοίας τὴν ἀλήθειαν—συνεξερευνήσωμαι: it is therefore safer to suppose (with Schanz and most Platonic scholars) that where ἀνήρ and ἄνθρωπος of the MSS. can only be translated by ‘the man’, the rough breathing should be restored: e.g. below, 315E and 267A, 268C; I. 331E.

20. προστῴῳ . The αὐλή was surrounded by cloisters. This προστῷον is doubtless that on which the passage from the πρόθυρον opened.

23. Πάραλος—Ἀντίμοιρος . The mother of Xanthippus and Paralus had once been the wife of Hipponicus, to whom she bare Callias: Plut. XXIV. 9. In 320A as well as in 94BPlato remarks that Pericles was unable to teach the art of statesmanship to either of his two sons: cf. 118E τὼ— Περικλέους υἱέε ἠλιθίω ἐγενέσθην . They both died of the plague. One of the most interesting fragments of Protagoras describes the fortitude of Pericles when his sons died: see Appendix 11, p. 221, Frag. 3. Charmides, son of the elder Glaucon, was Plato's maternal uncle: the dialogue is named after him. He was φιλόσοφός τε καὶ πάνυ ποιητικός ( 155A), and as remarkable for σωφροσύνη as for personal beauty ( ibid. 157D). He was afterwards one of the Ten, and fell along with Critias at the battle of Munychia in 404 B.C. (Xen. II. 4. 19). Of Philippides and his father Philomelus and Antimoerus of Mende (on the west coast of the peninsula Pallene) nothing further is known.

28. τούτων—ἐφαίνοντο . B and T omit οἵ , which was inserted by Stephanus: Heindorf prefers to insert ὧν after λεγομένων . The MSS. reading could only be defended if ἐφαίνοντο could be regarded as parenthetical = ὡς ἐφαίνοντο , like δοκεῖ μοι = ὡς δοκεῖ μοι (see on 314C above), but there seems to be no authority for such a use of φαίνομαι . It is hardly to be supposed that in ἐπακούοντες and ἐπήκοοι (in B below) there is an allusion to an outer circle of Protagorean students, deemed unworthy of the subtlest teaching of the master, in spite of 152C, where Socrates is hardly serious.

33. χορῷ . The tragic chorus consisted of 15 χορευταί , arranged in 3 rows ( στοῖχοι ) of 5 each, the leader ( κορυφαῖος ) being in the middle of the first. So here Protagoras occupies the middle of a row of 7. With the metaphor cf. 173B τοὺς— τοῦ ἡμετέρου χοροῦand ibid. C λέγωμεν δὴ—περὶ τῶν κορυφαίων ; 246Eff.

34. ηὐλαβοῦντο . The MSS. have εὐλαβοῦντο , but in the classic period εὐ‐ (and εἰ‐ ) were always augmented in the past tenses, while after 300 B.C. ηὐ‐ does not appear on inscriptions. See Meisterhans, 2 , p. 136.

3. ἐν θρόνῳ—ἐπὶ βάθρων . Hippias is depicted as a schoolmaster among his pupils.

Ἐρυξίμαχός τε—Ἄνδρων ὁ Ἀνδροτίωνος . Eryximachus and his father Acumenus were doctors. The former delivers one of the speeches in the defining his art as ἐπιστήμη τῶν τοῦ σώματος ἐρωτικῶν πρὸς πλησμονὴν καὶ κένωσιν ( 186C). Phaedrus, son of Pythocles, of the deme of Μυρρινοῦς in the tribe Πανδιονίς , also appears in the and gives his name to the dialogue Andron, son of Androtion (so T here, and both B and T in 487C—the Ἀριστίωνος of B is therefore a mistake), is in the described as one of those who shared the view of Callicles, that too much φιλοσοφία ruined men. His son is believed to have been the orator Androtion.

6. περὶ φύσεως . The primary substance which early Greek philosophy sought to discover was called by them φύσις : see Aristotle, IV. 4, p. 1014 b . 32 ff. τοῦτον γὰρ τὸν τρόπον καὶ τῶν φύσει ὄντων τὰ στοιχεῖά φασιν εἶναι φύσιν, οἱ μὲν ρῦπ, οἱ δὲ γῆν, οἱ δ’ ἀέπα, οἱ δ’ ὕδωπ, οἱ δ’ ἄλλο τι τοιοῦτον λέγοντες, οἱ δ’ ἔνια τούτων, οἱ δὲ πάντα ταῦτα . Thus (among others) Heraclitus, Anaxagoras, and Empedocles called their books περὶ φύσεως , i.e. on the substance which gives birth ( φύει ) to all else. By Plato's time the word had acquired a wider meaning equivalent to ‘de rerum natura’, our ‘Nature’, and it is in this sense that the word is here used.

7. ἀστρονομικὰ ἄττα . Schanz brackets ἀστρονομικά , but the word is meant to be taken rather with μετεώρων than with φύσεως . On Hippias' skill in astronomy cf. 367E ἦς (sc. ἀστρονομικῆς ) αὖ σὺ τέχνης ἔτι μᾶλλον ἐπιστήμων οἴει εἶναι ἢ τῶν ἔμπροσθεν .

10. Τάνταλόν γε . XI. 582 καὶ μὴν Τάνταλον εἰσεῖδον κρατέρ’ ἄλγε’ ἔχοντα. Prodicus is compared to Tantalus because of his physical wretchedness: see 395E καὶ ἀτεχνῶς ἔοικεν ὥσρεπ ἂν εἴ τις βουλόμενος ταλάντατον ὀνομάσαι, ἀροκπυρτόμενος ὀνομάσειε καὶ εἴροι ἀντ’ ἐκείνου Τάνταλον· τοιοῦτόν τι καὶ τούτῳ τὸ ὄνομα ἔοικεν ἐκπορίσαι ἡ τύχη τῆς φήμης.

ἐπιδημεῖ γὰρ ἄρα . Heindorf's correction ἐπιδημεῖ for ἐπεδήμει seems to be necessary, as Prodicus is still in Athens. Just so the presential tenses are used in 314D κινδυνεύει δὲ— ἄχθεσθαι and below, l. 14 κατάλυσιν πεποίηκεν . If the clause is taken as a parenthetical clause of explanation, there is nothing harsh in the construction: ἦν δὲ ἐν οἰκήματί τινι naturally enough follows on εἰσεῖδον . For γὰρ ἄρα (a somewhat rare combination) Heindorf quotes IV. 438A πάντες γὰρ ἄρα τῶν ἀγαθῶν ἐπιθυμοῦσιν, 205B ἀφελόντες γὰρ ἄρα(so T: B has only γάρ ) τοῦ ἔρωτός τι εἶδος ὀνομάζομεν , III. 698D συνάψαντες γὰρ ἄρα τὰς χεῖρας σαγηνεύσαιεν πᾶσαν τὴν Ἐρετρικήν. γὰρ ἄραalways introduces something supposed to be known to the person addressed: see 314B οἶμαι δὲ καὶ Πρόδικον τὸν Κεῖον .

20. τὴν δ’ οὖν ἰδέαν πάνυ καλός . Cf. 174A ταῦτα δὴ ἐκαλλωπισάμην ἵνα καλὸς παρὰ καλὸν(i.e. Ἀγάθωνα ) ἴω . There is no ground for reading καλόν for καλός with Stephanus: cf. 180E τὰ γὰρ μειράκια τάδε—διαλενόμενοι.

21. Ἀγάθωνα . It is at the poet Agathon's house that the scene of the is laid. From his speech in that dialogue 194D ff. we should infer that he had studied under Prodicus to some purpose. Born about 447 B.C., Agathon was first successful on the tragic stage at an early age: cf. 175E ἡ δὲ σὴ (sc. Ἀγάθωνος σοφία ) λαμπρά τε καὶ πολλὴν ἐπίδοσιν ἔχουσα, ἥ γε παρὰ σοῦ νέου ὄντος οὕτω σφόδπα ἐξέλαμψε καὶ ἐκφανὴς ἐγένετο ρπώην ἐν μάρτυσι τῶν Ἑλλήνων πλέον ἢ τρισμυρίοις . He visited the court of Archelaus about 407 B.C. and died probably about 400. That he was held in esteem by the Athenians appears from Ar. 83 ff. Ἀγάθων δὲ ποῦ ’σθ’; ὅπου ’στ’; ἀπολιπών μ’ οἴχεται ἀγαθὸς ποιητὴς καὶ ποθεινὸς τοῖς φίλοις . For παιδικὰ Παυσανίου cf. 193B.

22. τοῦτ’ ἦν τὸ μειράκιον . So BT: the idiom is like our ‘There was this young man’. The nearest parallels seem to be 59B ἦν δὲ καὶ Κτήσιππος ὁ Παιανιεύςand X. 615D ἦσαν δὲ καὶ ἰδιῶταί τινες. τοῦτό τε δή(Ast), τοῦτό τ’ οὖν (Heindorf), τοῦτ’ ἔνδον ἦν (Schirlitz), have been suggested, while Kroschel rejects the words, but the text is probably sound.

23. τὼ Ἀδειμάντω . Adimantus son of Cepis is not elsewhere named. The son of Leucolophidas was afterwards στρατηγός under Alcibiades, whose policy he supported (Xen. I. 4. 21): he was taken prisoner at the battle of Aegospotami, but spared by the Spartans. He was afterwards put on his trial by Conon on the charge, not improbably a true one, of having betrayed the Athenian fleet at this battle: see Kock on Ar. 1512.

26. πάσσοφος—ἁνήρ—καὶ θεῖος . For ἁνήρ BT read ἀνήρ , but see on 314E above. πάσσοφος is often ironical: e.g. 271C ὃ δὲ σὺ ἐρωτᾷς τὴν σοφίαν αὐτοῖν, θαυμασία, ὦ Κρίτων· πάσσοφοι ἀτεχνῶς τώ γε, οὐδ’ ᾔδη πρὸ τοῦ ὅ τι εἶεν οἱ παγκρατιασταί . For the meaning of θεῖος cf. 99C οὐκοῦν—ἄξιον τούτους θείους καλεῖν τοὺς ἄνδρας, οἵτινες νοῦν μὴ ἔχοντες πολλὰ καὶ μεγάλα κατορθοῦσιν ὧν πράττουσιν καὶ λέγουσι; θεῖος(in the form σεῖος ) was the favourite Spartan word of praise and was much affected by the Socratic circle in Athens.

3. Κριτίας the foremost of the Thirty so-called tyrants was related to Plato on the mother's side, Perictione Plato's mother being granddaughter of the elder Critias, who was father of Callaeschrus. He is one of the dramatis personae in the and and the dialogue is named after him. A scholiast on the 20A (quoted by Stallbaum) describes him as γενναίας καὶ ἁδρᾶς φύσεως, ἥπτετο δὲ καὶ φιλοσόφων συνουσιῶν, καὶ ἐκαλεῖτο ἰδιώτης μὲν ἐν φιλοσόφοις, φιλόσοφος δὲ ἐν ἰδιώταις . For the fragments of his poems see Bergk's 4 , II, 279 ff.

4. ἔτι σμίκρ’ ἄττα—διαθεασάμενοι. σμίκρ’ ἄτταis the plural of σμικρόν τι , and as ἔτι σμικρόν τι διατρίψαντες would mean ‘after a little further delay’, so ἔτι σμίκρ’ ἄττα διατρίψαντες means ‘after some little further delays’. The reference in ἔτι is to 314C ταῦτα in καὶ ταῦτα refers to the causes of the delays, which were, as usual, some topics of discourse: its antecedent is involved in διατρίψαντες .

6. πρὸς σέ τοι ἤλθομεν . Here, as in 314E both B and T have τι for τοι : but τι is impossible here, and this use of τοι is very frequent in address: Heindorf refers ( inter alia ) to Ar. 1100 σέ τοι σέ τοι λέγω, Καρίων, ἀνάμεινον .

15. οἴεταί οἱ μάλιστ’ ἂν γενέσθαι . B and T read μάλιστα without ἄν : see on 51E. It is more likely that ἄν should have dropped out after μάλιστα here than that Plato's usage should have varied with οἴομαι and the aorist infinitive referring to the future: cf. below, 353B ᾗ οἶμαι ἂν ἔγωγε κάλλιστα φανερὸν γενέσθαι , where ἄν is in all the MSS.

17. μόνος οἴει δεῖν—πρὸς μόνους. οἴει δεῖνis virtually like one verb, ἀξιοῖς or the like: it therefore takes the nom. (not the acc.) with inf.; cf. Demosth. 235 εὐθὺς ἡγούμην ἐν τούτοις—αὐτὸς περιεῖναι δεῖν αὐτῶν καὶ μεγαλοψυχότερος φαίνεσθαι , quoted by Stallbaum on 50E. Note that Protagoras here uses the more precise μόνος πρὸς μόνους , but above only μόνῳ (316 B)—unless (with Cobet and Schanz) we there insert μόνοι before μόνῳ .

19. ξένον γὰρ ἄνδρα καὶ ἰόντα κτλ. For καὶ ἰόντα T reads in the margin κατιόντα : Liebhold suggests καταλύοντα. ἰόντα , however, is parallel to ξένον and to πείθοντα , ‘a man who is a stranger and goes—and tries to persuade etc.’ The whole passage recalls 19E-20A τούτων (viz. Gorgias, Prodicus, Hippias) γὰρ ἕκαστος—οἷός τ’ ἐστὶν ἰὼν εἰς ἑκάστην τῶν πόλεων τοὺς νέους οἷς ἔξεστι τῶν ἑαυτῶν ρολιτῶν ρποῖκα ξυνεῖναι ᾧ ἂν βούλωνται—τούτους ρείθουσι τὰς ἐκείνων ξυνουσίας ἀρολιρόντας σφίσιν ξυνεῖναι χρήματα διδόντας καὶ χάριν προσειδέναι . Notice throughout the longwinded character of Protagoras' speeches: cf. Philostrat. I. 494 (quoted by Kroschel on 320C , γνοὺς δὲ τὸν Πρωταγόραν ὁ Πλάτων σεμνῶς μὲν ἑρμηνεύοντα, ὑπτιάζοντα δὲ τῇ σεμνότητι καί που καὶ μακρολογώτερον τοῦ συμμέτρου, τὴν ἰδέαν αὐτοῦ μύθῳ μακρῷ ἐχαρακτήρισεν .

29. τὸ ἐπαχθὲς αὐτῆς. ἐπαχθέςlike φορτικόν means what hearers will resent (think a burden or ἄχθος ) as in bad taste because presumptuous: cf. Demosth. de Pace, 4 τὸ λέγειν περὶ ὧν αὐτὸς εἶρέ τις καὶ ρεπὶ αὑτοῦ—οὕτως ἡγοῦμαι φοπτικὸν καὶ ἐπαχθὲς ὥστε κτλ.

30. τοὺς μὲν ποίησιν. σοφόςand its derivatives are often used especially in early writers in connection with poetry, e.g Theognis, 19-20 Κύρνε σοφιζομένῳ μὲν ἐμοὶ σφρηγὶς ἐπικείσθω τοῖσδ’ ἔπεσιν , Pindar, 4. 29-30 μελέταν δὲ σοφισταῖς Διὸς ἕκατι πρόσβαλον σεβιζόμενοι .

32. τοὺς ἀμφί τε Ὀρφέα καὶ Μουσαῖον = ‘Orpheus and Musaeus and their schools’. The Editors quote Ar. 1032 ff. Ὀρφεὺς μὲν γὰρ τελετάς θ’ ἡμῖν κατέδειξε φόνων τ’ ἀρέχεσθαι, Μουσαῖος δ’ ἐξακέσεις τε νόσων καὶ χπησμούς, Ἡσίοδος δὲ γῆς ἐργασίας, καρπῶν ὥρας, ἀρότους· ὁ δὲ θεῖος Ὅμηρος ἀπὸ τοῦ τιμὴν καὶ κλέος ἔσχεν πλὴν τοῦδ’ ὅτι χρήστ’ ἐδίδαξε, τάξεις, ἀρετάς, ὁπλίσεις ἀνδρῶν ;For the misplacement of τε cf. 43B ὲν τοσαύτῃ τε ἀγρυπνίᾳ καὶ λύπῃ εἶναι.

34. Ἴκκος—Ἡρόδικος . Iccus is mentioned in VIII. 839E ff. as a model of temperance. According to Pausanias, he won in the pentathlon at the Olympic Games and afterwards set up as a trainer ( γυμναστής ). Herodicus of Selymbria in Thrace combined the professions of the γυμναστής and ἰατρός ( μίξας γυμναστικὴν ἰατρικῇ ). He is described in III. 406Aas the inventor of the παιδαγωγικὴ τῶν νοσημάτων ἰατρική —by means of which he tortured first himself (by prolonging his own dying) and afterwards others: δυσθανατῶν—ὑπὸ σοφίας εἰς γῆρας ἀφίκετο ( ibid. B).

ὁ νῦν ἔτι ὤν—σοφιστής . Heindorf reads ἧττον in place of ἥττων , but νῦν ἔτι ὤν is ‘still living’ and οὐδενὸς ἥττων σοφιστής is a descriptive qualification.

37. Ἀγαθοκλῆς—Πυθοκλείδης . Agathocles is mentioned in 180Das teacher of Damon: according to the 118C Pythoclides also taught Pericles. A scholiast on the ( loc. cit. ) remarks: Πυθοκλείδης μουσικὸς ἦν τῆς σεμνῆς μουσικῆς διδάσκαλος, καὶ Πυθαγόρειος, οὗ μαθητὴς Ἀγαθοκλῆς, οὗ Λαμπροκλῆς ( Λάμπρος , van Prinsterer), οὗ Δάμων . With the sentiment compare the imitation in Plut. 4. 2 ὁ δὲ Δάμων ἔοικεν ἄκπος ὢν σοφιστὴς καταδύεσθαι μὲν εἰς τὸ τῆς μουσικῆς ὄνομα ρπὸς τοὺς πολλοὺς ἐπικρυπτόμενος τὴν δεινότητα .

40. ταῖς τέχναις ταύταις παραπετάσμασιν ἐχρήσαντο . Cf. Plut. οὐ μὴν ἔλαθεν ὁ Δάμων τῇ λύρᾳ παρακαλύμματι χρώμενος .

41. κατὰ τοῦτο εἶναι. εἶναιis redundant as in Isaeus, περὶ τοῦ Μενεκλέους κλήρου , 32 ἡμεῖς ὠμόσαμεν εὖ ποιεῖν ἀλλήλους— κατὰ δύναμιν εἶναι, καὶ λόγῳ καὶ ἔργῳ , and in the common ἑκὼν εἶναι : but no other instance of the precise phrase κατὰ τοῦτο εἶναι is quoted.

45. οἵ γε πολλοὶ—οὐδὲν αἰσθάνονται . Cf. 474A τοὺς δὲ πολλοὺς ἐῶ χαίρειν, καὶ ἕνα ἐπιψηφίζειν ἐπίσταμαι, τοῖς δὲ πολλοῖς οὐδὲ διαλέγομαι. For ὡς ἔπος εἰπεῖν (to be taken with οὐδὲν ) see on 17A. The phrase is used to qualify the comprehension of a word or phrase, generally οὐδείς or πάντες .

52. τούτων τὴν ἐναντίαν ἅπασαν ὁδόν . ‘h. e. τὴν ὁδὸν ἣ ἅπασά ἐστιν ἐναντία τούτων . pro τούτων eadem comparatione quae vocatur compendiaria dici potuit τούτοις . Andocid. or. I. 4 ἐγὼ πολὺ τὴν ἐναντίαν τούτοις γνώμην ἔχω ’ (Kroschel). For ἅπασαν with an adjective compare (with Kroschel) Xen. I. 5. I ἧν ἡ γῆ πεδίον ἅπαν ὁμαλὲς ὥσπερ θάλαττα and the common phrase πᾶν τοὐναντίον .

54. σοφιστὴς εἶναι καὶ παιδεύειν ἀνθρώπους . The σοφιστὴς σοφίζει , i.e. παιδεύει : see note on 312C above. 167C (where Socrates is speaking for Protagoras) ὁ σοφιστὴς τοὺς ραιδευομένους οὕτω δυνάμενος ραιδαγωγεῖν σοφός τε καὶ ἄξιος πολλῶν χρημάτων τοῖς παιδευθεῖσιν .

57. σὺν θεῷ εἰπεῖν . ‘Crebra modestiae formula, bene explicat Steph., quod ita dico, ut in eo opem divinam agnoscam ’. Heindorf on 151B καὶ σὺν θεῷ εἰπεῖν, πάνυ ἱκανῶς τοπάζω οἶς ἂν συγγενόμενοι ὄναιντο.

59. πολλά γε ἔτη ἤδη εἰμὶ ἐν τῇ τέχνῃ . 91E οἶμαι γὰρ αὐτὸν(i.e. Πρωταγόραν ) ἀποθανεῖν ἐγγὺς καὶ ἑβδομήκοντα ἔτη γεγονότα, τετταράκοντα δὲ ἐν τῇ τέχνῃ ὄντα . The phrases εἶναι ἐν τῇ τέχνῃ, ἐν φιλοσοφίᾳ and the like are common.

60. οὐδενὸς ὅτου οὐ . For the asyndeton Heindorf compares Hom. XIV. 96 ἧ γάρ οἱ ζωή γ’ ἦν ἄσπετος· οὔ τινι τόσση ἀνδρῶν ἡρώων. The nearest parallel to πάντων (which Cobet rejects) after οὐδενὸς ὅτου οὐ seems to be in Dem. adv. Lept. § 3 οὐ γὰρ ἔστιν ἐφ’ ὅτου τοῦτο οὐ πεπόνθατε τῶν πάντων (quoted by Heindorf).

62. ἁπάντων goes with τῶν ἔνδον ὄντων . It seems better to take περὶ τούτων with βούλεσθε than with τὸν λόγον ποιεῖσθαι : otherwise it is hardly possible to separate τούτων from ἁπάντων .

65. ἐνδείξασθαι καὶ καλλωπίσασθαι—ὅτι ἐρασταί. ὅτιdepends rather on ἐνδείξασθαι than on καλλωπίσασθαι : cf. below, 327A ἐδίδασκε καὶ ἐπέπληττε τὸν μὴ καλῶς αὐλοῦντα , and ibid. B τῶν δικαίων καὶ τῶν νομίμων οὐδεὶς φθονεῖ οὐδ’ ἀποκρύπτεται ὥσπερ τῶν ἅλλων τεχνημάτων , 94D τὰ μὲν ἀπειλοῦσα, τὰ δὲ νουθετοῦσα ταῖς ἐπιθυμίαις. See also notes on 327A and 335E below.

66. εἶμεν . BT read εἴημεν , but in Plato's time the short form of the optative plural is regular: see on 22E.

τί οὖν—οὐ—ἐκαλέσαμεν . See above on 310A and for ἐπακούσωσιν note on 315A

70. καθεζόμενοι . So T: B has καθιζόμενοι , but presently in 317E both MSS. read συνεκαθεζόμεθα .

72. καὶ αὐτοὶ ἀντιλαβόμενοι . This passage is difficult and different views are possible as to the construction and meaning. The MSS. read καὶ αὐτοί τε . If the MSS. reading is retained, τε must be regarded as answering to δέ in ἐν δὲ τούτῳ (cf. 3E), and καὶ αὐτοί = etiam ipsi, i.e. even without waiting for servants—a harsh explanation both for καί and αὐτοί . If τε is left out (with most of the editors, following Heindorf), καὶ αὐτοί means ‘ourselves also’ as opposed to Callias and Alcibiades in ἐν δὲ τούτῳ Καλλίας τε καὶ Ἀλκιβιάδης . There is still a slight difficulty in contrasting αὐτοί with these two, because they also would seem to be included in αὐτοί ( πάντες ἡμεῖς—καὶ αὐτοί ), but such slight inaccuracies of expression are common in actual dialogue, which Plato's style is intended to imitate.

75. Καλλίας . Hermann read Κριτίας in order to escape the slight difficulty remarked on in the preceding note: ‘sed hospitis erat Prodicum ut accederet invitare’ (Kroschel).

76. ἀναστήσαντες : but ἄγοντε just before. 273Boffers an exact parallel (if we follow the reading of the Bodleian MS.): ἰδόντες δὲ αὐτὸν ὅ τε Διονυσόδωρος καὶ ὁ Εὐθύδημος πρῶτον μὲν ἐριστάντε διελεγέσθην ἀλλήλοιν, ἄλλην καὶ ἄλλην ἀροβλέροντες εἰς ἡμᾶς : cf. ibid. 273A εἰσελθόντες—περιεπατείτην and D ἐγελασάτην γοῦν ἄμφω βλέψαντες εἰς ἀλλήλους . The dual in Plato as in Greek generally is apt to sink into the plural.

6. γάρ used like Latin enim to introduce a speech (introductory γάρ ): see on 20E.

7. ὅ τι—ἀποβήσεται—πυθέσθαι . For the situation cf. 455C-D, where Socrates says ἴσως γὰρ καὶ τυγχάνει τις τῶν ἔνδον ὄντων μαθητής σου βουλόμενος γενέσθαι, ὡς ἐγώ τινας σχεδὸν καὶ συχνοὺς αἰσθάνομαι, οἳ ἴσως αἰσχύνοιντ’ ἄν σε ἀνεπέσθαι· ὑρ’ ἐμοῦ οὖν ἀνεπωτώμενος νόμισον καὶ ὑρ’ ἐκείνων ἀνεπωτᾶσθαι· τί ἡμῖν, ὦ Γοπγία, ἔσται, ἐάν σοι συνῶμεν; ρεπὶ τίνων τῇ ρόλει συμβουλεύειν οἶοί τε ἐσόμεθα ;

12. ἐν τῇ ὑστεραίᾳ ταὐτὰ ταῦτα : sc. ἔσται σοι. ἐν (bracketed by Schanz after Hirschig) before τῇ ὑστεραίᾳ means ‘during’: cf. (with Kroschel) Thuc. I. 44. 1 τῇ μὲν προτέρᾳ οὐχ ἧσσον—ἀπεδέξαντο τοὺς λόγους, ἐν δὲ τῇ ὑστεραίᾳ μετέγνωσαν κτλ. and add below 321C ἡ εἱμαρμένη ἡμέρα παρῆν, ἐν ᾗ ἔδει καὶ ἄνθρωπον ἐξιέναι ἐκ γῆς .

17. ὃ μὴ τυγχάνεις ἐπιστάμενος. τυγχάνοις(suggested by Bekker) would be the usual periodic structure: but the indicative is looser and more direct: Kroschel compares 464D εἰ δέοι—διαγωνίζεσθαι—πότερος ἐπαίει. Notice the usual Socratic implication that virtue is knowledge.

18. ἀλλὰ μὴ οὕτως . Cf. 74D ἀλλὰ μή μοι οὕτωςand below 331C So μὴ γάρ, μὴ σύ γε, μή τοί γε and the like are often used without a verb expressed. ὥσπερ ἄν belongs to εἴποι ἄν in ll. 26 and 30: it is answered by οὕτω δή in l. 31. αὐτίκα here does not mean ‘for example’ but goes with μάλα (which intensifies it): ‘as if Hippocrates were to change his desire on the spot and conceive a desire ( ἐπιθυμήσειεν ) for etc.’ αὐτίκα μάλα is idiomatic, cf. 469D αὐτίκα μάλα δεῖν τεθνάναι, 384B αὐτίκα μάλα εἰδέναι.

20. νῦν νεωστί : nunc nuper as in 90A. The more general word is followed by the more specific exactly as in 161C βατράχου γυρίνου, where γυρῖνος according to a scholiast is τὸ ἐκ τοῦ βατράχου παιδίον .

21. Ζευξίππου . An obscure sculptor (not painter) of this name is mentioned in No. 1229 (quoted by Overbeck, p. 300). The Zeuxippus here mentioned is supposed to be the same as the famous painter Zeuxis of Heraclea (the Greek colony in Italy, most probably) who is mentioned by Plato in 453C, and several times by Xenophon ( I. 4. 3, 10. 1, 4. 63 τὸν Ἡρακλεώτην ξένον ): see Overbeck, pp. 311 ff. Zeuxis began to rise into fame probably about 424 B.C.: he would therefore be still a young man about the date when this dialogue is supposed to take place: see Introduction, p. xxxiii. Brunn would read Ζεύξιδος for Ζευξίππου here, but according to Fick (quoted by Sauppe) Ζεῦξις is nothing but a familiar diminutive of Ζεύξιππος : for other similar examples see Fick, pp. xxviii ff. ‐ιππος in proper names was aristocratic ( Ar. 62 ff.) and could upon occasion be omitted or added without prejudice to personal identity: cf. 929 οὐχὶ διδάξεις τοῦτον Κρόνος ὤνwith ibid. 1071 σὺ δ’ εἶ κρόνιππος , where ‐ιππος adds mock dignity to κρόνος .

27. Ὀρθαγόρᾳ τῷ Θηβαίῳ . Orthagoras is said by Athenaeus (IV. 184E) on the authority of Aristoxenus to have taught Epaminondas how to play the flute.

29. αὐτόν but ἐκείνου in 28 and ἐκείνω? in 30: see on 310D εἰς ὅ τι βελτίων is here used rather than ὅ τι or τί βελτίων (as in l. 25) to prepare for εἰς αὔλησιν , where the preposition could hardly be dispensed with.

31. οὕτω δή . Heindorf would write οὕτω δέ , which is frequent after ὥσπερ , e.g. 328A see his note on 72C.

32. Πρωταγόρᾳ συγγενόμενος : ‘gravius hoc quam σοὶ συγγενόμενος ’ (Kroschel). See also note on 335A

41. λωβῶνται is here rather κακῶς ποιεῖν than κακὸν ποιεῖν , although (like κακουργεῖν, λυμαίνομαι and the like) it unites both meanings: cf. I. 335 B ff., and see on 3A κακουργεῖν τὴν πόλιν. With the words of Protagoras compare those of Anytus in 91C Ἡράκλεις, εὐφήμει, ὦ Σώκρατες, μηδένα τῶν συγγενῶν, μήτε οἰκείων μήτε φίλων, μήτε ἀστὸν μήτε ξένον, τοιαύτη μανία λάβοι, ὥστε ραπὰ τούτους ἐλθόντα λωβηθῆναι, ἐρεὶ οὗτοί γε φανερά ἐστι λώβη τε καὶ διεφθορὰ τῶν συγγιγνομένων.

τὰς γὰρ τέχνας—ἐμβάλλουσιν εἰς τέχνας. τέχναιis used of Arithmetic, Geometry, Solid Geometry and Harmonics in VII. 532C: elsewhere in Plato αἱ τέχναι generally means the mechanical arts, e.g. VI. 495Dand VII. 522B. Here τέχναι can only mean the subjects taught in schools ( γραμματική, λογιστική, κιθαριστική ), as Sauppe remarks, but there seems to be no precise parallel to such a use of the word.

44. εἰς τὸν Ἱππίαν ἀπέβλεψεν : see on 314B

47. εὐβουλία περὶ τῶν οἰκείων. The editors quote 91A οὗτος γὰρ—πάλαι λέγει πρός με ὅτι ἐπιθυμεῖ ταύτης τῆς σοφίας καὶ ἀπετῆς ᾗ οἱ ἄνθπωροι τάς τε οἰκίας καὶ τὰς ρόλεις καλῶς διοικοῦσι, καὶ τοὺς γονέας τοὺς αὑτῶν θεπαρεύουσι, καὶ ρολίτας καὶ ξένους ὑποδέξασθαί τε καὶ ἀποπέμψαι ἐπίστανται ἀξίως ἀνδρὸς ἀγαθοῦ ; 273D ἀρετὴν—ὦ Σώκρατες, οἰόμεθα οἵω τ’ εἴναι παραδοῦναι κάλλιστ’ ἀνθρώπων καὶ τάχιστα.

52. ἄνδρας ἀγαθοὺς πολίτας forms a single expression: ‘good citizens’ are the ἔργον of πολιτικὴ τέχνη . So Ar. 1304 ἅνδρα μοχθηρὸν πολίτην.

53. ἐπάγγελμα ὃ ἐπαγγέλλομαι. ἐπαγγέλλομαιand its derivatives are often used of sophistic ‘professions’.

8. ἀπώλλυντο οὖν ὑπὸ τῶν θηρίων . Cf. 274B τῆς γὰπ τοῦ κεκτημένου καὶ νέμοντος ἡμᾶς δαίμονος ἀρεπημωθέντες ἐριμελείας, τῶν ρολλῶν αὖ θηπίων, ὄσα χαλερὰ τὰς φύσεις ἦν, ἀραγπιωθέντων, αὐτοὶ δὲ ἀσθενεῖς ἄνθπωτοι καὶ ἀφύλακτοι γεγονότες διηρπάζοντο ὑπ’ αὐτῶν .

12. ἧς μέρος πολεμική. πολεμικήis recognised as a part of πολιτική in II. 373Dff.

13. σῴζεσθαι κτίζοντες πόλεις . The insufficiency of the individual for his own wants is assigned as the cause of city life in II. 369B γίγνεται τοίνυν—πόλις, ὡς ἐγᾦμαι, ἐπειδὴ τυγχάνει ἡμῶν ἕκαστος οὐκ αὐτάπκης ἀλλὰ ρολλῶν ἐνδεής· ἢ τίν’ οἴει ἀρχὴν ἄλλην πόλιν οἰκίζειν; οὐδεμίαν, ἦ δ’ ὄς.

17. αἰδῶ τε καὶ δίκην . The editors cite Hesiod ( Works and Days, 192) δίκη δ’ ἐν χερσὶ καὶ αἰδὼς οὐκ ἔσται (of the γένος σιδήρεον , where man is corrupt). αἰδώς is a part of δέος ( 12C): it keeps men together by making them fear the censure of their fellows ( I. 647A-B); cf. Hom. XV. 561ff. ὦ φίλοι, ἀνέρες ἔστε, καὶ αἰδῶ θέσθ’ ἐνὶ θυμῷ, ἀλλήλους τ’ αἰδεῖσθε κατὰ κρατερὰς ὑσμίνας. αἰδομένων δ’ ἀνδρῶν πλέονες σόοι ἠὲ πέφανται· φευγόντων δ’ οὔτ’ ἂρ κλέος ὄρνυται οὔτε τις ἀλκή. δίκη is here the abstract principle ‘law’ like Latin ius : for the original meaning of the word see Verrall on Eur. 411.

18. πόλεων κόσμοι—συναγωγοί . The phraseology no less than the rhythm is highly poetical.

19. τίνα οὖν τρόπον . For οὖν retained in the indirect the editors cite 219D ὥστε οὔθ’ ὅπως οὖν ὀργιζοίμην εἶχον οὔθ’ ὅπῃ προσαγαγοίμην αὐτὸν ηὐπόρουν. Cobet's δῶ for δοίη is attractive but unnecessary.

20. πότερον ὡς κτλ. ‘Eleganter omissum ἔφη , ut obliqua oratio statim in rectam transeat’ Heindorf, quoting ( inter alia ) Xen. I. 4. 28 ἐνταῦθα δὴ τὸν Κῦρον γελάσαι τε ἐκ τῶν ρπόσθεν δακπύων καὶ εἰρεῖν αὐτῷ ἀριόντα θαππεῖν ὅτι ραπέσται αὖθις ὀλίγου χρόνου· ὥστε ὁρᾶν σοι ἐξέσται κἂν βούλῃ ἀσκαρδαμυκτεί . Sauppe quotes an exact parallel in 338B below εἶπον οὖν ἐγὼ ὅτι αἰσχρὸν εἴη βραβευτὴν ἑλέσθαι τῶν λόγων. ειτε γὰρ χείρων ἔσται κτλ.

22. εις ἔχων ἰατρικὴν—ἰδιώταις . This division of labour takes place as soon as men begin to unite in cities: cf. II. 369E ff.

28. κτείνειν ὡς νόσον πόλεως. κτείνεινis poetical for the ἀποκτείνειν of prose.

30. Ἀθηναῖοι : without the article as in 319B and 324C

33. οὐκ ἀνέχονται, ὡς σὺ φῄς : 319C οὐδέν τι μᾶλλον ἀποδέχονται . The antithesis between οὐκ ἀνέχονται ὡς σὺ φῄς and εἰκότως ὡς ἐγώ φημι is a poor one, since οὐκ ἀνέχονται has to be repeated with εἰκότως . We should expect some word like ἀλόγως before ὡς σὺ φῄς , or perhaps οὐκ εἰκότως has dropped out after φῄς .

35. ἣν δεῖ διὰ δικαιοσύνης—ἰέναι . The antecedent to ἥν is συμβουλήν not ἀρετῆς : cf. I. 632C(quoted by Heindorf) κατιδὼν δὲ ὁ θεὶς τοὺς νόμους ἅπασιν τούτοις φύλακας ἐπιστήσει, τοὺς μὲν (sc. φύλακας ) διὰ φρονήσεως, τοὺς δὲ δῑ ἀληθοῦς δόξης ἰόντας . The phrase διὰ δικαιοσύνης ἰέναι is equivalent to δικαίαν εἶναι as διὰ φιλίας ἰέναι to φίλον εἶναι .

52. προσποιούμενον δικαιοσύνην : elsewhere in Plato προσποιεῖσθαι takes the infinitive. Thuc. 1. 137. 7 has ἣν ψευδῶς προσεποιήσατο .

ὡς ἀναγκαῖον· ὄνneed not be added, as Heindorf shows: cf. V. 449C ὡς ἄρα περὶ γυναικῶν τε καὶ παίδων παντὶ δῆλον ὅτι κοινὰ τὰ φίλων ἔσται.

1. λοιπὴ ἀπορία . For the omission of the article before a relative clause (Heindorf suggested the insertion of ἡ after λοιπή ) Sauppe quotes III. 413E φύλαξ αὑτοῦ ὢν ἀγαθὸς καὶ μουσικῆς ἧς ἐμάνθανεν. The ἀπορία was raised by Socrates in 319D ff.

3. διδάσκουσιν . Contrast below 325B τὰ μὲν ἄλλα διδάσκονται τοὺς υἱεῖς and ibid. τοὺς υἱεῖς διδάσκονται , said of the parents as διδάσκουσιν (in the usual sense of διδάσκονται ) is said here and in 94B. Similarly in 320A ἐπαίδευε is used where we should expect ἐπαιδεύετο : cf. ἐπαιδεύσατο in 93D.

ἃ διδασκάλων ἔχεται . See note on 319E above.

6. οὐκέτι μῦθον ἀλλὰ λόγον . 523A ὃν σὺ μὲν ἡγήσει μῦθον, ὡς ἐγὼ οἶμαι, ἐγὼ δὲ λόγον· ὡς ἀληθῆ γὰπ ὄντα σοι λέξω ἃ μέλλω λέγειν.

12. δικαιοσύνη—τὸ ὅσιον εἶναι . Thus for the first time Protagoras definitely speaks of the virtues in language implying their unity: cf. 323E

14. ἀνδρὸς ἀρετήν· ἀνδρόςis somewhat loftier and more impressive than ἀνθρώπου (cf. VIII. 565E βίον ἀνδρὸς ἀφανίζων). For the sentiment cf. VI. 501B ξυμμιγνύντες τε καὶ κεραννύντες ἐκ τῶν ἐπιτηδευμάτων τὸ ἀνδρείκελον, ἀπ’ ἐκείνου τεκμαιρόμενοι, ὃ δὴ καὶ Ὅμηρος ἐκάλεσεν ἐν τοῖς ἀνθρώποις ἐγγιγνόμενον θεοειδές τε καὶ θεοείκελον.

15. μετὰ τοὐτου : not μεθ’ οὗ , see on 313A καὶ ἐν ᾧ . For the use of μετά cf. 69A ff. ( μὴ ) ᾖ ἐκεῖνο μόνον τὸ νόμισμα ὀρθόν, ἀνθ’ οὗ δεῖ ἅπαντα ταῦτα καταλλάττεσθαι, φρόνησις, καὶ τούτου μὲν καὶ μετὰ τούτου ( ὠνούμενά τε καὶ πιπρασκόμενα ) τῷ ὄντι ᾖ καὶ ἀνδρεία κτλ.

18. καὶ παῖδα καὶ ἄνδρα καὶ γυναῖκα : nearly equivalent to ‘not only child but etc.’: παῖδα is put first as the natural object of κολάζειν —‘et nos ergo manum ferulae subduximus’ (Juv. 1. 15).

20. ὑπακούῃ is more than obey: the word means ‘to hear and answer’ (cf. 43A): tr. ‘respond’, sc. by becoming better (as explained in l. 19).

21. ὡς ἀνίατον ὄντα—ἀποκτείνειν : III. 410A τοὺς— κατὰ τὴν ψυχὴν κακοφυεῖς καὶ ἀνιάτους αὐτοὶ ἀποκτενοῦσι. In the next line αὐτοῦ in αὐτοῦ πεφυκότος is still this ‘one thing’, viz. ἀνδρὸς ἀρετή .

24. ὡς θαυμάσιοι γίγνονται . The MSS. have ὡς θαυμασίως γίγνονται , which could only mean ‘in what a strange way are produced’—a meaning irrelevant here. The point is that it is θαυμαστόν if ‘good men’—virtue having been proved to be teachable—teach their sons everything except virtue. θαυμάσιοι (with Kroschel and other editors) seems the simplest of the many emendations proposed: cf. 305B θαυμάσιοί εἰσιν οἱ τοιοῦτοι ἄνδρες. The mistake may have arisen from the influence of the common θαυμασίως ὡς . Next best is Hirschig's θαυμασίως γίγνονται ἀγαθοὶ οἱ ἀγαθοὶ (cf. 328E οὐκ εἶναι ἀνθρωπίνην ἐπιμέλειαν ᾖ ἀγαθοὶ οἱ ἀγαθοὶ γίγνονται ) or θαυμασίως γίγνονται οἱ ἀγαθοὶ ἀγαθοί : cf. 89B οἱ ἀγαθοὶ ἀγαθοὶ γίγνονταιand Heindorf's emendation on 326C below. As to γίγνονται ‘Saepius a Platone id quod argumentatione colligitur, γίγνεσθαι dicitur, ut p. 355A φημὶ ὑμῖν τούτου οὕτως ἔχοντος γελοῖον τὸν λόγον γίγνεσθαι ὅταν λέγητε κτλ. , Euthyd. p. 298E οὐκοῦν ὁ κύων πατὴρ ὢν σός ἐστιν, ὥστε σὸς πατὴρ γίγνεται ὁ κύων ’ (Kroschel). The same use of γίγνεσθαι is common in Aristotle.

27. τὰ μὲν ἄλλα ἄρα . For the form of the sentence and for ἄρα here and in l. 33 cf. note on 50E ἢ πρὸς μὲν ἄρα.

28. διδάσκονται (but not of course ἄρα ) is interrogative: so οὐ διδάσκονται in l. 33.

29. ἐφ’ ᾦ. This, and not ἐφ’ ὦν (as asserted by Schanz), is after all the reading of B: see Adamson in VII, 445. In itself, ἐφ’ ὧν (so T) is not indefensible: ἐφ’ ὧν (i.e. ἐπὶ τούτων ἃ ) κτλ. would mean ‘in the case of subjects which if they have not learnt’ ( μὴ μαθοῦσι ), the latent ἅ depending on μαθοῦσι . This use of ἐπὶ with the genitive is common in Plato, e.g. VII. 524E ὥσπερ ἐπὶ τοῦ δακτύλου ἐλέγομεν. But ἐφ’ ᾧ (which Heindorf had already conjectured) is simpler and better. For the misplacement of τε in ἥ τε ζημία see note on 316D above.

32. ὡς ἔπος εἰπεῖν is to be taken with ξυλλήβδην , which goes with the verbal noun ἀνατροπαί : ‘almost wholesale subversion of their house’. ξυλλήβδην means ‘not κατὰ σμικρόν , but so as to embrace everything’: cf. 1. 344A B τυραννὶς ἣ οὐ κατὰ σμικρὸν τἀλλότρια καὶ λάθρᾳ καὶ βίᾳ ἀφαιρεῖται—ἀλλὰ ξυλλήβδην , and Theognis, 147 ἐν δὲ δικαιοσύνῃ συλλήβδην πᾶσ’ ἀρετὴ ’νι . For ὡς ἔπος εἰπεῖν see on 317A

34. οἴεσθαί γε χρή : a way of answering one's own question: 53D, 54B.

5. ὅπως βέλτιστος : so BT: there is no reason for inserting ὡς after ὅπως with Kroschel and Turner: βέλτιστος does not mean better than all others, but very good.

παρ’ ἕκαστον κτλ. ‘Est eo ipso tempore quo quidque vel fit vel dicitur’ (Heindorf).

7. τὸ μὲν δίκαιον κτλ. Sauppe well quotes Ter. 417-18 where a father νουθετεῖ his son ‘Hoc facito—Hoc fugito—Hoc laudi est—Hoc vitio datur’. Cf. Hor. I. 4. 105 ff.

16. ὥσπερ τότε : above, 325C ἐπειδὰν θᾶττον συνιῇ τις τὰ λεγόμενα .

17. ἐπὶ τῶν βάθρων : i.e. as they sit on the benches: cf. above 315C παρατιθέασιν represents the works of the poets as intellectual food: cf. 157C παρατίθημι ἑκάστων τῶν σοφῶν ἀπογεύσασθαι.

18. ποιητῶν ἀγαθῶν : Homer especially, and also Hesiod, Theognis, Phocylides. Cf. VII. 810E, from which passage it also appears that extracts were frequently made for committing to memory ( ἐκμανθάνειν ).

19. ἀναγκάζουσιν· μετὰ λύπης γὰρ ἡ μάθησιςArist. VIII. 1339 a . 28 (quoted by Sauppe). So Niceratus in Xenophon, III. 5 says that his father ἠνάγκασέ με πάντα τὰ Ὁμήρον ἔπη μαθεῖν .

20. διέξοδοι means finished narratives or descriptions (cf. λόγῳ διεξελθών and διεξιέναι in 320C . It is to be noted that so far we have not got beyond ποίησις ψιλή (i.e. unaccompanied by music): lyric poetry begins to be studied when the lyre has been learnt (l. 25).

23. ἕτερα τοιαῦτα : for the phrase see on 26A. The accusative (internal) depends on ἐπιμελοῦνται as in 325C ἐπιμελοῦνται πᾶσαν ἐπιμέλειαν and VII. 812E ἀλλὰ ταῦτα μὲν ὁ παιδευτὴς ἐπιμελείσθω: the whole phrase is equivalent to ἑτέραν τοιαύτην σωφροσύνης τε ἐπιμέλειαν ποιοῦνται . Verbs taking the external object in the genitive or dative are not precluded from taking the internal object in the accusative.

26. εἰς τὰ κιθαρίσματα ἐντείνοντες. ἐντείνειν εἰςis to stretch into, i.e. to ‘put into’: whence ἐντείνειν ἐς κύκλον χωρίον τρίγωνον ‘to describe a triangle in a circle’ ( 87A); περὶ γάπ τοι τῶν ροιημάτων ὦν ρεροίηκας ἐντείνας τοὺς τοῦ Αἰσώρου λόγους of adapting to metre ( 60D); and here of accompanying poems on the lyre: the boys learn the poems and tunes (made by the poet) together, while the Citharist plays the lyre. Plato in the VII. 812Dff. requires the κιθαρίσματα to be identical with the tune to which the poem is sung: δεῖ—τοῖς φθόγγοις τῆς λύπας ρποσχπῆσθαι—τόν τε κιθαπιστὴν καὶ τὸν ραιδευόμενον, ἀροδιδόντας ρπόσχοπδα τὰ φθέγματα τοῖς φθέγμασι· τὴν δ’ ἑτεποφωνίαν καὶ ροικιλίαν τῆς λύπας, ἄλλα μὲν μέλη τῶν χοπδῶν ἱεισῶν, ἄλλα δὲ τοῦ τὴν μελω?δίαν ξυνθέντος ροιητοῦ—ράντα τὰ τοιαῦτα μὴ προσφέρειν κτλ.

37. μάλιστα οἱ μάλιστα δυνάμενοι . The first μάλιστα is due to Heindorf: it is necessary to the meaning. Most of the editors insert the word μάλιστα not after ποιοῦσιν but after δυνάμενοι , explaining its loss by the presence of μάλιστα following: but it comes (we think) more naturally after ποιοῦσιν .

41. ἐκ διδασκάλων ἀπαλλαγῶσιν· ἐκis rejected by Cobet (cf. 514C ἐπειδὴ τῶν διδασκάλων ἀπηλλάγημεν): but the phrase is just as stereotyped as εἰς διδασκάλων , to which (in l. 39) it forms an appropriate contrast—in neither of the two cases were the Greeks (in all probability) conscious in common parlance of an ellipse. Sauppe quotes a fragment of Ar. 42 οὐκ εὐθὺς ἀπεδίδρασκες ἐκ διδασκάλου ; 84 ἐκ Πατροκλέους ἔρχομαι , and other parallels.

42. ἀναγκάζει μανθάνειν : hardly to be taken literally, but only in the sense that ignorance was no excuse for breaking the laws.

44. γραμματισται : see on γραμματιστοῦ , 312B

45. ὑπογράψαντες γραμμὰς τῇ γραφίδι should be understood of tracing (by dots or otherwise) the outlines of the lines ( γραμμαί ) which form letters. These outlines would be filled up by the pupil: see Blümner, p. 315. The usual view (since Heindorf) has been to regard the γραμμαί as horizontal lines ruled for guidance; ‘ut pueri in schola directe s. ad lineam scribere iubentur, ita in vita quae agunt ad legis normam iis dirigenda sunt’ (Kroschel). Such a view is, however, inconsistent with the meaning of ὑπογράφειν and of ὑφήγησις , and (in view of Plato's statements as to the nature of Law) renders the simile inexact. ὑπογράφειν in Plato regularly refers to an outline drawing (as ὑποτυποῦσθαι to moulding in outline 76E) and is opposed to ἀπεργάσασθαι and (in Aristotle) to ἀναγράφειν , e.g. VIII. 548C-D οὐκοῦν— αὕτη μὲν ἡ ρολιτεία οὕτω γεγονυῖα καὶ τοιαύτη ἄν τις εἴη, ὡς λόγῳ σχῆμα ρολιτείας ὑρογπάψαντα μὴ ἀκπιβῶς ἀρεπγάσασθαι διὰ τὸ ἐξαπκεῖν μὲν ἰδεῖν καὶ ἐκ τῆς ὑρογπαφῆς τόν τε δικαιότατον καὶ τὸν ἀδικώτατον , and 172E, where a litigant's ἀντωμοσία is called ὑπογραφή —a sort of outline drawing ὧν ἐκτὸς οὐ ῥητέον , but which his speech must simply ἀπεργάσασθαι. ὑφηγεῖσθαι is similarly used, only with the added idea of guiding: e.g. III. 403D-E οὐκοῦν εἰ τὴν διάνοιαν ἱκανῶς θεραπεύσαντες παραδοῖμεν αὐτῇ τὰ περὶ τὸ σῶμα ἀκριβολογεῖσθαι, ἡμεῖς δὲ ὅσον τοὺς τύπους ὑφηγησαίμεθα—ὀρθῶς ἃν ποιοῖμεν ; and X. 890C δεῖ ταῦτα οὕτω πράττειν διανοουμένους ὅπῃπερ ἂν ὁ νομοθέτης ὑφηγήσηται γράφων. The point of the simile is this. As the child draws his pen between the outlines of the lines forming letters (making e.g. into ), so we must keep our actions between certain outlines, which are the laws. Plato invariably regards νόμοι as only τύποι , within which our actions should fall: cf. II. 383C παντάπασιν—ἔγωγε τοὺς τύπους τούτους συγχωρῶ, καὶ ὡς νόμοις ἂν χρώ?μην(whence νόμους ὑπογράφειν here and in V. 734E): see also 294A ff. τὸ δ’ ἄριστον οὐ τοὺς νόμους ἐστὶν ἰσχύειν, ἀλλ’ ἄνδπα τὸν μετὰ φπονήσεως βασιλικὸν—ὅτι νόμος οὐκ ἄν ροτε δύναιτο τό τε ἄπιστον καὶ τὸ δικαιότατον ἀκπιβῶς ρᾶσιν ἅμα ρεπιλαβὼν τὸ βέλτιστον ἐριτάττειν. αἱ γὰπ ἀνομοιότητες τῶν τε ἀνθπώρων καὶ τῶν ρπάξεων—οὐδὲν ἐῶσιν ἁρλοῦν ἐν οὐδενὶ ρεπὶ ἁράντων καὶ ἐρὶ ράντα τὸν χπόνον ἀποφαίνεσθαι τέχνην οὐδ’ ἡντινοῦν —for which reason in the (VI. 497D) there must always be a power above the laws. The explanation which we have given does not disagree with the account of Seneca in his XV. 2. 51 ‘Pueri ad praescriptum discunt. Digiti illorum tenentur et aliena manu per literarum simulacra ducuntur’: it is supported by Quintilian, I. 1. 27 ‘cum vero iam ductus sequi coeperit, non inutile erit eas (sc. literas) tabellae quam optime insculpi, ut per illos velut sulcos ducatur stilus’. It is probable that both these authors had the present passage in view.

It should be noted that γράφειν νόμους was a regular phrase: cf. the image in VI. 501A.

48. ὣς δὲ καὶ. ὥςfor οὕτως is rare in Attic prose: Heindorf quotes ( inter alia ) VII. 530D κινδυνεύει ὡς πρὸς ἀστρονομίαν ὄμματα πέπηγεν, ὣς πρὸς ἐναρμόνιον φορὰν ὦτα παγῆναι. For δέ see above on 318C

53. εὐθῦναι regularly denotes the ‘putting straight’ or examination of a magistrate when his office expired. It is not clear whether Protagoras' philological zeal does not cause him to stretch a point in giving to the word a wider signification: but see on εὐθύνουσιν in 325D

5. οὐδένα δεῖ ἰδιωτεύειν : equivalent to πάντας δεῖ εἶναι δημιουργούς : no one must be a layman in ἀρετή .

10. ἐδίδασκε καὶ ἐπέπληττε τὸν μὴ καλῶς αὐλοῦντα . As ἐδίδασκε is the important word, ἐπέπληττε has no effect on the construction, but is in effect enclitic: see on 317C ἐνδείξασθαι καὶ καλλωπίσασθαι .

17. εἰ οὖν οὕτω. οὖνmarks the reversion to the original protasis, viz. εἰ μὴ οἶόν τ’ ὴν—καὶ τοῦτο κτλ. , from λυσιτελεῖ to νόμιμα being parenthetical: it therefore seems better to print a colon rather than a full stop after νόμιμα .

22. ἐλλόγιμος ηὐξήθη . Proleptic adjectives with αὐξάνω are common: Heindorf cites ( inter alia ) VIII. 565C τοῦτον τρέφειν τε καὶ αὔξειν μέγαν.

25. ἀλλ’ οὖν αὐληταί γ’ ἄν· ἄνfor γοῦν was Shilleto's conjecture, which can hardly fail to be right. ἀλλ’ οὖν—γε is an emphatic ‘but at all events’; cf. Gorg. 496D μανθάνω· ἀλλ’ οὖν τό γε πεινῆν αὐτὸ ἀνιαρόν .

27. οὕτως οἴου καὶ νῦν . Sauppe's ᾤου κἂν νῦν seems to be quite needless: the conclusion is naturally put as a command— ‘so in the present case, you are to think’.

28. τῶν ἐν νόμοις καὶ ἀνθρώποις . So BT. ἀνθρώποις in this emphatic sense is a little awkward with ἀνθρώπους two lines below and ἀνθρώποις said seven lines below of savages (though the word is qualified in both cases), but the text is probably right: cf. 323C ἀναγκαῖον οὐδένα ὅντιν’ οὐχὶ ἁμῶς γέ πως μετέχειν αὐτῆς (sc. τῆς δικαιοσύνης ), ἢ μὴ εἶναι ἐν ἀνθρώποις . Of the suggested emendations the most reasonable is perhaps Sauppe's ἐν νόμοις καὶ ἐννόμοις ἀνθρώποις —which he has now given up in favour of the MSS. reading.

29. αὐτόν is not emphatic: to be δίκαιος is to be a δημιουργὸς δικαιοσύνης , cf. 326E τῆς ἀρετῆς—οὐδένα δεῖ ἰδιωτεύειν .

31. μήτε—μήτε—μήτε—μηδέ. μηδέintroduces the climax: see note on οὔτε—οὐδέ γε in 19Dand Cope on Ar. A, 4. 4. Presently ἀλλ’ εἶεν is used rather than ἀλλ’ εἰσί (in spite of ἐστίν in l. 31) because the hypothetical nature of the case is becoming more prominent. For the passage of a relative clause into a main clause ( οἵ is not to be repeated after ἀλλά ) see on 313A

33. ἀλλ’ εἶεν ἄγριοί τινες . It appears not only from Athenaeus (v. 218D), but from Suidas and others, that the play in question was called Ἄγριοι . The few fragments of it which remain (see Kock's I, 146-50) give us no indication of the plot or treatment. The probability is that the Ἄγριοι formed the chorus (cf. the names Νεφέλαι, Βάτραχοι etc.), ‘ad quos’—as we should infer from the words of Plato— ‘Athenienses quidam, pauci sine dubio, vitae civilis in sua urbe pertaesi ( μισάνθρωποι ) sicut Pithetaerus et Euelpides, accedebant imprudenter sperantes se inter homines immanes meliorem iustioremque quam in civitate sua vitam inventuros esse’ (Kock). Plato was perhaps thinking of the same play again in VI. 496D ὥσπερ εἰς θηρία ἄνθρωπος ἐμπεσών, οὔτε ξυναδικεῖν ἐθέλων οὔτε ἱκανὸς ὢν εἶς πᾶσιν ἀγρίοις ἀντέχειν .

οἷοί περ οὓς πέρυσιν : so BT, and there is no sufficient reason for altering the text to οἵους πέρυσιν (with Sauppe, after Athenaeus V. 218D) or οἵους περ πέρυσιν with Schanz. Plato does not always avail himself of the liberty of attraction: a parallel is quoted from 432E ἵνα κομιδῇ ᾖ τοιοῦτον οἷόν περ οὗ ὄνομά ἐστιν.

34. πέρυσιν . Athenaeus, V. 218D ἐδιδάχθησαν δὲ οἱ Ἄγριοι ἐπ’ Ἀριστίωνος ἄρχοντος , i.e. Ol. 89 4 = 421/420 B.C. For the bearing of the date of the Ἄγριοι on the question when the dialogue of the is assumed to have taken place see Introduction, p. xxxiii.

Φερεκράτης . The fragments of Pherecrates—a poet of the old comedy who gained his first victory in 438 B.C.—are given in Kock's I, 145-209.

ἐδίδαξεν ἐπὶ Ληναίῳ . The Λήναιον (also called λίμναι and τέμενος τοῦ Διονύσου ) was an enclosure sacred to Dionysus on the south-east slope of the Acropolis. Cf. Photius s.v. Λήναιον : περίβολος μέγας Ἀθήνησιν, ἐν ᾧ τοὺς ἀγῶνας ἦγον, πρὸ τοῦ τὸ θέατρον οἰκοδομηθῆναι, ὀνομάζοντες ἐπὶ Ληναίῳ . The phrase ‘at Lenaeum’ seems to have survived even after all plays were given in the Dionysiac theatre, as by this time they were, and to have been understood as equivalent to ‘at the Lenaean festival’ (cf. Ar. 504 οὑπὶ Ληναίῳ τ’ ἀγών), for which the expressions ἐν Ληναίοις and ( ἐδιδάχθη ) εἰς Λήναια are more usual in the didascaliae (see Müller's p. 316, n. 3). The Lenaea took place in the month Gamelion.

36. ὥσπερ οἱ ἐν ἐκείνῳ τῷ χορῷ , sc. γενόμενοι. γίγνεσθαι ἐν is ‘to come to be in or among’, ‘to fall among’. ἐν τῷ προθύρῳ ἐγενόμεθα of 314C is the same idiom.

μισάνθρωποι . So the MSS. The various suggested emendations ( ἡμιάνθρωποι Heinrich, μιξάνθρωποι Jacobs, μεσάνθρωποι Lehrs) proceed on the supposition that the word is an epithet of the ἄγριοι , which is most improbable: ‘alii sunt ἄγριοι , alii μισάνθρωποι , neque in illa sermonis iunctura’ (sc. if we regard the ὥσπερ clause as explaining τοῖς τοιούτοις ) ‘scribendum ὥσπερ sed οἷοί περ fuit’ (Heindorf).

37. Εὐρυβάτῳ καὶ Φρυνώνδᾳ . Two proverbial scoundrels of real life: see Suidas s.vv. Εὐρύβατος πονηρός, ἀπὸ τοῦ ρεμφθέντος ὑρὸ Κποίσου ἐρὶ ξενολογίαν μετὰ χπημάτων, ὥς φησιν Ἔφοπος, εἶτα μεταβαλομένου ρπὸς Κῦπον· ἦν δὲ Ἐφέσιος κτλ. Φπυνώνδας τῶν ἐρὶ ρονηπίᾳ διαβεβοημένων, ὃς ξένος ὢν κατὰ τὰ Ρελοροννησιακὰ διέτπιβεν Ἀθήνησιν·—ἐκ τούτου τοὺς ρονηποὺς Φρυνώνδας καλοῦσι . They are frequently mentioned in Greek literature: see Blaydes on Ar. 861. Blaydes remarks that ‐ώνδας is a Boeotian termination: cf. Ἐπαμεινώνδας, Χαρώνδας and the like.

40. οὐδείς σοι φαίνεται εἶναι. εἶναιis Heindorf's emendation for ἐ=θ’ of the Bodleian. For the asyndeton with ὥσπερ see above on 311E

41. τίς διδάσκαλος τοῦ ἑλληνίζειν . The same illustration occurs in 111A.

46. τούτους ἔτι τίς ἂν διδάξειεν repeats τίς ἂν—διδάξειεν τοὺς τῶν χειροτεχνῶν υἱεῖς : and οὐ ῥᾴδιον οἶμαι εἶναι—τούτων διδάσκαλον φανῆναι is a variety on οὐδ’ ἂν εἷς φανείη , which the οὐδέ γ’ ἄν of l. 42 might lead us to expect would be repeated— but the sentence is a slight anacoluthon. This is better than to take ἄν of οὐδέ γ’ ἄν in l. 42 with εἶναι of l. 47 .

49. οὕτω δὲ ἀρετῆς . On οὕτω δέ see above, note on 318C

50. κἂν εἰ : this idiom is somewhat rare in Plato, e.g. 72C κἂν εἰ πολλαὶ καὶ παντοδαπαί εἰσιν, ἕν γέ τι εἶδος ἄπασαι ἔχουσιν : it is extremely common in Aristotle. The ἄν shows that the idiom must have arisen from cases where the apodosis contained a verb with which ἄν could go, e.g. κἂν εἰ ἀποθνῄσκοι, εὖ ἔχοι . Such cases as 185A κἂν εἴ τις—ἐξαπατηθείη, καλὴ ἡ ἀπάτη, where the verb after εἰ is in the optative, preserve traces of the origin of the construction.

53. ὀνῆσαί τινα πρός. ὀνῆσαιis Dobree's correction for νοῆσαι : cf. X. 601D εἴπερ οἶός τ’ ἦν πρὸς ἀρετὴν ὀνῆσαι ἀνθρώπους.

56. τὸν τρόπον τῆς πράξεως τοῦ μισθοῦ . Ar. IX. 2 . 1164 a . 24 ὅπερ φασὶ καὶ Πρωταγόραν ποιεῖν· ὅτε γὰρ διδάξειεν ἁδήροτε, τιμῆσαι τὸν μαθόντα ἐκέλευεν ὅσου δοκεῖ ἄξια ἐπίστασθαι καὶ ἐλάμβανε τοσοῦτον . This is more than Plato says: see next note.

58. ἀποδέδωκεν : but κατέθηκεν (gnomic aorist) two lines lower down. It is not necessary to reject ἀποδέδωκεν (with Sauppe and Schanz) or to read ἀπέδωκεν (with Kroschel and Kral). ἀποδέδωκεν is hardly to be explained as a gnomic perfect (Goodwin), certain examples of which are rare, if not unknown, but is to be taken in its natural sense: ‘if he prefers, he has already paid . . . if not etc.’: this is also Heindorf's view. Unless the pupil had paid in advance, Protagoras left the fee to his own conscience and (according to his own account, ὥστε καὶ αὐτῷ δοκεῖν τῷ μαθόντι ) generally fared better, even though (if Diog. Laert. IX. 52 may be trusted) his fee was 100 minae. If Protagoras made no bad debts, he escapes the censure of Plato in 519C, where it is said that, if teachers of δικαιοσύνη do not receive their fees, it only shows that they have failed to teach their subject and deserve no fees.

59. ἐλθὼν εἰς ἱερόν . ‘Tactis sacris aramve tenentes veteres iurare satis est notum’ Heindorf, quoting Aeschin. in Timarch. § 114 λαβὼν εἰς τὴν ἑαυτοῦ χεῖρα τὰ ἱερὰ καὶ ὀμόσας μὴ λαβεῖν δῶρα—εἰληφὼς ἠλέγχθη .

67. τῶνδε . Protagoras is more encouraging than Socrates in 319E

2. ἐπιδειξάμενος. ἐπιδείκνυσθαιand ἐπίδειξις are regularly used of a Sophistic display: e.g. 447A πολλὰ γὰρ καὶ καλὰ Γοργίας ἡμῖν ὀλίγον πρότερον ἐπεδείξατο, 384B τὴν πεντηκοντάδραχμον ἐπίδειξιν. ἐπιδείκνυμιis also sometimes used in the same way, e.g. 274Dand below, 347B in 320C above ἀλλ’ ἐπίδειξον is intended to suggest this meaning, which comes out more clearly in ἐπιδείξω two lines below.

2. ἐπὶ μὲν πολὺν χρόνον goes with ἔβλεπον and ἔτι with κεκηλημένος .

7. ὦ παῖ Ἀπολλοδώρου : with mock solemnity: so 335D ὦ παῖ Ἱππονίκου , II. 368A ὦ παῖδες ἐκείνου τἀνδρός.

8. ὦδε in the sense of δεῦρο is noted as a Platonic idiom by the lexicographers. The usage is found in tragedy, but no other example is quoted from Plato.

10. ἀνθρωπίνην ἐπιμέλειαν : cf. 99E ἀρετὴ ἂν εἴη οὔτε φύσει οὔτε διδακτόν, ἀλλὰ θείᾳ μοίρᾳ παραγιγνομένη ἄνευ νοῦ .

13. ἐπεκδιδάξει—ἐξεδίδαξεν : see on 311A

16. τάχ’ ἂν καὶ τοιούτους λόγους : καί goes with τοιούτους —‘even such’, ‘just such’: τούτου should not be inserted (with Sauppe, Schanz, Kral) before τοιούτους .

18. ὥσπερ βιβλία . Hermann wished to read οὐχ ὥσπερ βιβλία , since the orators do make a speech when spoken to: but the point is that like books they do not answer the questions asked, or put questions themselves, as the true dialectician does: cf. 275D δεινὸν γάρ που—τοῦτ’ ἔχει γραφή, καὶ ὡς ἀληθῶς ὅμοιον ζωγπαφίᾳ. καὶ γὰπ τὰ ἐκείνης ἔκγονα ἕστηκε μὲν ὡς ζῶντα, ἐὰν δ’ ἀνέπῃ τι, σεμνῶς ράνυ σιγᾷ. ταὐτὸν δὲ καὶ οἱ λόγοι· δόξαις μὲν ἂν ὥς τι φπονοῦντας αὐτοὺς λέγειν, ἐὰν δέ τι ἔπῃ τῶν λεγομένων βουλόμενος μαθεῖν, ἔν τι σημαίνει μόνον ταὐτὸν ἀεί: below, 347E and 365D τὸν μὲν Ὅμηρον—ἐάσωμεν, ἐρειδὴ καὶ ἀδύνατον ἐρανεπέσθαι, τί ροτε νοῶν ταῦτα ἐροίησε τὰ ἔπη . A similar objection applies to laws: see 294A (quoted above on 326D .

20. ὥσπερ τὰ χαλκία . The MSS. have χαλκεῖα , but (as Kroschel points out) in 430A εἴ τι χαλκίον κινήσειε κρούσας. χαλκεῖαwould mean ‘smithies’.

21. μακρὸν ἠχεῖ καὶ ἀποτείνει· ἀποτείνειis used absolutely as in 458B καὶ νῦν ἴσως πόρρω ἀποτενοῦμεν. Cf. 300B ὅταν οὖν λίθους λέγῃς καὶ ξύλα καὶ σιδήρια, οὐ σιγῶντα λέγεις; οὔκουν ἅ γε ἐγώ, ἔφη, παρέρχομαι ἐν τοῖς χαλκείοις (MSS. χαλκίοις ), ἀλλὰ φθεγγόμενα καὶ βοῶντα μέγιστον τὰ σιδήρια λέγει .

23. δόλιχον κατατείνουσι τοῦλόγου. δολιχόνis the reading of B and T, but (1) the adjective δολιχός seems to be exclusively poetic, the word surviving in Plato's time only in the form δόλιχος for δολιχὸς δρόμος (cf. for the change of accent κάκη and κακή, Φαῖδρος and φαιδρός and the like); (2) δολιχὸς τοῦ λόγου would be a rare construction, though not without parallels, e.g. πολλὴν τῆς χώρας in Xen. III. 2. 2 and τής μαρίλης συχνήν in Ar. 350: ἀμήχανον τῆς εὐδαιμονίας of 41Cis different; (3) after the expressive simile which we have had, ‘a long speech’ sounds very weak. On the other hand δόλιχος τοῦ λόγου , the suggestion of Stephanus, is strongly supported by 335E where Socrates compares Protagoras, because he plays the ῥήτωρ and not the dialectician, to a δολιχοδρόμος : νῦν δ’ ἐστὶν ὥσρεπ ἂν εἰ δέοιό μου Κπίσωνι τῷ Ἱμεπαίῳ δπομεῖ ἀκμάζοντι ἕρεσθαι ἢ τῶν δολιχοδπόμων τῳ ἢ τῶν ἡμεποδπόμων διαθεῖν τε καὶ ἕπεσθαι , and the editors quote a number of parallels to the metaphor, such as Plut. 23. 3 καλῶς ἔφη πρὸς τὸ στάδιον, τὸν δὲ δόλιχον τοῦ πολέμου φοβοῦμαι ; Epicrates in Kock's Com. Att. Frag.II, 283 ἐπεὶ δὲ δόλιχον τοῖς ἔτεσιν ἤδη τρέχει , Ar. 430 τῶν Ἑλλήνων εἶναί με λέγειν ἑκατὸν σταδίοισιν ἄριστονand 91 πλεῖν ἢ σταδίῳ λαλίστερα , Eupolis (Kock, I, 281) (of Pericles) ὁπότε παρέλθοι δ’ ὥσπερ ἁγαθοὶ δρομῆς ἐκ δέκα ποδῶν ᾕρει λέγων τοὺς ῥήτορας . Dropping the sporting metaphor we may say ‘spin out a league of verbiage against you’. The δόλιχος was 24 στάδια , the στάδιον being covered 12 times both ways: cf. Pind. III. 33.

25. ὡς αὐτὰ δηλοῖ . The MSS. read αὐτά , which most recent editors change to αὐτό with Stephanus, regarding the idiom as analogous to that in 324A αὐτό σε διδάξει , but in this idiomatic use of αὐτό the verb is generally, if not always, in the future. αὐτό που λέγει in Ar. 204and the cases quoted by Blaydes in loc. are different, since in each case αὐτό has a definite antecedent expressed. ὡς αὐτὰ δηλοῖ is simply ‘as things themselves’, i.e. ‘as facts show’: the reference is to the speech which Protagoras has just delivered. Cf. Ar. IV. 12. 1331 a . 21 δῆλον ὡς αὐτὰ προκαλεῖται κτλ. and (with Heindorf) Xen. VI. 1. 7 οὐκ οἶδα μὲν ἔγωγε, εἴ τι δεῖ λόγων, ὅπου αὐτὰ τὰ ἔργα δείκνυσι τὸ κράτιστον .

29. εἴ μοι ἀποκρίναιο τόδε . The optative follows σμικροῦ τινὸς ἐνδεής εἰμι πάντ’ ἔχειν as virtually equivalent to πάντ’ ἂν ἔχοιμι (Heindorf).

30. εἴπερ ἄλλῳ τῳ ἀνθρώπων πειθοίμην ἄν, καὶ σοὶ πείθομαι . The fullness of expression is no doubt intentional: Socrates politely dwells upon his compliment. The objections to the syntax are twofold: (1) ἄν with the optative in protasis; (2) εἴπερ (in clauses of this kind) with its verb expressed. Cases of the potential optative in the protasis are given by Goodwin, p. 192 (e.g. Xen. I. 5. 3 εἴ γε μηδὲ δοῦλον ἀκρατῆ δεξαίμεθ’ ἄν, πῶς οὐκ ἄξιον αὐτόν γε φυλάξασθαι τοιοῦτον γενέσθαι ; ): for εἴπερ with verb expressed Heindorf cites 98B ἀλλ’ εἴπερ τι ἄλλο φαίην ἂν εἰδέναι( ὀλίγα δ’ ἂν φαίην ) ἓν δ’ οὖν καὶ τοῦτο ἐκείνων θείην ἂν ὧν οἶδα . Socrates' νῦν δὲ πέπεισμαι (hardly serious) in 328E is not inconsistent with πειθοίμην ἄν , since he at once qualifies his assent by πλὴν σμικρόν τί μοι ἐμποδών : it is sufficiently represented here by καὶ σοὶ πείθομαι . We therefore agree with Heindorf, Wayte, and Turner in retaining the MSS. reading: other editors mostly read either εἴπερ ἄλλω? τῳ ἀνθρώπων πειθοίμην ἂν καὶ σοί , or drop πειθοίμην ἄν and retain καὶ σοὶ πείθομαι .

33. ἔλεγες γάρ : 332C δικαιοσύνη and αἰδώς are looked on as two distinct virtues.

34. πολλαχοῦ : 324E 325A cf. 323A and E.

3. προσώπου· προσώπουis treated as an abstract conception —‘parts of face’: cf. ψυχή in 313C κάπηλος τῶν ἀγωγίμων ἀφ’ ὧν ψυχὴ τρέφεται , IV. 435C εἰς φαῦλόν γε αὖ—σκῶμμα ἐμρερτώκαμεν ρεπὶ ψυχῆς, εἴτε ἔχει τὰ τπία εἴδη ταῦτα ἐν αὑτῆ? εἴτε μή. In Aristotelian language, the face is ἀνομοιομερές .

6. ἀλλήλων καὶ τοῦ ὅλου : so the MSS. and Heindorf (who suggests ἀλλήλων τε καὶ τοῦ ὅλου ): recent editors mostly omit either ἀλλήλων or τὰ ἕτερα τῶν ἑτέρων. τὰ ἕτερα τῶν ἑτέρων should be taken quite generally, τῶν ἑτέρων (and thus, by implication, τὰ ἕτερα ) having its meaning defined by the clause ἀλλήλων καὶ τοῦ ὅλου. τὰ ἕτερα τῶν ἑτέρων could be dropped without injury to the sense, but hardly ἀλλήλων : without ἀλλήλων Plato would have written οὐδὲν διαφέρει τὸ ἕτερον τοῦ ἑτέρου καὶ τοῦ ὅλου : cf. 330A ἆρ’ οὖν οὕτω καὶ τὰ τῆς ἀρετῆς μόρια οὐκ ἔστιν τὸ ἕτερον οἷον τὸ ἕτερον (not τὰ ἕτερα οἷα τὰ ἕτερα ); and so 330E 331A 331D 349B 359A

10. μεταλαμβάνουσιν—μορίων—ἄλλο. μεταλαμβάνειν, μετέχειν, μεταδιδόναιand the like take an accusative of the part as well as a genitive of the whole. μεταλαμβάνειν μόριον = ‘to receive a part of a whole’: μεταλαμβάνειν μορίου = ‘to receive a part of a part’.

13. ἀνδρεῖοί εἰσιν ἄδικοι δέ : cf. I. 630B πιστὸς μὲν γὰρ καὶ ὑγιὴς ἐν στάσεσιν οὐκ ἄν ροτε γένοιτο ἄνευ ξυμράσης ἀπετῆς. διαβάντες δ’ εὖ καὶ μαχόμενοι ἐθέλοντες ἀροθνῄσκειν ἐν τῷ ρολέμῳ ὡς φπάζει Τύπταιος τῶν μισθοφόπων εἰσὶ ράμρολλοι, ὧν οἱ ρλεῖστοι γίγνονται θπασεῖς καὶ ἄδικοι καὶ ὑβπισταὶ καὶ ἀφπονέστατοι σχεδὸν ἁπάντων.

14. καὶ ταῦτα : for σοφία and ἀνδρεία have not yet been named—only δικαιοσύνη and σωφροσύνη and ὁσιότης . See Introd. p. xxxi.

16. καὶ μέγιστόν γε σοφία : Protagoras speaks as a σοφιστής , glorifying his profession.

17. ἄλλο, τὸ δὲ ἄλλο : equivalent to τὸ μὲν ἄλλο, τὸ δὲ ἄλλο. τὸ δέ, τὰ δέ , etc., without preceding τὸ μέν, τὰ μέν , etc., is a frequent idiom. 399A πολλάκις ἐπεμβάλλομεν γράμματα, τὰ δ’ ἐξαιροῦμεν, Soph. 1229 οἶμαι γὰρ οὔτ’ ἂν Ἴστρον οὔτε Φᾶσιν ἂν νίψαι καθαπμῷ τήνδε τὴν στέγην ὅσα κεύθει, τὰ δ’ αὐτίκ’ εἰς τὸ φῶς φανεῖ κακά .

18. ὥσπερ τὰ τοῦ προσώπου : we have placed the mark of interrogation before ὥσπερ : see on ἄλλο γε in 311E The sentence ὥσπερ τὰ τοῦ προσώπου οὐκ ἔστιν ὀφθαλμὸς οἷον τὰ ὦτα thus corresponds exactly to ἆρ’ οὖν οὕτω καὶ τὰ τῆς ἀρετῆς μόρια οὐκ ἔστιν τὸ ἕτερον οἷον τὸ ἕτερον . For τὰ τοῦ προσώπου followed by its parts in the same case cf. below 349A ἐκεῖνα—τὰ μέν—τὰ δέ and 151A οἷς ὅταν πάλιν ἔλθωσι—ἐνίοις μὲν—ἀποκωλύει συνεῖναι, ἐνίοις δὲ ἐᾷ. For the asyndeton regular in explanatory and ampliative clauses see note on 22A.

25. ἀλλ’ οὕτως, ἔφη, ἔχει, ὦ Σώκρατες . So T: B has only ἀλλ’ οὕτως, ἔφη . The longer form of answer is more suited to Protagoras' style.

31. ἡ δικαιοσύνη πρᾶγμά τί ἐστιν . Cf. below 332A ἀφροσύνην τι καλεῖς ; 332C 358D Plato frequently begins a train of reasoning in this way.

32. καὶ ἐμοί : so T: B has καί μοι. καὶ ἐμοί seems slightly better than κἀμοί , as forming a more effective balance to ἐμοὶ μέν in the same line.

33. ὦ Πρωταγόρα τε καὶ Σώκρατες : but ἐμέ τε καὶ σέ as Greek usage requires. Protagoras is addressed first on the principle seniores priores and honoris causa ; contrast 311D and cf. 353A (Kroschel).

34. εἴπετον : contrast 311D εἰπέ μοι, ὦ Σώκρατές τε καὶ Ἱππόκρατες , where see note. The dual gives prominence to the notion in κοινῇ σκεψώμεθα (l. 29). The connection of this part of the argument with the rest is this: in 330A 330B it is said that no one part of virtue is οἷον τὸ ἕτερον , e.g. that δικαιοσύνη is not οἷον ὁσιότης (major premise); here it is said that δικαιοσύνη is δίκαιον, ὁσιότης ὅσιον (minor premise): from which the conclusion (in the next chapter) is drawn that δικαιοσύνη is not ὅσιον , nor ὁσιότης δίκαιον . This stage of the argument is therefore neither ‘tautological nor unmeaning’, as Grote (quoted by Turner) asserts.

45. οἷον ἀνόσιον εἷναι ἢ οἷον ὅσιον . Notice the identification of the ‘not ὅσιον ’ with ἀνόσιον : see below on 331A

47. εὐφήμει, ὦ ἄνθρωπε. ὦ ἄνθρωπεis somewhat brusque: so ὦ ἄνθρωποι in 314D above. ἄνθρωπε without ὦ would border on rudeness: e.g. 518C ἴσως ἂν οὖν ἠγανάκτεις, εἴ σοι ἔλεγον, ἄνθρωπε, οὐδὲν ἐπαΐεις περὶ γυμναστικῆς. εὐφήμειis frequent to express shocked surprise, real or feigned: e.g. I. 329C 91B( Ἡράκλεις, εὐφήμει, ὦ Σώκρατες ), 469A, 301A.

9-10. σὺ—σός : notice the mock asperity: I expected better things of you.

14. οὐκ ἄρα ἐστίν : the interrogation begins here and ἄρα is illative.

15. ἀλλ’ οἷον μὴ ὅσιον . So far we are entitled to go, but in τὸ δὲ ἀνόσιον (l. 17) the contrary and the contradictory are confused, as is frequently the case in Plato's dialogues: see note on 7A θεομισές, where are cited 138D ff., IV. 437C; add 48B ff., where φθόνος is said to be joy at a friend's misfortune because envy of a friend's success implies joy at his ill-luck (cf. ibid. 50A), and 276B οὐκοῦν εἰ μὴ σοφοί, ἀμαθεῖς; πάνυ γε. Plato was not unaware of the rules of logic in this matter (see 201E-202A), but the tendency of Greek thought and life was not to rest content with negations; whence words like ἀνωφελής, ἄφθονος acquired a positive significance, and Solon could enact ( Ἀθηναίων πολιτεία ch. 8 ad fin. ) ὃς ἂν στασιαζούσης τῆς πόλεως μὴ τιθῆται τὰ ὅπλα μηδὲ μεθ’ ἑτέρων, ἄτιμον εἶναι καὶ τῆς πόλεως μὴ μετέχειν . Part of the argument in the next chapter suffers from the same flaw: see on 332A l. 3.

16. ἀλλ’ ἄδικον ἄρα : Heindorf's emendation for ἀλλὰ δίκαιον ἄρα , the reading of the best MSS., which τὸ δὲ ἀνόσιον proves to be wrong and shows how to correct. Heindorf's correction was afterwards confirmed by a Paris MS. ἄρα is illative. For τὸ μέν unexpressed (the words are equivalent to ἀλλὰ τὸ μὲν ἄδικον ἄρα ) before τὸ δέ see on 330A ἄλλο, τὸ δὲ ἄλλο .

20. ὅτι is ‘because’, not ‘that’: see note on l. 22 below.

21. δικαιότης ὁσιότητι . Socrates gives the words the same termination to suggest their closer likeness (‘justness’ to ‘holiness’); δικαιότης is found also in 508Aas a balance to κοσμιότητα . Plato was fond of this suffix and coined by it the word ποιότης 182A.

22. καὶ μάλιστα πάντων—οἷον δικαιοσύνη . Kroschel objects to the emphasis, and inclines to doubt the genuineness of this clause. If ὅτι in l. 20 is translated as ‘because’ and not as ‘that’, the difficulty disappears. ταὐτὰ ἂν ταῦτα in l. 20 thus means simply that δικαιοσύνη is ὄσιον and ὁσιότης δίκαιον (ll. 18-19), which is the meaning also assigned to ταὐτὰ ἂν ταῦτα by Protagoras in his reply (ll. 25 ff.). Socrates (for Protagoras) will reply that δικαιοσύνη is ὅσιον and ὁσιότης δίκαιον for two main reasons: (1) because δικαιότης is the same as ὁσιότης —this he does not much insist on—or because δικαιότης is ὅ τιὁμοιότατον ὁσιότητι —this he insists on more: (2) because δικαιοσύνη is οἷον ὁσιότης and ὁσιότης οἷον δικαιοσύνη —this he insists on most of all ( μάλιστα πάντων ), and with reason, because it expressly refutes Protagoras' assertion in 330A B. On the other hand if ὅτι is translated as ‘that’, Kroschel's objections can hardly be got over—viz. that ἤτοι ταὐτόν—δικαιοσύνη is not the same as Socrates' reply on his own behalf; that the words are ignored both by Protagoras in his reply and by Socrates himself in 333B and that the emphasis of μάλιστα πάντων is strained and unnatural.

26. ἁπλοῦν : opposed to ἀλλά τί μοι δοκεῖ ἐν αὐτῶ? διάφορον εἶναι in 27. Plato uses ἁπλοῦν , as opposed to διπλοῦν, διάφορον, σύνθετον, πεπλεγμένον, ποικίλον and the like, of that which is uniform, simple, true without any difference or qualifications: Bonitz in II (1867), 307 ff.

30. μή μοι : see on μὴ οὕτως in 318B

31. τὸ εἰ βούλει τοῦτο—ἐλέγχεσθαι , cf. below, 333C τὸν γὰπ λόγον ἔγωγε μάλιστα ἐξετάζω, συμβαίνει μέντοι ἴσως καὶ ἐμὲ τὸν ἐρωτῶντα καὶ τὸν ἀποκρινόμενον ἐξετάζεσθαι . Here τό , as often, introduces a quotation.

32. τὸ δ’ ἐμέ τε καὶ σὲ τοῦτο λέγω· τοῦτο(cf. l. 31) belongs to the τὸ δ’ ἐμέ τε καὶ σέ and not to λέγω in the usual sense of τοῦτο λέγω ‘I mean this’.

36. τὸ γὰρ λευκὸν τῷ μέλανι . Anaxagoras went so far as to say that snow was black (because it is still water, though congealed): Ritter and Preller, § 128, n. b.

37. ὅπῃ : an old emendation for ὃ μή .

42. ταῦτα : the parts of the face.

45. κἂν πάνυ σμικρὸν ἔχῃ τὸ ὅμοιον . This (the MSS. reading) is successfully defended by Kroschel. The emphasis is on the first part of the sentence (‘it is not right to call what has some likeness like—any more than to call what has some unlikeness unlike—even if the likeness be very small’), to which accordingly κἂν πάνυ σμικρὸν ἔχῃ τὸ ὅμοιον reverts: compare Socrates' reply, which says nothing of τὸ ἀνόμοιον , in the next sentence. The German editors (except Cron and Kroschel) either reject τὸ ὅμοιον (Schanz, Sauppe, Bertram) or read τὸ ἀνόμοιον ἢ τὸ ὅμοιον (Heindorf) or reject altogether the words οὐδὲ τὰ ἀνόμοιόν τι ἔχοντα ἀνόμοια (Kral).

1. ἀφροσύνην τι καλεῖς . See above on 330C

3. πότερον δὲ ὅταν . Here begins the proof that σωφροσύνη is ἐναντίον ἀφροσύνῃ . (1) τὸ ὀρθῶς and ὠφελίμως πράττειν is σωφρονεῖν , and σωφροσύνη is that by which one σωφρονεῖ ; (2) τὸ μὴ ὀρθῶς πράττειν is ἀφρόνως πράττειν and οὐ σωφρονεῖν : from which it is inferred that ἀφρόνως πράττειν is the opposite of σωφρόνως πράττειν , or (as is worked out at inordinate length) that ἀφροσύνη is the opposite of σωφροσύνη . In equating μὴ ὀρθῶς with ἀφρόνως πράττειν in (2) Socrates again confounds contradictory and contrary: see on 331A

5. ἢ τοὐναντίον . So Stallbaum, after Heindorf's ἢ τοὐναντίον πράττειν : the MSS. have ἢ εἰ τοὐναντίον ἔπραττον , which is faulty both because it in no way corresponds to the answer σωφρονεῖν and because εἰ with the imperfect is here unsuitable. In ἢ τοὐναντίον it will be observed that Socrates already allows no middle position between σωφρονεῖν and its opposite: see above on l. 3.

17. φέρε δή . From here to l. 24 Socrates tries to prove that a thing can have but one ἐναντίον . This is true only if we confine ἐναντίον to the meaning of ‘contradictorily opposite’ throughout, e.g. if we are always content merely to assert that the ἐναντίον of καλόν is μὴ καλόν : as soon as we say that its ἐναντίον is αἰσχρόν we have given to the thing two opposites (one of them multiform)—since μὴ καλόν is not αἰσχρόν but may be anything in the whole world except καλόν . This part of the argument is therefore also vitiated by neglecting the difference between contrary and contradictory terms.

24. ἀναλογισώμεθα : of reckoning up and reflecting as in I. 330E X. 618C.

30-1. ὑπὸ σωφροσύνης—ὑπὸ ἀφροσύνης . This has not been admitted in so many words, but with σωφροσύνῃ and ἀφροσύνῃ for ὑπὸ σωφροσύνης and ὑπὸ ἀφροσύνης : 332B l. II: cf. C, l. 16.

37. ἔμπροσθεν : 332A

41. τό does not belong to ἕν but to the clause ἓν ἑνὶ μόνον ἐναντίον εἶναι : cf. τὸ εἰ βούλει τοῦτο in 331C ἐκεῖνον refers to ch. XVIII.

46. οὐ πάνυ μουσικῶς—συναρμόττουσιν . ‘In his lenem agnosco Sophistae irrisionem, qui supra §43’ ( 326B ‘ πάντα τὸν βίον τοῦ ἀνθρώπου εὐρυθμίας τε καὶ εὐαρμοστίας δεῖσθαι dixisset’, Heindorf.

49. πλείω δὲ μή . B reads πλείοσιν, Τ πλείοσι : πλείω is a suggestion of Heindorf's, adopted by most editors. The nominatives ἓν μόνον in l. 48 and σοφία and σωφροσύνη in 50 are strongly in favour of πλείω . If πλείοσιν is retained, it must be regarded as a blemish in Plato's style.

56. τἀ λοιπά . If δικαιοσύνη = ὁσιότης and σωφροσύνη = σοφία it remains to identify either δικαιοσύνη or ὁσιότης with either σωφροσύνη or σοφία in order to prove the identity of these four virtues. Socrates begins to prove that δικαιοσύνη = σωφροσύνη .

57. ὅτι ἀδικεῖ : ὅτι (cf. below, 333D and 155E) is equivalent to κατὰ τοῦτο ὅ : cf. I. 340D ἐπεὶ αὐτίκα ἰατρὸν καλεῖς σὺ τὸν ἐξαμαπτάνοντα ρεπὶ τοὺς κάμνοντας κατ’ αὐτὸ τοῦτο ὃ ἐξαμαρτάνει; ὅ τιthe relative and ὅτι the conjunction shade into one another here. There is much to be said for Cron's view that the words should be written alike: see on Appendix II, p. 123.

59. πολλοί γέ φασιν . In I. 348Bff., Thrasymachus asserts that ἀδικία is εὐβουλία, ἀρετή, καλόν and ἰσχυρόν : compare also Polus in 469Bff. Notice that Protagoras' own opinion —that ἀδικία is not compatible with σωφροσύνη —makes for the identification of δικαιοσύνη and σωφροσύνη .

62. τὸν τῶν πολλῶν : on account of the τῶν here, Heindorf wished to insert οἱ before πολλοί in l. 59, but such a view could hardly be said to be held by the majority: cf. I. 348E εἰ γὰρ λυσιτελεῖν μὲν τὴν ἀδικίαν ἐτίθεσο, κακίαν μέντοι ἢ αἰσχπὸν αὐτὸ ὡμολόγεις εἶναι ὥσρεπ ἄλλοι τινές, εἴχομεν ἄν τι λέγειν κατὰ τὰ νομιζόμενα λέγοντες κτλ.As Sauppe points out, the article τῶν refers only to the above-mentioned πολλοί . With the situation cf. VII. 527Eff. σκόπει οὖν αὐτόθεν πρὸς ποτέρους διαλέγει· ἢ οὐ ρπὸς οὐδετέπους ἀλλὰ σαυτοῦ ἕνεκα τὸ μέγιστον ροιεῖ τοὺς λόγους, φθονοῖς μὴν οὐδ’ ἂν ἄλλω? εἴ τίς τι δύναιτο ἀπ’ αὐτῶν ὄνασθαι ;

63. εἴτ’ οὖν—εἴτε : see on 34E.

1. ἐκαλλωπίζετο . Cf. 236D παῦσαι πρός με καλλωπιζόμενος· σχεδὸν γὰρ ἔχω ὃ εἰπὼν ἀναγκάσω σε λέγειν; so also τρυφᾶν , cf. 11 E ἐπειδὴ δέ μοι δοκεῖς σὺ τρυφᾶν, αὐτός σοι ξυμπροθυμήσομαι , where see note.

5. ἔστω . Protagoras has already disclaimed this view in 333B

7. ὅτι ἀδικοῦσιν : Hirschig suggested ὅ τι ἀδικοῦσιν , but see on 333B Schanz rejects the words, but without sufficient reason.

8. εὖ πράττουσιν ἀδικοῦντες. εὖis here primarily of success, and κακῶς of failure, but the collocation of εὖ πράττειν with ἀδικεῖν is intended to suggest that the view which Protagoras is defending is a paradox. Plato frequently makes use of the double sense (as we should call it; but see on 3A) of εὖ πράττειν, εὖ ζῆν and the like, to imply that the virtuous life is the happy one: e.g. 116B ff.; I. 353Eff.; 172A, 173D; cf. 281C.

9. λέγεις οὖν ἀγαθά . Socrates proceeds to elucidate the definition of εὖ in εὖ πράττειν .

11. ὠφέλιμα τοῖς ἀνθρώποις : Xen. IV. 6. 8 ἄλλο δ’ ἄν τι φαίης ἀγαθὸν εἶναι ἢ τὸ ὠφέλιμον; οὐκ ἔγωγ’, ἔφη. τὸ ἄπα ὠφέλιμον ἀγαθόν ἐστιν ὅτῳ ἂν ὠφέλιμον ᾖ; δοκεῖ μοι, ἔφη . Plato, II. 379B τί δέ; ὠφέλιμον τὸ ἀγαθόν; ναί.

14. ἀγωνιᾶν καὶ παρατετάχθαι . The martial metaphor in the desiderative ἀγωνιᾶν is carried on by παρατετάχθαι , which is the MSS. reading, and means literally ‘to be set in array’, cf. Thuc. IV. 43. 3 οὗτοι γὰρ παρατεταγμένοι ἦσαν ἔσχατοι . For the metaphorical use cf. III. 399B ἐν πᾶσι τούτοις παρατεταγμένως καὶ καρτερούντως ἀμυνουμένου τὴν τύχην. The whole clause describes the outward demeanour of Protagoras, which causes Socrates to fear an imminent personal conflict. Kock's παρατετάσθαι , accepted by Schanz, does not suit, since παρατετάσθαι means ‘to be tired, worn out’, e.g. 303B καὶ γελῶντε καὶ κροτοῦντε . . . ὀλίγου παρετάθησαν, cf. Ar. 213. It is noteworthy that συντεταγμένως and συντεταμένως are frequently found as variants in Platonic MSS.

19. οὐδαμῶς, ἔφη . The argument is here broken off, and not (in its present form) again resumed. If Socrates had continued, Protagoras would doubtless have asserted that what is ὠφέλιμον for man is for him ἀγαθόν . The argument has therefore been: σωφρονεῖν , it is said, is ἀδικεῖν . But σωφρονεῖν = εὖ φρονεῖν = εὖ βουλεύεσθαι if (but only if) εὖ πράττειν (i.e. ὠφέλιμα πράττειν ) accompanies εὐβουλία . In other words ὠφέλιμα πράττειν = ἀδικεῖν is the stage which we have reached—a thesis which it is the object of the and of a large part of the to refute.

πολλὰ οἶδ’ ἅ . The speech which follows may be from some work by Protagoras (so recently Zeller in the Archiv für Geschichte der Philosophie, V, 2, 177), perhaps his περὶ ἀρετῶν , mentioned by Diog. Laert. IX. 8. 55 (since ἀρετή is a general word for excellence). Zeller points out that just as the notion ‘true’ was conceived by Protagoras (in his πάντων μέτρον ἄνθρωπος ) as relative, so here the notion ‘good’ is represented as relative and varying with that to which it is relative.

20. ἀνθρώποις μὲν—ὠφέλιμα : ἀνθρώποις μέν has no antithesis expressed: were the antitheses expressed in full, the sentence would be very cumbrous, running somewhat as follows: ἔγωγε πολλὰ οἶδ’ ἃ ἀνθρώποις μὲν <τὰ μὲν> ἀνωφελῆ ἐστί, καὶ σιτία—μυπία, <τοῖς δὲἄλλοις ζῴοις καὶ φυτοῖς ὠφέλιμα>, τὰ δέ γε ὠφέλιμα <τοῖς ἀνθρώποις, τοῖς δὲ ἄλλοις ἀνωφελῆ> . For μέν without a balancing clause see on τὸ μὲν καταγελασθῆναι in 3C, and on 330A above for τὰ δέ without a preceding τὰ μέν. ἀνωφελῆ is ‘hurtful’, as frequently: see above on οἷον μὴ ὅσιον in 331A With Protagoras' classification of ὠφέλιμα—ἀνωφελῆ—οὐδέτερα cf. 467E ἆρ’ οὖν ἔστιν τι τῶν ὄντων, ὃ οὐχὶ ἤτοι ἀγαθόν γ’ ἐστὶν ἢ κακὸν ἢ μεταξὺ τούτων οὔτε ἀγαθὸν οὔτε κακόν(where see Thompson's note) and below, 351D

23. οὐδενί : Schanz follows Naber in reading οὐδέσι against the MSS. both here and in 302C. The plural would be more appropriate here, but what is denied of all individuals is denied of the species, and the dative plural of οὐδείς seems to lack authority.

27. εἰ δ’ ἐθέλοις—πάντα ἀπόλλυσιν . Heindorf suggests εἰ δ’ ἐθέλεις —the indicative being generally used in this phrase, e.g. 122B εἰ δ’ αὖ ἐθέλεις εἰς πλούτους ἀποβλέψαι κτλ., αἰσχυνθείης ἂν ἐπὶ σαυτῷ —but Goodwin, p. 188, § 500, quotes a precisely similar example from Thuc. II. 39. 5 καίτοι εἰ π(αθυμίᾳ μᾶλλον ἢ ρόνων μελέτῃ—ἐθέλοιμεν κινδυνεύειν, ρεπιγίγνεται ἡμῖν κτλ. In both examples the present in the apodosis contains a reference to the future.

29. φυτοῖς—πάγκακον . Theophr. IV. 16. 5 πολέμιον γὰρ δὴ καὶ τοῦτο (sc. τὸ ἔλαιον ) πᾶσι· καὶ ἔλαιον ἐπιχέουσι τοῖς ὑρολείμμασι τῶν π(ιζῶν· ἰσχύει δὲ μᾶλλον τὸ ἔλαιον ἐν τοῖς νέοις καὶ ἄρτι φυομένοις· ἀσθενέστερα γάρ κτλ.

ταῖς θριξὶν—ἀνθρώπου . Arist. περὶ ζῴων γενέσεως V. 5. 785 a . 30 ff. says that a mixture of oil and water is a remedy against grey hairs.

33. ἐνταῦθα , i.e. ἐν τῷ ἀνθρώπῳ .

35. ταὐτόν . This form, not ταὐτό , of the neuter of ὁ αὐτός is almost regular in Plato: see on 314B

37. μέλλει , sc. ὁ ἀσθενῶν : see on 319D

38. ὅσον μόνον ‘just enough to’. Cf. III. 416E δέχεσθαι μισθὸν τῆς φυλακῆς τοσοῦτον, ὅσον μήτε ρεπιεῖναι αὐτοῖς. . .μήτε ἐνδεῖν, 161B ἐγὼ δὲ οὐδὲν ἐπίσταμαι πλέον πλὴν βραχέος, ὅσον λόγον παρ’ ἑτέρου σοφοῦ λαβεῖν καὶ ἀποδέξασθαι μετρίως .

τὴν δυσχέρειαν—ὄψοις . The nausea is of course that felt by a sick person at the smell of food: so far there is therefore no reason for holding ῥινῶν to be corrupt and reading χυμῶν (as Kroschel formerly read), still less εὐκρινῶν with Bergk. But it is not clear how oil could thus prevent nausea, nor does there seem to be any parallel among the ancients to such a statement. It is known that the ancients (like many modern peoples) used oil in cooking much as we use butter (see Blümner's p. 228), but the present passage seems rather to point to the use of some kind of fragrant oil sprinkled on the food after it was cooked. The Greeks were at all events not unfamiliar with the use of scents in banqueting: see Xen. II. 2 ff. Kroschel thinks Plato is making fun of Protagoras by putting into his mouth the word ‘ ῥινῶν pro πικρῶν vel χυμῶν ’, but this view will hardly command assent.

9. καὶ βραχυτέρας ποίει . Hirschig would reject the words, but they explain the metaphor in σύντεμνε : see on 314A κυβεύῃς τε καὶ κινδυνεύῃς .

13. ὅσα ἐμοὶ δοκεῖ . Cf. 461D τί δέ; οὐκ ἐξέσται μοι λέγειν ὁπόσα ἂν βούλωμαι ;

17. καὶ αὖ βραχέα . Sauppe cites 449C καὶ γὰρ αὖ καὶ τοῦτο ἕν ἐστιν ὧν φημί, μηδένα ἂν ἐν βπαχυτέποις ἐμοῦ τὰ αὐτὰ εἰρεῖν. Τούτου μὴν δεῖ, ὦ Γοπγία· καί μοι ἐρίδειξιν αὐτοῦ τούτου ροίησαι, τῆς βραχυλογίας, μακρολογίας δὲ εἰσαῦθις; cf. also 461Dand 267B.

22. ὡς ὁ ἀντιλέγων—οὕτω διελεγόμην . Asyndeton is frequent in such sentences: cf. II. 359B(cited by Heindorf) εἰ τοιόνδε ποιήσαιμεν τῇ διανοίᾳ, δόντες ἐξουσίαν—εἶτ’ ἐπακολουθήσαιμεν κτλ. See on 311E

24. ἐγένετο Πρωταγόρου ὄνομα . With ἐγένετο ὄνομα (for which Kroschel reads ἐλέγετο ὄνομα ) cf. 20D τί ποτ’ ἔστιν τοῦτο ὃ ἐμοὶ πεποίηκε τό τε ὄνομα καὶ τὴν διαβολήν: γίγνεσθαι is the usual passive to ποιεῖν . With the sentence generally cf. 4E-5A, where Euthyphro says οὐδὲν γὰρ ἄν μου ὄφελος εἴη, ὦ Σώκπατες, οὐδέ τῳ ἂν διαφέποι Εὐθύφπων τῶν ρολλῶν ἀνθπώρων, εἰ μὴ κτλ. : see note in loc. for more parallels.

25. οὐκ ἤρεσεν—καὶ—οὐκ ἐθελήσοι . The combination of indicative and optative in the same sentence in indirect speech is fairly common: Goodwin, p. 261, § 670.

28. ἡγησάμενος . Heindorf conjectures ἡγησάμενος οὖν , but οὖν may be dispensed with after a short parenthesis.

36. τὰ μακρὰ ταῦτα ἀδύνατος . 94B ἴνα δὲ μὴ ὀλίγους οἴη? καὶ τοὺς φαυλοτάτους Ἀθηναίων ἀδυνάτους γεγονέναι τοῦτο τὸ πρᾶγμα; VI. 478A ἐφ’ ἑτέρῳ ἄρα ἕτερόν τι δυναμένη ἑκατέρα αὐτῶν πέφυκεν.

38. ἵνα συνουσία ἐγίγνετο . For this construction see note on 44D εἰ γὰρ ὤφελον—οἷοί τ’ εἶναι—ἴνα οἷοί τ’ ἦσαν. A secondary tense of the indicative is found with ἴνα and ὅπως (without ἄν ) in final clauses dependent on a wish which can no longer be fulfilled, or on the apodosis to an impossible protasis: Goodwin, p. 120, § 333. The idiom is frequent in Plato and sometimes corrupted by scribes, e.g. 89B ἡμεῖς ἂν παραλαβόντες—ἐφυλάττομεν—ἵνα μηδεὶς αὐτοὺς διέφθειρεν, ἀλλ’ ἐπειδὴ ἀφίκοιντο εἰς τὴν ἡλικίαν, χρήσιμοι γίγνοιντο ταῖς πόλεσιν—where perhaps we should read ἀφίκοντο and ἐγίγνοντο . See Cobet's pp. 102, 359.

45. τρίβωνος . The ordinary clothing of an Athenian consisted of a χιτών , or shirt, with a ἱμάτιον or short mantle worn above it. The τρίβων was a short cloak of coarse material worn by the Spartans and sometimes by the Laconisers in other states to take the place of both χιτών and ἱμάτιον . It was the ordinary garment of Socrates ( 219B), and was afterwards much affected by the more ascetic philosophers, such as the Cynics (Diog. Laert. VI. I. 13). Its use was supposed to be a mark of sturdy simplicity and austerity of manners.

48. οὐδ’ ἂν ἑνός . The two parts of οὐδείς are sometimes separated by ἄν or a preposition, often with the effect (as here) of increasing the emphasis by making οὐδέ = ne—quidem, cf. below 343D οὐδὲ πρὸς ἔνα λόγον .

51. ὦ παῖ Ἱππονίκου : see on 328D above. With ἀεὶ μὲν— ἀτάρ Sauppe compares II. 367E καὶ ἐγὼ ἀκούσας ἀεὶ μὲν δὴ τὴν φύσιν τοῦ τε Γλαύκωνος καὶ τοῦ Ἁδειμάντου ἠγάμην, ἀτὰπ οὖν καὶ τότε πάνυ γε ἥσθην.

52. φιλοσοφίαν . The original meaning of the word φιλοσοφία was ‘love of knowledge for its own sake’, no particular kind of knowledge being specified. The verb φιλοσοφεῖν is first found in Hdt. I. 30 ὡς φιλοσοφέων γῆν πολλὴν θεωρίης εἵνεκεν ἐπελήλυθας . In the present passage (as in Thuc. II. 40 φιλοσοφοῦμεν ἄνευ μαλακίας ) the word retains its original wider meaning, viz, ‘love of knowledge’ in general; but side by side with this in Plato it has the more restricted meaning of ‘Philosophy’, e.g. VI. 496A πάνσμικρον δή τι. . .λείπεται τῶν κατ’ ἀξίαν ὁμιλούντων φιλοσοφίᾳ. The wider meaning is also found in Aristotle, where he calls Theology the ‘Highest Study’ ( πρώτη φιλοσοφία ), and elsewhere.

55. Κρίσωνι . Criso of Himera, a famous σταδιοδρόμος , won three times at Olympia, viz. Olymp. 83, 84 and 85 (448, 444 and 440 B.C.). His chastity during his prime as an athlete is mentioned in VIII. 840A. δρομεῖ ἀκμάζοντι is rejected by some editors, but δρομεῖ should be taken with Ἱμεραίῳ and ἀκμάζοντι with ἕπεσθαι : ‘to keep up with Criso, the runner of Himera, when he was in his prime’—presumably at the assumed date of this dialogue he was past his ἀκμή .

56. δολιχοδρόμων : δολιχοδρόμοι εἰσὶν οἱ τὸν δόλιχον τρέχοντες , says the scholiast: see on 329A

ἡμεροδρόμων . Heindorf quotes Livy, XXXI. 24. 4 ‘Hemerodromos vocant Graeci ingens die uno cursu emetientes spatium.’ Such was Phidippides; see Hdt. VI. 105 ἡμεροδρόμην τε καὶ τοῦτο μελετῶντα .

57. διαθεῖν τε καὶ ἕπεσθαι. δια‐in compounds frequently denotes competition: for διαθεῖν cf. 148C εἶτα διαθέων τοῦ ἀκμάζοντος καὶ ταχίστου ἡττήθης. As the emphasis is on ἕπεσθαι (‘keep up with’), διαθεῖν τε καὶ ἕπεσθαι is equivalent to διαθέοντα ἕπεσθαι : cf. note on 317C ἐνδείξασθαι καὶ καλλωπίσασθαι , where καὶ καλλωπίσασθαι is (so to speak) enclitic as διαθεῖν τε καί is proclitic (cf. the English ‘Some—must be talked over by the hour before they could reach the humblest decision, which they only left the office to return again (ten minutes later) and rescind ’).

61. συγκαθεῖναι : intransitive as in VIII. 563A οἱ δὲ γέροντες ξυγκαθιέντες τοῖς νέοις: so παρέχω for παρέχω ἐμαυτόν below, 348A ἕτοιμός εἰμί σοι παρέχειν ἀποκρινόμενος : cf. also ἐκεῖνοί τε ἀπέκρυψαν (sc. ἑαυτούς ) Thuc. V. 65. 5. The omission of the reflexive pronoun is a well-recognised way of making transitive verbs into intransitive.

66. χωρίς . Soph. 808 χωρὶς τό τ’ εἰπεῖν πολλὰ καὶ τὰ καίρια . ‘In talibus locis δίχα, χωρίς etc. non, ut vulgo, seiunctionem et separationem, sed diversitatem et oppositionem significant’ (Kroschel). The usage is frequent in Plato. δημηγορεῖν is contionari —platform oratory.

68. ὁρᾷς . Almost like our colloquial ‘don't you know’ (when used parenthetically and without interrogative force): the ἀλλά belongs in strict logic to δίκαια δοκεῖ λέγειν . Heindorf quotes parallels from Aristophanes, e.g. 330-1 οὐκἂν ὀρχησαίμεθ’, εἴπερ ὠφελήσαιμέν τί σε. ἀλλ’ ὁρᾶτ’, οὔπω πέπαυσθε . Where ὁρᾷς or ὁρᾶτε stand in this way as the first word of a sentence, editors generally regard the usage as interrogative, e.g. Eur. 581 ὁρᾷς; Ὀδυσσέως ἄλοχον οὐ κατέκτανε .

70. καὶ σύ . So the original hand in T: B has καί σοι . If we read καὶ σοί , the construction is ἀξιῶν αὑτῷ τε ἐξεῖναι διαλέγεσθαι ὅπως βούλεται καὶ σοὶ ἐξεῖναι κτλ. , but has nowhere asked that Socrates should be permitted to converse as he likes: quite the contrary. With σύ the construction is καὶ σὺ δίκαια δοκεῖς λέγειν ἀξιῶν διαλέγεσθαι ὅπως ἂν κτλ. , i.e. and your demand that Protagoras should converse as you wish likewise seems fair. Protagoras asked to be allowed to use his own style in 335A Socrates requested that Protagoras should converse as Socrates wished in 334D 335C 335E Alcibiades' defence of Socrates in the next chapter seems also to imply the reading σύ here. The only objection to this view lies in the position of τε after αὑτῷ : we should expect it to follow Πρωταγόρας. τε is however frequently misplaced (see above on 316D . σοί can only be retained if we either (1) take καὶ σοί as altogether independent of ἀξιῶν — which is barely possible, or (2) regard Plato as guilty of inaccurate writing.

5. λόγον τε δοῦναι καὶ δέξασθαι . We should expect τε to follow δοῦναι : see on 316D

10. ἐκκρούων . ‘Vox ducta a pugilatu, cuius proprium κρούειν de rep. IV. p. 422B οὐδ’ εἰ ἐξείη—ὑποφεύγοντι ( τῷ πύκτῃ ) τὸν πρότερον αἰεὶ προσφερόμενον ἀναστρέφοντα κρούειν etc.’ Heindorf. The simple verb is used metaphorically in 154E of beating arguments with arguments, ἤδη ἂν—συνελθόντες σοφιστικῶς εἰς μάχην τοιαύτην ἀλλήλων τοὺς λόγους τοῖς λόγοις ἐκρούομεν , and the compound in 228E ἐκκέκρουκάς με ἐλπίδος. Here the idea is of beating off, staving off by force, i.e. here by μακρηγορία : ἐκκρούειν δίκην is used by Demosthenes of staving off a trial by συκοφαντία and the like.

14. οὐχ ὅτι : ‘not but what’. The idiom (as if οὐ λέγω ὅτι , omitto quod ) is tolerably common in Plato, e.g. 450E οὐχ ὅτι τῷ ῥήματι οὕτως εἶπες, 157B τὸ δ’ εἶναι πανταχόθεν ἐξαιρετέον, οὐχ ὅτι ἡμεῖς—ἠναγκάσμεθα—χρῆσθαι αὐτῷ.

16. χρὴ γὰρ κτλ. : said apologetically.

20. φιλόνικος . The MSS. read φιλόνεικος and presently συμφιλονεικεῖν . It is however clear that the word comes from φιλο‐ and νίκη , not from φιλο‐ and νεῖκος (in which case the form would be φιλονεικής : compare φιλοκερδής, φιλοκυδής , but φιλότιμος, φιλόδοξος, φιλόθηρος and the like). Schanz has found only two traces of the original spelling with iota in Plato's MSS. (A, B, T): viz. in XI. 935B, where Paris A has ἀριστείων πέρι φιλονικήσῃ , and 122C, where φιλονικίαν appears as a correction for φιλονεικίαν in T. We might in consequence be tempted to suppose that Plato himself wrote φιλόνεικος through the influence of a false etymology, were it not that the derivation from νίκη alone suits the meaning, and that in more than one passage he shows himself conscious of the connection of the word with νίκη , notably in IX. 586C τί δέ; περὶ τὸ θυμοειδὲς οὐχ ἕτερα τοιαῦτα ἀνάγκη γίγνεσθαι, ὃς ἂν αὐτὸ τοῦτο διαπράττηται ἢ φθόνῳ διὰ φιλοτιμίαν ἢ βίᾳ διὰ φιλονικίαν( φιλονεικίαν Α ) ἢ θυμῷ διὰ δυσκολίαν, ρλησμονὴν τιμῆς τε καὶ νίκης καὶ θυμοῦ διώκων ἄνευ λογισμοῦ τε καὶ νοῦ ; Cf. also ibid. 581A-B. The orthography of this word is an old subject of dispute (see Stallbaum on VIII. 545A); Schanz (Preface to vol. VI, p. X) declares himself, after a full discussion, for φιλόνικος .

24. ὁ Πρόδικος—ἔφη . Prodicus contrives to make his remarks an ἐπίδειξις on ὀρθότης ὀνομάτων , which was his leading subject of instruction: see 277Eand cf. above on 314C The distinctions drawn by Prodicus are on the whole sound if somewhat wiredrawn and pedantic. The carefully balanced style of the speech finds a parallel in the story of Heracles in Xen. II. I. 21 ff.: cf. especially §§ 31-3.

26. κοινοὺς—ἀκροατάς—ἴσους. κοινόςand ἴσος are found as epithets of ἀκροατής in the orators: e.g. Dem. 7 τὰ τοῦ λέγοντος ὑστέπου δίκαια εὐνοϊκῶς ρποσδέξεται καὶ ραπασχὼν ἑαυτὸν ἴσον καὶ κοινὸν ἀμφοτέποις ἀκποατὴν οὕτω τὴν διάγνωσιν ποιήσεται περὶ πάντων , and Andoc. in Alcib. § 7 δέομαι δ’ ὑμῶν, τῶν λόγων ἴσους καὶ κοινοὺς ἡμῖν ἐπιστάτας γενέσθαι (Heindorf).

29. τῷ μὲν σοφωτέρῳ πλέον κτλ. In other words ἰσότης γεωμετρική and not ἰσότης ἀριθμητική should be observed by the audience ( 508A): the regard paid to the speakers should be in proportion to their merit.

33. ἀμφισβητεῖν—ἐρίζειν . Cicero's translation of this sentence is preserved by Priscian (Nobbe's p. 1313): ‘Nunc a vobis, a Protagora et Socrate ( leg. o Protagora et Socrates), postulo, ut de isto concedatis alter alteri, et inter vos de huiuscemodi rebus controversemini, non concertetis.’

38. εὐδοκιμοῖτε καὶ οὐκ ἐπαινοῖσθε . Heindorf suggests that ἀλλ’ οὐκ should be written for καὶ οὐκ , or καί omitted; but καί is occasionally used in this way.

39. ἄνευ ἀπάτης is opposed to παρὰ δόξαν ψευδομένων : εὐδοκιμεῖν implies that he with whom one εὐδοκιμεῖ is sincere. ἐν λόγῳ as opposed to παρὰ ταῖς ψυχαῖς suggests Shak. ‘Mouthhonour, breath, which the poor heart would fain deny, but dare not’.

42. εὐφραίνεσθαι is connected by Prodicus with φρόνησις ( φρονήσεως μεταλαμβάνοντα ): for a fantastic derivation of the kindred εὐφροσύνη see 419D παντὶ γὰρ δῆλον ὅτι ἀπὸ τοῦ εὖ τοῖς ρπάγμασι τὴν ψυχὴν ξυμφέπεσθαι τοῦτο ἔλαβε τὸ ὄνομα, εὐφεροσύνην, τό γε δίκαιον· ὅμως δὲ αὐτὸ καλοῦμεν εὐφροσύνην. Cf. Ar. II. 6, p. 112 b . 22 Πρόδικος διῃρεῖτο τὰς ἡδονὰς εἰς χαρὰν καὶ τέρψιν καὶ εὐφροσύνην . The Greek usage of this word hardly conformed to the rule laid down by Prodicus.

44. αὐτῇ τῇ διανοίᾳ : αὐτῆ? is ‘by itself’, i.e. without the body, as αὐτῷ in αὐτῷ τῷ σώματι is without the mind. ἡδύ is introduced to give the derivation of ἥδεσθαι .

8. ἀλλὰ δή like at enim introduces a possible objection: see on 54A. The fuller form of the comparative ( βελτίονα and the like) is less common in Plato than the contracted.

10. ὥστε . Heindorf cites ( inter alia ) 269D τὸ μὲν δύνασθαι, ὦ Φαῖδρε, ὥστε ἀγωνιστὴν τέλεον γενέσθαι. The insertion of ὥστε with the usual infinitive after ἀδύνατον increases the emphasis of ἀδύνατον by making it appear more than a mere auxiliary notion. Badham's conjecture ἀδύνατον ὂν ὑμῖν ἴστε deserves mention for its ingenuity. For Πρωταγόρου see above, note on 318D

27. πάνυ μὲν οὐκ ἤθελεν : οὐκ ἤθελεν is virtually one word: ‘was very reluctant’. The phrase is quite different from οὐ πάνυ : see on οὐ πάνυ τι in 321B above.

8. νῦν διαλεγόμεθα : so B and T: the editors mostly read νῦν δὴ ( νυνδὴ ) διελεγόμεθα. νῦν does not mean ‘at this present moment’, but simply ‘now’, ‘on the present occasion’, = ἐν τῇ νῦν συνουσίᾳ : translate ‘about the same subject as you and I are now discussing’, i.e. about the subject of our present discussion. A discussion which has never been finished (see on 334A and is to be resumed (cf. 338E ἐπειδὰν ἱκανῶς ἐρωτήσῃ, πάλιν δώσειν λόγον , and 338D ought not to be spoken of as past. We have in fact in νῦν διαλεγόμεθα an indication that the subject of the dialogue is the same throughout: see Introduction pp. xvii ff. νῦν in καὶ δὴ καὶ νῦν (l. 6) is simply ‘in the present case’ and introduces the application of the general statement contained in ἡγοῦμαι—λόγον δοῦναι : cf. 17D-18A ὥσπερ οὖν ἂν—καὶ δὴ καὶ νῦν .

περὶ ἀρετῆς . According to Diog. Laert. IX. 8. 55, Protagoras wrote a book having the title περὶ ἀρετῶν .

10. Σιμωνίδης . It is a saying of Simonides of Ceos (ca. 556-468 B.C.) that forms the text on which the discussion in the first book of the is based: see I. 331Dff. Plato seems also to allude to him in two other places, viz. II. 365C ἐπειδὴ τὸ δοκεῖν, ὡς δηλοῦσί μοι οἱ σοφοί, καὶ τὰν ἀλάθειαν βιᾶται καὶ κύριον εὐδαιμονίας κτλ.and VI. 489B οὐ γὰρ ἔχει φύσιν— τοὺς σοφοὺς ἐπὶ τὰς πλουσίων θύρας ἰέναι ἀλλ’ ὁ τοῦτο κομψευσάμενος ἐψεύσατο(compare Ar. II. 16. 1391 a . 8 ff. with Cope's note).

Σκόπαν . The Scopadae were a ruling family at Crannon and Pharsalus in Thessaly. Simonides seems to have frequently been their guest, and wrote poems in their honour: the most famous is that referred to by Cicero, de Or. II. § 352-3.

12. ἀγαθόν is here more than morally good: it includes bodily and external as well as internal well-being: whence χερσίν τε καὶ ποσί as well as νόῳ : see also note on 344B l. 4 below. The notion of external well-being belonged to the word from very early times: see Grote, III, 45, n. 3: ‘good’ and ‘bad’ are applied in Theognis and Solon ‘to wealth as contrasted with poverty— nobility with low birth—strength with weakness—conservative and oligarchical politics as opposed to innovation’. This sense survived in classical times in the political meaning of καλὸς κἀγαθός , e.g. Xen. II. 3. 12, Pl. VIII. 569A.

13. τετράγωνον . Simonides avails himself of a Pythagorean notion: among the Pythagoreans the number 4 was sacred, as being the first square number: see Ritter and Preller 7 , § 54. The expression τετράγωνος ἀνήρ became afterwards almost proverbial for a perfect man: Sauppe refers to Ar. III. 11. 1411 b . 27 ολ(=ον τὸν ἀγαθὸν ἄνδρα φάναι εἶναι τετράγωνον· ἄμφω γὰρ τέλεια .

24. ἐμμελέως—νέμεται. ἐμμελέωςbelongs to εἰρημένον , and νέμεται is poetic for νομίζεται : cf. (with Sauppe) Soph. 1080 ἐγὼ δ’ ἐμαυτὸν παῖδα τῆς τύχης νέμων . The story is (schol. to Plato, 304E quoted by Sauppe) that Pittacus, when ruler of Mitylene, on hearing of Periander's rapid conversion into a tyrant, sat down at an altar and begged to be released of his rule, assigning as his reason ὡς χαλεπὸν ἐσθλὸν ἔμμεναι . The scholiast adds that Solon when he heard the remark capped it with χαλεπὰ τὰ καλά . Pittacus is mentioned side by side with Bias and Simonides as one of the σοφοὶ καὶ μακάριοι ἄνδρες in I. 335E.

29-31. ὁμολογεῖσθαι—ὁμολογεῖν . The middle is said of things, and the active of persons, according to the general rule.

34. ὀλίγον δὲ τοῦ ποιήματος εἰς τὸ πρόσθεν : these words as well as προϊόντος τοῦ ᾄσματος in 339C are in favour of the supposition that some verses are wanting between ἄνευ ψόγου τετυγμένον and οὐδέ μοι ἐμμελέως : see Appendix I, pp. 213-14.

40. τὸ πρότερον : adverbial.

41. θόρυβον παρέσχεν καὶ ἔπαινον. θόρυβοςdenotes only ‘noise’, ‘tumult’ (cf. VI. 492C θόρυβον—τοῦ ψόγου καὶ ἐπαίνου): καὶ ἔπαινον is needed to show that the noise was favourable. So above, 334C ἀνεθορύβησαν ὡς εὖ λέγοι . With παρέχειν in this sense cf. the phrase παρέχειν γέλωτα in 473E and 174C.

43. ὡσπερεὶ ὑπὸ ἀγαθοῦ πύκτου πληγείς . For the metaphor cf. 303A ἐγὼ μὲν οὖν—ὥσπερ πληγεὶς ὑπὸ τοῦ λόγου ἐκείμην ἄφωνος. Socrates describes the effect of Protagoras' questions on himself in words which recall the descriptions of his own dialectic in its effect on others: cf. e.g. 80 A ff.

50. παρακαλεῖν is future: cf. 183D ἀλλά μοι δοκῶ— οὐ πείσεσθαι αὐτῷ, 228C δοκεῖς σὺ οὐδαμῶς με ἀφήσειν: tr. ‘therefore I think will call you to my assistance’. ἐγώ and σέ are contrasted in view of the illustration which is about to follow.

ὥσπερ ἔφη κτλ. : ὥσπερ corresponds to καὶ ἐγώ in l. 55: as, according to Homer, Scamander called on Simois, so look you, I call upon you. For ὥσπερ used in this way see above, note on 330A The other editors take δοκῶ οὖν ἐγὼ παρακαλεῖν σέ with the ὥσπερ clause, and, regarding παρακαλεῖν as a present, print a full stop after σχῶμεν in the quotation; but (1) there is a certain awkwardness in the repetition ‘I think I am summoning you’ and ‘so look you, I am summoning you’; (2) the quotation does not finish with σχῶμεν , but ἐκπέρσῃ in l. 56 belongs to it also—a point which is against separating φίλε κασίγνητε—σχῶμεν from the following clause. In the view which we have taken a fresh start begins with ὥσπερ , after which the actual summons follows in the present ἀτὰρ καὶ ἐγὼ σὲ παρακαλῶ .

51. ἔφη Ὁμηρος . XXI. 305 ff. οὐδὲ Σκάμανδρος ἔληγε τὸ ὃν μένος ἀλλ’ ἔτι μᾶλλον χώετο Ρηλείωνι—Σιμόεντι δὲ κέκλετ’ ἀύσας· Φίλε κασίγνητε, σθένος ἀνέπος ἀμφότεποί ρεπ σχῶμεν, ἐρεὶ τάχα ἄστυ μέγα Πριάμοιο ἄνακτος ἐκπέρσει. This ἐκπέρσει explains the occurrence of ἐκπέρσῃ in l. 56, although (as Heindorf shows) the same metaphor is occasionally found in tragedy (but hardly in prose), e.g. 1104 τυφλῆς ὑπ’ ἄτης ἐκπεπόρθημαι τάλας .

55. ἀτάρ : cf. above 335D

57. μουσικῆς : ‘culture’, as often; here not without some sarcasm, in reference to Prodicus' ὀνομάτων ὀρθότης , exemplified in 337A ff.

58. βούλεσθαι—ἐπιθυμεῖν. βούλεσθαιis of will, ἐπιθυμεῖν of desire. The distinction is generally well marked in Plato: see note on 25Cand cf. Cope on Ar. II. 19. 9. Prodicus does not touch on this example in his speech in 337, but it is quite in Plato's way to select a fresh example (cf. 147A-B, 166E by the side of 159C, 169A-B), which may in this case be supposed to come from Prodicus' lectures (cf. 341A . We should expect the article to be repeated with ἐπιθυμεῖν , as the two words are to be distinguished (cf. below in l. 63 τὸ γενέσθαι καὶ τὸ εἶναι ); but the article is sometimes dropped with the second of two words even when the words are contrasted, e.g. 9C τὸ ὅσιον καὶ μή. Here the effect of its omission is perhaps to suggest that the two notions are after all more like than different.

63. γενέσθαι—εἶναι . The distinction though long ago recognised by the philosophers was not always present in ordinary speech: otherwise (as Kroschel points out) Protagoras' censure of Simonides would be too absurd, and Socrates' pretended bewilderment out of place. But that Simonides in this poem drew a distinction between γενέσθαι and εἶναι is certain: whether it was the same distinction as Socrates himself draws later is another question; see on γενόμενον δὲ—ἀδύνατον in 344B

69. τὸ αὐτόν . B and T here have τὸ αὐτόν , which Schanz retains: the form occurs on inscriptions and once or twice in Plato's MSS.: v. Schanz, XII, vii.

70. ἔλεγεν, τὸ χαλεπὸν γενέσθαι . So Heindorf. ἔλεγεν is ‘said’, not ‘meant’, and the sentence is intended to prove what is stated in the last sentence, viz. that Pittacus οὺ τὸ αὐτὸν ἑαυτῷ ἔλεγεν (said), ἀλλ’ ἄλλο. τό goes with the whole clause χαλεπὸν γενέσθαι ἐσθλόν ; the emphasis is on γενέσθαι : for which reason τὸ ἔμμεναι in the next line (for τὸ ἔμμεναι ἐσθλὸν χαλεπόν ) suffices. It is unnecessary to read (with Schanz and Kroschel) ἔλεγε ( ν ) χαλεπόν, τὸ γενέσθαι ἐσθλόν , or ἔλεγε χαλεπόν, γενέσθαι ἐσθλόν with Sauppe.

75. Πρόδικος ὅδε καὶ ἄλλοι πολλοί . Socrates dissociates himself from the others, because he is about to give a different solution of Simonides' apparently contradictory statements (in 344B ff.).

76. Ἡσίοδον : in 289 ff. τῆς δ’ ἀρετῆς ἱδπῶτα θεοὶ ρποράποιθεν ἔθηκαν ἀθάνατοι· μακπὸς δὲ καὶ ὄπθιος οἶμος ἐς αὐτὴν καὶ τπηχὺς τὸ ρπῶτον· ἐρὴν δ’ εἰς ἄκπον ἵκηται, ῥηιδίη δἤπειτα πέλει, χαλεπή περ ἐοῦσα . This passage was very famous in antiquity: Plato cites it again in II. 364C, IV. 718E. Hesiod in effect says it is difficult to become good, but easy to be good. Simonides himself refers to Hesiod's verses in Frag. 58 (Bergk) ἔστι τις λόγος τὰν ἀρετὰν ναίειν δυσαμβάτοις ἐπὶ ρέτπαις, νῦν δέ μιν θεῶν χῶπον ἁγνὸν ἀμφέρειν. . .οὐδ’ ἀραντᾶν βλεφάποις θνατῶν ἔσορτον, ᾧ μὴ δακέθυμος ἱδπὼς ἔνδοθεν μόλῃ θ’ ἵκηταί τ’ ἐς ἄκρον ἀνδρείας .

80. ἐκτῆσθαι : to be taken with ῥηϊδίην . B and T read κτῆσθαι , but cf. 340E For the form see above on 319A

6. ἰώμενος μεῖζον τὸ νόσημα ποιῶ . Socrates is thinking of the proverb κακὸν κακῷ ἰᾶσθαι .

8. ποιητοῦ—ἐκτῆσθαι . The poet is not of course Hesiod but Simonides. Socrates nowhere said that in censuring Pittacus Simonides implies that it is easy to ‘keep virtue’, i.e. to be virtuous: the quotation from Hesiod was put in the mouth of Prodicus and ἄλλοι πολλοί ( 340C . At the same time Protagoras might fairly turn Socrates' fallacies (see on 331A against himself and say that if Simonides said it was ‘not difficult’ to be virtuous, he meant that it was ‘easy’ to be so.

12. κινδυνεύει γάρ τοι : the art of Prodicus (Socrates means) can lay as good claims to antiquity as yours: cf. 316D

13. θεία τις εἶναι πάλαι . Kroschel's reading εἶναι καὶ παλαιά would somewhat change the meaning, which is ‘has long been an art divine’: divine because practised by poets and the like, cf. I. 331E ἀλλὰ μέντοι—Σιμωνίδῃ γε οὐ ῥᾴδιον ἀπιστεῖν· σοφὸς γὰρ καὶ θεῖος ἁνήρ: above 315E and note. In ἤτοι ἀπὸ Σιμωνίδου there is an allusion to 316D

16. οὐχ ὥσπερ ἐγώ , sc. εἰμί . The idiom is frequent in Plato, e.g. 179E οὐχ ὥσπερ Ἀχιλλέα—ἐτίμησαν καὶ εἰς μακάρων νήσους ἀπέπεμψαν.

μαθητής . Socrates calls himself a disciple of Prodicus also in 384B, 96D, 163D.

18. τὸ χαλεπὸν τοῦτο. τόmarks χαλεπόν as a quotation: see above on 331C

20. ἀλλ’ ὥσπερ κτλ. The application of the illustration follows in 341B ll. 27 ff. ἴσως οὖν καὶ τὸ χαλεπὸν—ὑπολαμβάνουσιν , where see note.

22. Πρωταγόρας σοφὸς καὶ δεινός ἐστιν . The ἢ ἄλλον τινά is neglected, and Πρωταγόρας takes us back to σέ : cf. note on ἐνδείξασθαι καὶ καλλωπίσασθαι in 317C The collocation σοφὸς καὶ δεινός is tolerably frequent in ironical characterisations, e.g. 173B δεινοί τε καὶ σοφοὶ γεγονότες, ὡς οἴονται.

23. ἐρωτᾷ explains νουθετεῖ , whence the asyndeton explicativum : see on 335A above.

24. τὸ γὰρ δεινὸν—κακόν ἐστιν . Prodicus' canon—which rests on the derivation of δεινόν from δέος —is not borne out by Greek usage, except to this extent, that when a man is called δεινός , it is generally implied that he is more clever than good.

25. δεινοῦ πλούτου κτλ. Genitives of exclamation in the Platonic dialogues are generally (as Turner remarks) preceded by an interjection, e.g. 303A πυππὰξ ὦ Ἡράκλεις καλοῦ λόγουand ibid. ὦ Πόσειδον δεινῶν λόγων . Here of course the exclamation is left out as irrelevant: the only relevant point is the use of δεινός .

27. ἴσως οὖν καὶ τὸ χαλεπόν . Sauppe remarks that we should expect οὕτω καὶ τὸ χαλεπόν to introduce the apodosis to the ὥσπερ clause ( 341A l. 20); καί is however enough to show that we have reached the application; οὖν is introduced on account of the parenthesis from τὸ γὰρ δεινόν to κακοῦ ὄντος ; and ἴσως marks the suggestion as only tentative.

31. φωνήν : ‘dialect’ as often, e.g. 62A καὶ ὁ Κέβης—ἴττω Ζεύς, ἔφη, τῇ αὑτοῦ φωνῇ εἰπών.

32. κακόν, ἔφη . Prodicus enters into the spirit of the joke: in view of C and D it would be absurd to take this seriously: see note on ἀλλὰ παίζειν in D below.

37. τὰ ὀνόματα—ὀρθῶς διαιρεῖν : cf. above, 340A and below, 358A τὴν δὲ Προδίκου τοῦδε διαίρεσιν τῶν ὀνομάτων παραιτοῦμαι . Prodicus pretends to regard Simonides (cf. 341 A) as a teacher like himself of ὀνομάτων διαίρεσις .

38. ἅτε Λέσβιος ὤν : had he been Κεῖος , he would have learnt ὀνομάτων διαίρεσις forsooth in its natural home.

39. ἐν φωνῇ βαρβάρῳ : a malicious exaggeration inspired by the odium philologicum.

47. ἀλλὰ παίζειν . The editors suppose that Socrates is here turning the tables on Prodicus, who it is supposed meant his criticism seriously, but the tone of the passage seems to imply that Prodicus is in league with Socrates to make fun of Protagoras, who is represented throughout the whole dialogue as lacking all sense of humour. It would not be wit, but sheer buffoonery, in Plato to represent Prodicus as seriously believing that Simonides had censured Pittacus for having said: ‘It is bad to be good.’

48. καὶ σοῦ δοκεῖν ἀποπειρᾶσθαι. δοκεῖνis not pleonastic after οἶμαι but means ‘think fit’—a very idiomatic use, cf. Aesch. 16 ὅταν δ’ ἀείδειν ἢ μινύρεσθαι δοκῶ . See III, 148, where Mr Arthur Sidgwick discusses and illustrates this usage.

53. οὐ δήπου τοῦτό γε λέγων. οὐ δήπουgoes with λέγων , which is ‘meaning’ not ‘saying’, and τοῦτο is explained by κακὸν ἐσθλὸν ἔμμεναι . For the asyndeton see on 335A and for the use of εἶτα cf. 311A and 200A πότερον ἔχων αὐτὸ οὗ ἐπιθυμεῖ τε καὶ ἐρᾷ, εἶτα ἐπιθυμεῖ τε καὶ ἐρᾷ, ἢ οὐκ ἔχων ;

54. τοῦτο γέρας . Heindorf reads τοῦτο τὸ γέρας as in 344C but γέρας may be regarded as predicative and going closely with ἀπένειμε , so that τοῦτο—τοῦτο balance each other. Sauppe compares 179C εὐαριθμήτοις δή τισιν ἔδοσαν τοῦτο γέρας οἱ θεοί.

55. ἀκόλαστον—οὐδαμῶς Κεῖον . This seems to be the earliest passage making allusion to the sobriety and uprightness of the Ceans, to which, perhaps, Aristophanes sarcastically alludes in 970 ( οὐ Χῖος, ἀλλὰ Κεῖος ). In I. 638B(quoted by Sauppe) Plato cites the subjugation of the Ceans by the Athenians as a proof that victory does not always favour the more virtuous side. Strabo (X. 486) quotes from Menander the lines καλὸν τὸ Κείων νόμιμόν ἐστι, Φανία· ὁ μὴ δυνάμενος ζῆν καλῶς οὐ ζῇ κακῶς and explains them by saying that a Cean law required those above the age of 60 to take hemlock so as to make their country's produce suffice to feed the others.

59. ὃ σὺ λέγεις τοῦτο refers to περὶ ἐπῶν 338E For the idiom cf. the usual τὸ σὸν δὴ τοῦτο , e.g. 221B.

5. ἐξαρνοῦνται καὶ σχηματίζονται. σχηματίζεσθαιof ‘posing’ as in 268A ἀγνοεῖ ταῦτα ἃ πρὸς τοὺς ἄλλους ὡς εἰδὼς ἐσχημάτισται.

13. ἱμάντας περιειλίττονται . The ἱμάντες were thongs of leather bound round the knuckles for greater efficacy in boxing: Hom. XXIII. 685. The caestus, being loaded with balls of lead, was a much more brutal instrument (Virg. v. 404-5).

14. βραχείας ἀναβολάς : ‘short cloaks’ in imitation of the τρίβων (the national Spartan dress: see on 335D . ἀναβολή (here almost concrete) and ἀναβάλλεσθαι were said of the ἱμάτιον , to wear which rightly and like a gentleman was ἐπὶ δεξιὰ ἀναβάλλεσθαι ( 175 E), not ὲπ’ ἀριστερά ( Ar. 1567-8, a passage which seems decisive against reading ἐπιδέξια in this phrase). From Suidas (s.v. ἀναβάλλει—ἀναβάλλεσθαι δὲ τὸ ἱμάτιον, οὐ περιβάλλεσθαι λέγουσιν ) we may infer that ἀναβολή refers not to the throwing back of the ἱμάτιον over the shoulder (since in point of fact it was thrown back over the left shoulder) but to pulling it round the back (from left to right) before throwing the end over the left shoulder in front.

15. ὡσδὴ—κρατοῦντας is ‘quasi vero—his superent’ (Kroschel). For the construction cf. (with Kroschel) I. 345E οὐκ ἐννοεῖς ὅτι οὐδεὶς ἐθέλει ἄπχειν ἑκών, ἀλλὰ μισθὸν αἰτοῦσιν, ὡς οὐχὶ αὐτοῖσιν ὠφέλειαν ἐσομένην ἐκ τοῦ ἄρχειν ἀλλὰ τοῖς ἀρχομένοις ;Madvig's p. 168.

19. ξενηλασίας . Heindorf quotes ( inter alia ) Ar. 1012 ὥσπερ ἐν Λακεδαίμονι ξενηλατοῦνταιand Plut. 27. 6 τοὺς ἀθροιζομένους ἐπ’ οὐδενὶ χρησίμῳ καὶ παρεισρέοντας εἰς τὴν πόλιν ἀπήλαυνεν (sc. Λυκοῦργος ), οὐχ, ὡς Θουκυδίδης (II. 39) φησί, δεδιὼς μὴ τῆς ρολιτείας μιμηταὶ γένωνται καὶ ρπὸς ἀπετήν τι χπήσιμον ἐκμάθωσιν, ἀλλὰ μᾶλλον ὄρως μὴ διδάσκαλοι κακοῦ τινος ὑπάρξωσιν . The reasons assigned by Plutarch are no doubt the true ones.

22. οὐδένα ἐῶσιν—ἐξιέναι . Plut. 27. 5 οὐδ’ ἀποδημεῖν ἔδωκε ( Λυκοῦργος ) τοῖς βουλομένοις καὶ πλανᾶσθαι ξενικὰ συνάγοντας ἤθη καὶ μιμήματα βίων ἀπαιδεύτων καὶ πολιτευμάτων διαφοράν, ἀλλὰ καὶ τοὺς ἀθροιζομένους etc. (see last note). In XII. 950Cff. (quoted by Sauppe) Plato lays down similar though less stringent regulations as to ἀποδημία .

23. ὥσπερ οὐδὲ Κρῆτες . There seems to be no other authority for this statement, but (as Sauppe remarks) the resemblance between the Cretan and Spartan institutions is well known.

26. ἀλλὰ καὶ γυναῖκες . Women in Sparta held a position of much greater power and influence than in the rest of Greece, partly at least in consequence of their superior education, physical and otherwise: cf. Ar. II. 9. 1269 b . 32 πολλὰ διῳκεῖτο ὑπὸ τῶν γυναικῶν ἐπὶ τῆς ἀρχῆς αὐτῶν (sc. τῶν Λακεδαιμονίων ) and see Grote, 11, 383 ff. Heindorf remarks that wise and pregnant sayings by Spartan women (such as are given in pseudo-Plutarch's Λακαινῶν ἀποφθέγματα , e.g. the famous τέκνον, ἢ τὰν ἢ ἐπὶ τᾶς ) were probably already current in Plato's time.

28. εἰ γὰρ ἐθέλει—εὑρήσει . Cf. 324A

32. ἐνέβαλεν ῥῆμα—συνεστραμμένον . The aorist is like ‘behold! he has thrown in’: it expresses rapidity by representing the action as no sooner begun than over. The idiom is very frequent in Plato: Turner refers to Stallbaum on III. 406D ἐὰν δέ τις αὐτῷ μικρὰν δίαιταν προστάττῃ—ταχὺ εἶπεν ὅτι οὐ σχολὴ κάμνειν . With συνεστ ραμμένον cf. Ar. II. 24. 1401 a . 5 συνεστ ραμμένως—εἰπεῖν ; the metaphor is apparently from an animal gathering itself for a spring (cf. Ar. IX. 48. 631 a . 27 συστρέψαντες ἑαυτοὺς φέρονται ὥσπερ τόξευμα and Plato, I. 336B συστρέψας ἑαυτὸν ὥσπερ θηρίον ἧκεν ἐφ’ ἡμᾶς).

39. Θαλῆς ὁ Μιλήσιος κτλ. This list of the seven wise men is remarkable as excluding Periander, who was canonised later (Diog. Laert. I. 13). Plato refused to allow that a tyrant could be truly σοφός ( IX. 587D) or even (in the true sense of the term) powerful: see I. 336A οἶμαι αὐτὸ(sc. the view that justice is doing good to friends and evil to enemies) Περιάνδρου εἶναι ἢ Ρεπδίκκου ἢ Ξέπξου ἢ Ἰσμηνίου τοῦ Θηβαίου ἤ τινος ἄλλου μέγα οἰομένου δύνασθαι πλουσίου ἀνδρός .

(the least known of the seven) figures as early as Hipponax (Frag. 45, quoted by Sauppe) καὶ Μύσων ὃν Ὡπόλλων ἀνεῖπεν ἀνδρῶν σωφρονέστατον πάντων . According to a tradition preserved in Diog. Laert. I. 106 the Pythian priestess, being asked by Anacharsis to say if there was any man wiser than himself, replied Οἰταῖόν τινά φημι Μύσων’ ἐνὶ Χηνὶ γενέσθαι σοῦ μᾶλλον πραπίδεσσιν ἀρηρότα πευκαλίμη?σι : but another account placed his birthplace in Chen in Laconia, and a third (reading Ἠτεῖόν τινά φημι for Οἰταῖόν τινά φημι in the oracle) in Etea, which was variously placed in Laconia and in Crete. Sauppe remarks that the presence of Λακεδαιμόνιος with Χίλων seems to show that Plato did not regard Laconia as Myson's birthplace, but favoured the view which made him a native of Chenae by Mount Oeta. ( fl. ca. 585 B.C., the eclipse of which year he is said to have predicted) is mentioned in two other passages in Plato (in neither of which is his philosophical teaching—that the ἀρχή is ὕδωρ —referred to), once as an author of useful inventions ( X. 600A), and once as the hero of an anecdote illustrating the philosopher's want of worldly wisdom ( 174A). ( fl. ca. 612 B.C.) and of Priene in Ionia (contemporary with or earlier than Hipponax, who refers to him in Diog. Laert. I. 84) are mentioned together again in I. 335E Βίαντα ἢ Πιττακὸν—ἤ τιν’ ἄλλον τῶν σοφῶν τε καὶ μακαρίων ἀνδρῶν. of Lindus in Rhodes and of Sparta (both about the beginning of the 6th cent. B.C.) are not again referred to by Plato.

The traditions relating to the wise men and many of the aphorisms with which they are credited are given in Diog. Laert. I. 22-122: for the authorities for their lives, and for their sayings, see Mullach's II, 203-34. This passage of the apparently the earliest in which seven are named together, probably contributed in large measure to the canonisation of the wise men.

42. Λακεδαιμόνιος : Heindorf would read ὁ Λακεδαιμόνιος , but Plato may well have said ‘a Lacedaemonian, Chilon’.

46. εἰρημένα· οὗτοι . The sentence beginning with οὗτοι shows how one is to learn αὐτῶν τὴν σοφίαν τοιαύτην οὖσαν ; the asyndeton (as Heindorf observes) resembles that after σημεῖον δέ, τεκμήριον δέ and the like. Here οὗτοι καὶ κτλ. is so far removed from καὶ καταμάθοι—οὖσαν that we might have expected οὗτοι γὰρ καί or (as Kroschel reads) ὅτι for οὗτοι , but the emphatic οὗτοι (parallel to οὗτοι in l. 43 above) renders the explanatory particle unnecessary. Hermann's correction εἰρημένα ἅ for εἰρημένα —adopted by Sauppe—gives a wrong meaning; for ἀπαρχή ‘first-fruits’ (l. 47) coming after καὶ κοινῇ ξυνελθόντες , in marked antithesis to ἑκάστῳ εἰρημένα , cannot mean merely the sayings of each individual—as it will have to mean if ἅ is read, ἅ being then in apposition to ἀπαρχήν . Kral's ῥήματα βραχέα ἀξιομνημόνευτα <σκοπῶν> ἑκάστῳ εἰρημένα, ἃ οὗτοι κτλ. suffers from the same fault, besides that it is very unlikely that σκοπῶν should have fallen out.

47. κοινῇ ξυνελθόντες—ἀνέθεσαν . The editors cite Pausanias, X. 24. I ἐν δὲ τῷ προνάῳ τῷ ἐν Δελφοῖς γεγραμμένα ἐστὶν ὠφελήματα ἀνθπώροις εἰς βίον—οὗτοι οὖν οἱ ἄνδπες ἀφικόμενοι ἐς Δελφοὺς ἀνέθεσαν τῷ Ἀρόλλωνι τὰ ἀ?δόμενα Γνῶθι σαυτὸν καὶ Μηδὲν ἄγαν . The same explanation of the presence of these maxims on the temple at Delphi meets us in other authors; but in each case the author is obviously borrowing the story from Plato. Plato states that these two maxims were the cream of the wisdom of the wise men: it would be hardly too much to say that upon them the whole structure of Greek ethical philosophy was based. For the construction, and for the practice of thus dedicating wisdom to a god, Kroschel aptly quotes Diog. Laert. IX. 6 ἀνέθηκε (sc. Ἡράκλειτος ) δ’ αὐτὸ (sc. τὸ περὶ φύσεως βιβλίον ) εἰς τὸ τῆς Ἀρτέμιδος ἱερόν .

51. τοῦ δὴ ἕνεκα ταῦτα λέγω ; Plato frequently enlivens his style by such self-interrogations: Sauppe quotes ( inter alia ) 457E τοῦ δὴ ἕνεκα λέγω ταῦτα ; Apol.40B τί οὖν αἴτιον εἶναι ὑπολαμβάνω ;

τῶν παλαιῶν τῆς φιλοσοφίας. τῶν παλαιῶνin this emphatic place suggests the contrast with Protagoras and the νεώτεροι whose τρόπος is μακρολογία .

53. καὶ δὴ καί marks the application to the present case: cf. 18A ὥσπερ οὖν ἂν κτλ.—καὶ δὴ καὶ νῦν.

56. καθέλοι : cf. 344C Sauppe quotes Theocr. 22. 115 (of Polydeuces overcoming Amycus) Διὸς υἱὸς ἀδηφάγον ἄνδρα καθεῖλεν .

59. τούτου ἕνεκα , i.e. τοῦ εὐδοκιμεῖν ἕνεκα. τούτῳ in τούτῳ ἐπιβουλεύων is τούτῳ τῷ ῥήματι : with the construction cf. IV. 443B εὐθὺς ἀρχόμενοι τῆς πόλεως οἰκίζειν and see Stallbaum on I. 342B.

59. κολοῦσαι . Hdt. VII. 10. 5 φιλέει γὰρ ὁ θεὸς τὰ ὑπερέχοντα πάντα κολούειν (Sauppe).

5. οὐδὲ πρὸς ἕνα λόγον : see on 335D πρὸς λόγον = εὔλογον or εὐλόγως is frequent in Plato: see on 351E

6. ἐὰν μή τις ὑπολάβῃ κτλ. The natural interpretation of the μέν (since it follows ἀγαθόν , not γενέσθαι ) is this. While it is difficult ‘ γενέσθαι ’ truly good , it is yet possible to reach a fair standard of ἀρετή . The antithesis is implied if not actually expressed in the sequel: see 345D and 346C Cf. Introduction, p. xxiv.

11. ἐπὶ τούτῳ—τὴν ἀλήθειαν. τούτῳis τῷ ἀγαθῷ , and τὴν ἀλήθειαν = τὸ ‘ ἀλαθέως ’.

13. εὔηθες—Σιμωνίδου . As well might Socrates say it is εὔηθες to use the word ἄριστος .

14. ὑπερβατόν is used in its technical grammatical sense, Heindorf quotes Long. de Sublim. 22 ἔστιν δὲ (sc. τὰ ὑπ ερβατὰ ) λέξεων ἢ νοήσεων ἐκ τοῦ κατ’ ἀκολουθίαν κεκινημένη τάξις καὶ οἱονεὶ χαρακτὴρ ἐναγωνίου πάθους ἀληθέστατος .

15. ὑπειπόντα . There is no need to explain ὑπειπόντα here as ‘interpreting’ (a sense for which no exact parallel is quoted). ὑπειπεῖν is used here in its regular sense ‘to say before’, ‘to say first’, as in Ar. , 54-5 φέρε νυν κατείπω τοῖς θεαταῖς τὸν λόγον, ὀλίγ’ ἄτθ’ ὑπειπὼν πρῶτον αὐτοῖσιν ταδί. The meaning is: if we would arrive at Simonides' meaning, we must regard the adverb ἀλαθέως as transposed, and speak the words of Pittacus first ( ὑπειπόντα τὸ τοῦ Πιττακοῦ ): i.e. instead of saying ἄνδρ’ ἀγαθὸν μὲν ἀλαθέως γενέσθαι χαλεπόν first, and quoting Pittacus' saying afterwards, we should begin with χαλεπὸν ἐσθλὸν ἔμμεναι (line 18 εἰπόντα Ὦ ἄνθρωποι, χαλεπὸν ἐσθλὸν ἔμμεναι ), and make ἄνδρ’ ἀγαθόν etc. follow as Simonides' answer (ll. 19-22), putting ἀλαθέως last. οὑτωσί πως shows that ὑπειπόντα is to be explained in the sequel, viz. from ὥσπερ ἄν to ἀλαθέως in l. 22.

18. εἰπόντα : i.q. τὸν μὲν εἰπόντα : see on 330A Note the precision in the use of the tenses ( εἰπόντα , but ἀποκρινόμενον ).

20. οὐ γὰρ εἶναι ἀλλὰ γενέσθαι μέν . Cf. , 43D οὔτοι δὴ ἀφῖκται ἀλλὰ δοκεῖ μέν μοι ἤξειν κτλ. μέν without following δέ is frequent in a clause introduced by ἀλλά , the antithesis being contained in the preceding negative clause.

29. τύπον . Socrates' criticism of the poem so far has been δῑ ἀκριβείας : cf. II. 414A ὡς ἐν τύπῳ, μὴ δῑ ἀκριβείας εἰρῆσθαι. See also note on ὑπογράψαντες γραμμὰς τῇ γραφίδι in 326D above.

30. παντὸς μᾶλλον is frequent in Plato for ‘assuredly’: cf. , 49B.

6. ἀλλὰ θεὸς ἂν μόνος—γέρας . Plato, 204A θεῶν οὐδεὶς φιλοσοφεῖ οὐδ’ ἐπιθυμεῖ σοφὸς γενέσθαι· ἔστι γάρ.

8. ἄνδρα δ’ οὐκ ἔστι—καθέλῃ : see below on 344E l. 30.

12-14. οὐ τὸν κείμενον—τὸν δὲ κείμενον οὔ . For the repetition compare (with Sauppe) 521E οὐ πρὸς χάριν λέγων—ἀλλὰ πρὸς τὸ βέλτιστον, οὐ πρὸς τὸ ἥδιστον.

15. ὄντα ποτὲ—καθέλοι. ποτέgoes with καθέλοι as with καταβάλοι in l. 13.

19. καὶ ἰατρὸν ταὐτὰ ταῦτα. ταὐτὰ ταῦταis virtually adverbial as in , 90D οὐκοῦν καὶ περὶ αὐλήσεως καὶ τῶν ἄλλων τὰ αὐτὰ ταῦτα ;

20. ἐγχωρεῖ κακῶ? γενέσθαι . Socrates says γενέσθαι , although Simonides said ἔμμεναι ; see on E (l. 26) below.

21. παρ’ ἄλλου ποιητοῦ . Xen. 1. 2. 20 μαρτυρεῖ δὲ καὶ τῶν ποιητῶν ὅ τε λέγων ‘ ἐσθλῶν μὲν γὰρ ἀπ’ ἐσθλὰ διδάξεαι· ἢν δὲ κακοῖσι συμμίσγῃς, ἀπολεῖς καὶ τὸν ἐόντα νόον ’, καὶ ὁ λέγων ‘ αὐτὰρ ἀνὴρ ἀγαθὸς τοτὲ μὲν κακός, ἄλλοτε δ’ ἐσθλός ’. The first of these quotations is known (see , 95D) to be from Theognis, so that it seems clear that the second must be from some other poet. It is difficult to resist the impression that the author of this line intended ἀγαθός, κακός , and ἐσθλός to have a political sense, and was alluding to the ups and downs of an aristocrat's life in times of civil dissension: cf. Theognis, 1109-10 Κύρν’, οἱ πρόσθ’ ἀγαθοὶ νῦν αὖ κακοί, οἱ δὲ κακοὶ ρπὶν νῦν ἀγαθοί· τίς κεν ταῦτ’ ἀνέχοιτ’ ἐσορῶν ; Euripides, , 595 ff. (quoted by Cron) pointedly contradicts the sentiment of this line (as interpreted by Socrates): ἄνθ ρωποι δ’ ἀεὶ ὁ μὲν πονηρὸς οὐδὲν ἄλλο πλὴν κακός, ὁ δ’ ἐσθλὸς ἐσθλός, οὐδὲ συμφορᾶς ὕπο (alluding to Simonides' poem) φύσιν διέφθειρ’, ἀλλὰ χρηστός ἐστ’ ἀεί .

25. εὐμήχανον καὶ σοφὸν καὶ ἀγαθόν. εὐμήχανοςis identified with ἀγαθός by means of the middle step σοφός, σοφία being ἀγαθόν because it is one of the virtues. See on τῷ ἐπιστάτῃ καὶ ἐπαἳοντι in , 47B.

26. οὐκ ἔστι μὴ οὐ κακὸν ἔμμεναι . If Simonides had consistently carried out the distinction between γένεσις and οὐσία attributed to him (in part rightly: see on 344B l. 4), he would have used γενέσθαι , not ἔμμεναι here. Socrates throughout interprets ἔμμεναι as equivalent to γενέσθαι in this part of the poem, in spite of his previous distinction: but so (apparently) did Simonides: see on 344B

28. τὸ δ’ ἐστίν κτλ. τὸ δέ‘whereas in point of fact’ is very frequent in Plato: see on 23A. Notice how Socrates reverts to the beginning of the poem: see note on 346D and Appendix I, p. 214.

29. (δυνατὸν δἒ ἐσθλόν . There is not sufficient ground for rejecting (with most of the editors) ἐσθλόν : its position is a trifle awkward, but not more, since δυνατὸν δέ is parenthetical, the μέν after γενέσθαι being balanced by δέ in ἔμμεναι δέ . Heindorf reads δυνατὸν δέ, ἐσθλὸν δ’ ἔμμεναι with slight MS. authority.

30. πράξας μὲν γὰρ εὖ κτλ. γάρis probably due to Plato, who represents this sentence as adducing a reason for ἔμμεναι δὲ ἀδύνατον : see Appendix I, p. 217. πράξας εὖ is ‘if he has prospered’: the whole sentiment is the converse of ἄνδρα δ’ οὐκ ἔστι μὴ οὐ κακὸν ἔμμεναι and is characteristic of the ordinary Greek moral code: cf. Hom. XVIII. 136-7 τοῖος γὰρ νόος ἐστὶν ἐπιχθονίων ἀνθρώπων οἷον ἐπ’ ἦμαρ ἄγῃσι πατὴρ ἀνδρῶν τε θεῶν τε. The view that affliction involves moral degeneracy appears in the transition of meaning in πονηρός and μοχθηρός from ‘afflicted’ to ‘depraved’: the common view that ‘prosperity’ brings virtue is involved in the usual equivocation on εὖ πράττειν ‘fare well’ and ‘do well’; see on , 47E and , 3A. After κακὸς δ’ εἰ κακῶς is to be understood ἔπραξεν .

32. τίς οὖν εἰς γράμματα κτλ. In order to read into Simonides the doctrine that virtue is knowledge and vice ignorance, Socrates assigns to πράξας εὖ in the poem the meaning of acting well, rather than faring well.

36. κακὸς δὲ κακῶς , sc. πράξας : a free rendering of κακὸς δ’ εἰ κακῶς of the poem, which Kral (following Ast) reads here against the MSS.

40. κακῶς πράξαντες , i.e. εἰ κακῶς πράξαιμεν , as the words of the poem show. Socrates' reasoning is: to become a bad doctor by practising badly, you must first have been a good doctor: for if you cannot become a doctor by practising badly, obviously you cannot become a bad doctor. The argument is as fallacious as it is ingenious: it assumes that κακὸς ἰατρός is a twofold notion, and more than ἰατρός , whereas it is a single notion and less. It would be more in conformity with experience to say that the ἰδιώτης does become by practising badly a κακὸς ἰατρός .

45. αὕτη γὰρ μόνη—στερηθῆναι . This sentence (necessary as the converse of the statement in 345A that good action comes from knowledge) is introduced as an explanation of ὑπὸ χρόνου κτλ. , because χρόνος, πόνος, νόσος , etc., produce ἐπιστήμης στέρησις .

47. μέλλει—γενέσθαι . See on 312C above.

50. διατελοῦντα ἀγαθόν explains ἀγαθόν more precisely: cf. 344B γενόμενον δὲ διαμένειν ἐν ταύτῃ τῇ ἕξει καὶ εἶναι ἄνδρα ἀγαθόν. διατελεῖν without a participle seems not to be elsewhere found in Plato, though it occurs in Thucydides and others.

9. ἐπί θ’ ὑμῖν εὑρὼν ἀπαλλελέω : after Bergk's emendation for the ἔπειθ’ ὑμῖν of the MSS., which cannot be made to scan: see Appendix I, p. 217. It might be possible (in view especially of φησίν in the next line) to regard ἔπειτα here as no part of the poem, were it not for 346D where it must be part. ἐπί θ’ ὑμῖν is the most probable of the many emendations proposed and accounts by far the most easily for the reading of the MSS. ἐπί is to be taken with εὑρών , and τε is like the quasignomic τε in Homer, e.g. I. 218 ὅς κε θεοῖς ἐπιπείθηται, μάλα τ’ ἔκλυον αὐτοῦ. Translate ‘“trust me, I will tell you when I've found your man”, says he’. ὑμῖν (a kind of ethic dative) does not belong so much to ἀπαλλελέω as to ἐφευρών . What Simonides himself wrote was no doubt ἐπί τ’ ὔμμιν (availing himself of the Aeolic form as in the next line); this Plato put in Attic as ἐπί θ’ ὑμῖν , from which to ἔπειθ’ ὑμῖν the step was easy. Sauppe's ἐπὶ δή μιν εὑρών and Kroschel's ἐπεὶ οὔτιν’ εὑρών are hardly happy.

10. φησίν applies only to the last line: the former φησὶ γάρ covers the first sentence.

12. πάντας ὅστις : so below in l. 16 ὥστε τούτους φάναι ἐπαινεῖν ὃς ἄν and in l. 23 ὃς ἂν—τούτων : see also note on 319D

14. ἀνάγκῃ δ’ οὐδὲ θεοὶμάχονται . Proverbial: cf. , VII. 818B ἔοικεν ὁ τὸν θεὸν πρῶτον παροιμιασάμενος—εἰπεῖν ὡς οὐδὲ θεὸς ἀνάγκῃ μή ποτε φανῇ μαχόμενος . Aars quotes the words of Agamemnon in XIX. 86 ἐγὼ δ’ οὐκ αἴτιός εἰμι, ἀλλὰ Ζεὺς καὶ Μοῖρα καὶ ἠεροφοῖτις Ἐρινύς.

17. ὃς ἂν ἑκὼν—ποιῇ : but ἑκὼν ὅστις ἕρδῃ without ἄν above, as often in poetry: Goodwin, p. 208, § 540.

19. οὐδεὶς τῶν σοφῶν—ἑκόντα ἐξαμαρτάνειν . The doctrine that no one sins willingly—a corollary of the view that vice is only ignorance—is characteristic of the ethical teaching both of Socrates and Plato (compare note on 324A and on 2C), but not of all Greek sages, and it is only by the most perverse sophistry that Socrates here reads it into Simonides, ignoring entirely the words ἀνάγκῃ δ’ οὐδὲ θεοὶ μάχονται .

23. καὶ δὴ καί : see on 343B above.

26. καλὸν κἀγαθόν : see on 21D.

27. καὶ ἐπαινέτην . The words φιλεῖν καὶ ἐπαινεῖν which follow in the MSS. were ejected by Heindorf.

28. μητέρα ἢ πατέρα ἀλλόκοτον . Sauppe remarks that from Homer onwards μήτηρ generally comes first in such enumerations—an interesting survival, perhaps, of the greater importance assigned to the mother in primitive Greece. ἀλλόκοτον (‘eccentric’ as in VI. 487D) from ἄλλος (in its sinister sense) and κότος (i.q. τρόπος, ἦθος, ὀργή ) according to Phrynichus (quoted by Kroschel).

29. ἢ πατρίδα : see , 50E-51C.

34. ἔτι μᾶλλον : not = μᾶλλον ἢ κατ’ ἀξίαν , but—as is presently explained—because they ‘add voluntary feuds to those which they cannot avoid’.

36. ἀναγκαίαις : Heusde's correction for ἀνάγκαις of MSS.

ἐπικρύπτεσθαι—ἀναγκάζεσθαι . Plato is probably thinking of Socrates after his trial as he depicts him in the ἀναγκάζεσθαι ‘are constrained’ is of course passive.

39. παραμυθεῖσθαι , like mulcere , as often. Sauppe quotes Hor. XIII. 18 ‘deformis aegrimoniae dulcibus alloquiis ’.

41. ἡγήσατο—ἀναγκαζόμενος : ‘believed—that he had praised, etc.’ Plato deals a sly thrust at Simonides' notorious avarice, as Pindar (quoted by Sauppe) does in 11. 6 ἁ Μοῖσα γὰρ οὐ φιλοκερδής πω τότ’ ἦν οὐδ’ ἐργάτις . The words ἀλλ’ ἀναγκαζόμενος contain the sting; for the life of the χρηματιστής is βίαιος (Ar. 1. 3. 1096 a . 5).

46. ἔμοιγ’ ἐξαρκεῖ κτλ. See Appendix I, pp. 214 ff. for the arrangement of this part of the poem.

48. γ’ ὀνησίπολιν . The MSS. read γ’ ὀνήσει πόλιν , which G. Hermann emended to τ’ ὀνησίπολιν , Bergk to ὀνασίπολιν .

49. οὐ μήν : so the MSS., and so, most probably, Plato; but Simonides can hardly have written this, which will not scan; see Appendix I, p. 217.

51. οὐ γάρ εἰμι φιλόμωμος probably belongs to an earlier part of the poem (before ἔμοιγ’ ἐξαρκεῖ : cf. οὐ διὰ ταῦτά σε ψέγω, ὅτι εἰμὶ φιλόψογος in line 45): see Appendix I, p. 217.

53. γενέθλα is Stephanus' correction for γένεθλα of the MSS., a mistake due to supposing that ἀπείρων (not from ἄπειρος ) went with ἠλιθίων : it belongs to γενέθλα .

56. πάντα τοι καλά—μέμικται sums up the whole moral teaching of the poem.

60. καὶ οὐ ζητῶ κτλ. Socrates recapitulates part of the poem by way of interpreting the final text: see note on τὸ δ’ ἐστὶν γενέσθαι κτλ. in 344E above and Appendix I, p. 214.

62. τούτου γ’ ἕνεκα : ‘si hoc spectetur s. requiratur, τὸ πανάμωμον εἶναι ’ (Heindorf): cf. 85B ἀλλὰ τούτου γ’ ἕνεκα λέγειν τε χρὴ καὶ ἐρωτᾶν ὅ τι ἂν βούλησθε . The sense is: if I must wait for perfection before beginning to praise, I shall never praise anyone.

66. ὡς πρὸς Πιττακὸν λέγων : not serious, of course, nor true.

67. διαλαβεῖν : i.e. make a division, ‘pause’, virtually ‘punctuate’. The editors compare the use of διαστίξαι in Ar. III. 5. 1407 b . 14 τὰ γὰρ Ἡρακλείτου διαστίξαι ἔργον διὰ τὸ ἄδηλον εἶναι ποτέρω? πρόσκειται .

68. ἄκων δ’ ἔστιν οὕς : see on 344B

70. νῦν δὲ—σφόδρα γάρ . Cf. 38B νῦν δὲ—οὐ γὰρ ἔστινand note in loc. Here διὰ ταῦτα sums up the clause σφόδρα γὰρ— λέγειν .

6. ἐπιδείξω : see on ἐπιδειξάμενος in 328D

8. ὡμολογησάτην : in 338D where, however, the terms of the agreement are somewhat more stringent.

13. περὶ μὲν ᾀσμάτων—ἐάσωμεν . Heindorf quotes , 113D ἐάσαντες οὖν περὶ αὐτῶν σκοποῦσιν ὁπότερα συνοίσει πράξασιν .

16. καὶ γὰρ δοκεῖ μοι κτλ. This passage (from τὸ περὶ ποιήσεως down to κἂν πάνυ πολὺν οἶνον πίωσιν ) is quoted by Athenaeus, 111. 51. It is doubtful whether Plato is here animadverting on Xenophon's , in which (2. 1 and 9. 3 ff) both a flute-girl and a dancing-girl appear: the words τοῖς συμποσίοις τοῖς τῶν φαύλων καὶ ἀγοραίων ἀνθρώπων would contain a very pretty hit at Xenophon, if the evidence for the allusion were more complete. There is a similar passage in the of Plato (176E) εἰσηγοῦμαι τὴν μὲν ἄρτι εἰσελθοῦσαν αὐλητρίδα χαίρειν ἐᾶν, αὐλοῦσαν ἑαυτῇ ἢ ἂν βούληται ταῖς γυναιξὶ ταῖς ἔνδον : this passage Athenaeus (XI. 112) cites in support of his theory of a literary rivalry between Plato and Xenophon.

18. ἀγοραίων : like the Latin circumforanei , ‘ ἀγοραῖος , vel ut critici veteres volunt, ἀγόραιος , est ὁ ἐν ἀγορᾷ τεθραμμένος , s. qui totos dies forum conterit, quem ἀγορᾶς περίτριμμα Comicus appellat, vilis et ex ima plebe homo, Aristoph. Eqq. 181 ὁτιὴ πονηρὸς κἀξ ἀγορᾶς εἶ καὶ θρασύς : 214 τὰ δ’ ἄλλα σοι πρόσεστι δημαγωγικά, φωνὴ μιαρά, γέγονας κακῶς, ἀγόραιος εἶ ’ (Heindorf).

21. τιμίας ποιοῦσι τὰς αὐλητρίδας : ‘run up the price of flute-players’: τὸ γὰρ σπάνιον—τίμιον ( 304B). Heindorf aptly quotes Xen. IV. 10 χρυσίον ὄταν πολὺ παραφανῇ αὐτὸ μὲν ἀτιμότερον γίγνεται, τὸ δὲ ἀργύριον τιμιώτερον ποιεῖ .

25. καὶ πεπαιδευμένοι . So Athenaeus (III. 51) and the second hand in T: B and T read πεπαιδευμένοι . Schanz brackets πεπαιδευμένοι , but ὑπὸ ἀπαιδευσίας in l. 21 is in its favour.

οὐκ ἂν ἴδοις—ψαλτρίας . Cf. 173D δεῖπνα καὶ σὺν αὐλητρίσι κῶμοι, οὐδὲ ὄναρ πράττειν προσίσταται αὐτοῖς (i.e. τοῖς ὡς ἀληθῶς φιλοσόφοις ).

30. κἂν πάνυ πολὺν οἶνον πίωσιν : like Socrates in the : see 220A, 223C. In , I. 637B ff. Plato argues that wine properly used is the means of teaching self-control, since the man who will be σώφρων when drunk will a fortiori be σώφρων when sober: cf. ibid. II. 673E ff.

31. ἐὰν μὲν κτλ. μένsuggests that in different society the same would not be the case—an antithesis which is already expressed in 347C D. Cf. 17B εἰ μὲν γὰρ τοῦτο λέγουσιν, ὁμολογοίην ἂν ἔγωγε οὐ κατὰ τούτους εἶναι ῥήτωρ. For λάβωνται cf. (with Sauppe) 218A νέου ψυχῆς μὴ ἀφυοῦς ὅταν λάβωνταιcf. (sc. οἱ ἐν φιλοσοφίᾳ λόγοι ).

33. οὔτε ἀνερέσθαι—ἐπαγόμενοί τε : for the sentiment see on 329A above: for οὔτε—τε on 309B for αὐτούς after οὕς on 313A ἐπάγεσθαι is the regular word for ‘cite’, e.g. II. 364C μάρτυρας ποιητὰς ἐπάγονται.

36. ὃ ἀδυνατοῦσιν : ὅ , not in B and T, was restored by Heindorf: it has since been found in a Vienna MS.

38. ἐῶσιν : the subject is supplied from ἀνδρῶν οἷοίπερ ἡμῶν οἱ πολλοί φασιν εἶναι .

41. καταθεμένους : for the asyndeton see on 330A Both here and in 59C κατατίθεσθαιis not used in the sense of laying aside for good, but rather (as generally) putting away to be resumed again. ‘Putting the poets on their shelves’ would give the effect of the Greek.

45. παρέχειν : see on συγκαθεῖναι in 336A above.

51. καὶ νῦν : as before 336B

54. ἵνα τούτῳ μὲν ταῦτα συνειδῶμεν . 193E εἰ μὴ συνῄδη Σωκράτει τε καὶ Ἀγάθωνι δεινοῖς οὖσι περὶ τὰ ἐρωτικά. In σύνοιδα ἐμαυτῷ ἀγαθῷ ὄντι συν‐ goes with ἐμαυτῷ —I am conscious with myself that I am good—by the convenient fiction which separates the observing ego from the observed. Cases where the dative is not a reflexive pronoun are later in development and presuppose only knowledge about, not knowledge along with, the object.

56. ὥς γέ μοι . B has ὥστε μοι , T ὡς γ’ ἐμοί ( sic ). There is no occasion for the emphatic form of the pronoun.

58. τῶν ἄλλων σχεδόν τι. σχεδόν τιneed not be accompanied by πάντες : compare , 59C ἄλλος δέ τις παρῆν; σχεδόν τι οἶμαι τούτους παραγενέσθαι .

59. προὐτράπετο . No other example of the aorist middle of προτρέπω is quoted from any Attic author, but ἐτραπόμην is common, e.g. 21B μόγις πάνυ ἐπὶ ζήτησιν αὐτοῦ τοιαύτην τινὰ ἐτραπόμην.

60. ἐκέλευεν : the imperfect of this verb is used in narrative style for the aorist.

6. εὐπορώτεροι—ἐσμεν , i.e. οὕτως ἔχοντες (viz. σὺν δύο ἐρχομένω ); it is hardly necessary to read οὕτως πως or to change πως into οὕτως as Heindorf suggests.

8. αὐτίκα περιιὼν—ἐντύχῃ . The ἐπιδείξηται and βεβαιώσηται —deliberative subjunctives—of the MSS. would imply too much eagerness: the change to ἐπιδείξεται and βεβαιώσεται is adopted by Sauppe and others. This mistake (or the reverse) is tolerably frequent in Plato's MSS., cf. 489A( βεβαιώσωμαι T, βεβαιώσομαι B), 505E ( ποιήσωμεν BT), 510A ( ἀδικήσωμεν BT).

10. ἕνεκα τοὐτου refers to the following clause introduced by ἡγούμενος . Heindorf refers to his note on , 102D λέγω δὲ τοῦδ’ ἕνεκα, βουλόμενος δόξαι σοὶ ὅπρ ἐμοί .

17. ὥστε καὶ κτλ. The reference is to 316D ff. καί does not go with ἄλλων but with the whole sentence (as Sauppe points out).

19. ὑποκηρυξάμενος ‘having had yourself heralded as a sophist’: compare Aeschin. in Ctes. 41 ἄλλοι δέ τινες ὑποκηρυξάμενοι τοὺς αὑτῶν οἰκέτας ἀφίεσαν ἀπελευθέρους . The force of ὑπο‐ is probably the same as in ὑπεῖπον : see note on 343E Notice the effect of the double σεαυτόν (both of them after a natural pause): σεαυτὸν ὑποκηρυξάμενος—σεαυτὸν ἀπέφηνας : the ‐αυτόν is brought out so strongly as to suggest that Protagoras is the only true Αὐτός or Master.

22. μισθὸν—ἄρνυσθαι : 328B where however it is not said that Protagoras was the first to take a fee. The Homeric word ἄρνυσθαι is occasionally used by prose writers in the phrase μισθὸν ἄρνυσθαι : whence μισθαρνεῖν .

25. ἐκεῖνα—τὰ μὲν—τὰ δέ : see above on 330A

27. ἦν δὲ—τόδε : 329C ff.

39. ἕκαστα attracted for ἔκαστον : so VIII. 546C δεομένων ἑνὸς ἑκάστων, i.e. ‘lacking—each of them—one’.

42. οὐδέν σοι ὑπόλογον τίθεμαι. ὑπόλογον τίθεμαιlike ὑπόλογον ποιοῦμαι ( 189B) and ὑπολογίζομαι ( 28B) is ‘to set down against’, ‘per contra’, whence ‘object’ as here.

2. μόρια μέν . The antithesis to μέν was already expressed in ἔφησθα οὖν σὺ οὐκ ὀνόματα ἐπὶ ἑνὶ εἶναι 349B

4. ἡ δὲ ἀνδρεία κτλ. Protagoras therefore yields to Socrates' arguments so far as they have yet gone, and takes his stand on the only virtue the relation of which to the others has not yet been discussed: see on 333C and D and Introduction, p. xiii.

7. ἀκολαστοτάτους—ἀνδρειοτάτους δέ : like Otho (Tac. II. 49), apropos of whose death Merivale quotes the lines of Byron, which well illustrate the sentiment of Plato:

And strange to say, the sons of pleasure, They who have revelled beyond measure In beauty, wassail, wine and treasure, Die calm, and calmer oft than he Whose heritage was misery.

8. ἀνδρειοτάτους δὲ διαφερόντως . The extreme difference (cf. πάνυ πολὺ διαφέρον in l. 4) between courage and the other virtues is brought out by representing those most lacking in the other virtues as sometimes ‘supremely brave beyond all others’: below in 359B the διαφερόντως is omitted as unnecessary in a recapitulation. Sauppe quotes 23C ἡ νῦν Ἀθηναίων οὖσα πόλις ἀρίστη πρός τε τὸν πόλεμον καὶ κατὰ πάντα εὐνομωτάτη διαφερόντως: cf. also 487B αἰσχυντηροτέρω μᾶλλον τοῦ δέοντος. Various suggestions have been proposed, but the text is sound.

ἔχε δή . ‘Hac sistendi formula aut monetur, qui rectam viam ingressus est, ut caveat, ut hic et Gorg. 460A, aut revocatur, qui a recta aberravit, ut Prot. 349D rsquo; (Wohlrab on 186B).

9. πότερον τοὺς ἀνδρείους θαρραλέους κτλ. Socrates proceeds by reasoning thus: (1) ἀνδρεῖοι are θαρραλέοι ; (2) ἐπιστήμονες are θαρραλέοι ( 349E l. 16 to 350B l. 25); (3) none who are θαρραλέοι without ἐπιστήμη are ἀνδρεῖοι ( 350B ll. 25-30). From this he infers that σοφοί (i.e. ἐπιστήμονες ) are ἀνδρεῖοι , i.e. that σοφία is ἀνδρεία . The reasoning is far from cogent. In the first place, we have to assume (it is nowhere stated) that θαρραλέοι contains two classes and no more, viz. θαρραλέοι with knowledge and θαρραλέοι without knowledge: the assumption would be (to Socrates) a natural one, since (according to the reasoning in ch. XIX ff.) every one who is not ἐπιστήμων is ἀνεπιστήμων . Now as ἀνδρεῖοι are θαρραλέοι and no ἀνεπιστήμονες (in the class θαρραλέοι ) are ἀνδρεῖοι , it follows that ἀνδρεῖοι are ἐπιστήμονες , but even then the conclusion of Socrates is not warranted— that ἐπιστήμονες are ἀνδρεῖοι , since ἀνδρεῖοι may be only a part of ἐπιστήμονες . Socrates —consciously or unconsciously—covers his erroneous reasoning by another fallacy when about to draw his conclusion in 350B , ll. 31 and 34, where see note.

10. καὶ ἴτας γ’, ἔφη : i.e. they not only have θάρρος (which may be quiescent) but they put it into action. In ἰέναι Protagoras contrives to give the derivation of ἴτης : Sauppe refers to the scholiast on Ar. , 444 ἴτης. ἰταμός, ἀναιδής, καὶ δῑ αὐτῶν χωρῶν τῶν πραγμάτων. ἀνδρεῖοςis coupled with ἴτης in 203D and with θρασύς in Ar.

11. φέρε δή κτλ. This section (from φέρε δή to ὡς οἷόν τε μάλιστα in l. 16) is intended to prepare the way for the proof of the third proposition (see on l. 9) in 350B see note on l. 29, and cf. 192C, where the proof that ἄφρων καρτέρησις is not ἀνδρεία is introduced in much the same way: σχεδὸν γάρ τι οἶδα—ὅτι τῶν πάνυ καλῶν πραγμάτων ἡγεῖ σὺ ἀνδρείαν εἶναι .

14. εἰ μὴ μαίνομαί γε : ‘as I'm a sane man’. This and similar phrases are frequent in Plato, e.g. 283E, X. 608D ( εἰ μὴ ἀδικῶ γε ), 511B οἶδα—εἰ μὴ κωφός γ’ εἰμι.

τὸ μέν τι—τὸ δέ τι . See on 12A. Kroschel quotes 13C τὰς μὲν εἶναί τινας ἀγαθὰς ἡδονάς, τὰς δέ τινας—κακάς .

17. αολυμβῶσιν . This and the next example are given also in 193 B ff. Sauppe thinks that the object of such diving (an art in which the Greeks were very expert: see Thuc. IV. 26. 8) may have been to clean the wells and the like.

21. τίνες δὲ πέλτας ἔχοντες . See Introduction, p. xxxii.

24. αὐτοὶ ἑαυτῶν—ἤ . The notion ‘than’ is expressed twice: see note on 44C καίτοι τίς ἂν αἰσχίων εἴη ταύτης δόξα ἢ δοκεῖν κτλ., and cf. below, 350E and (with Sauppe) Hdt. VIII. 86 ἐγένοντο—μακρῷ ἀμείνονες αὐτοὶ ἑωυτῶν ἢ πρὸς Εὐβοίῃ .

29. αἰσχρὸν μεντἄν—ἀνδρεία : and if αἰσχρόν , not ἀρετή (which it is), since all ἀρετή is καλόν ( 349E . Cf. 182C ἡ δέ γε ἀνδρεία ὡμολογεῖτο καλὸν εἶναιand with the general sentiment 88B οἶον ἀνδρεία, εἰ μὴ ἔστι φρόνησις ἡ ἀνδρεία ἀλλ’ οἶον θάππος τι· οὐχ ὅταν μὲν ἄνευ νοῦ θαππῇ ἄνθπωρος, βλάρτεται, ὅταν δὲ σὺν νῷ, ὠφελεῖται ;

30. λέγεις : the present is idiomatically used in referring to an earlier part of a discussion not yet ended: see on ὅπερ λέγω in 21A. With λέγεις followed by an accusative in this sense compare 199E πειρῶ δὴ καὶ τὸν Ἔρωτα εἰπεῖν· ὁ Ἔρως ἔρως ἐστὶν οὐδενὸς ἢ τινός ;

31. οὐχὶτοὺς θαρραλέους εἶναι . This Protagoras did not say, but only that οἱ ἀνδρεῖοι are θαρραλέοι : see 349E For this reason Sauppe and others reject τούς before θαρραλέους , and in favour of this urge the reply of Protagoras ( καὶ νῦν γε ): but inasmuch as (1) οὐκοῦν οὗτοι—μαινόμενοι φαίνονται seems to be intended to be incompatible with τοὺς ἀνδρείους οὐχὶ—εἷναι , (2) θαρραλεώτατοι δὲ ὄντες ἀνδρειότατοι in l. 35 clearly implies that θαρραλέοι are conceived of as ἀνδρεῖοι (no less than ἀνδρεῖοι as θαρραλέοι ), we must, if we regard the argument as a whole, retain the MSS. reading. Protagoras' καὶ νῦν γε is an unwary admission: he does not at first catch the difference between οἱ ἀνδρεῖοί εἰσιν οἱ θαρραλέοι and οἱ ἀνδρεῖοί εἰσι θαρραλέοι , and Socrates avails himself of his opponent's slip to hasten to his conclusion—which but for this misrepresentation could only be expressed as ἀνδρεῖοι are σοφοί ( = ἐπιστήμονες ), not σοφοί are ἀνδρεῖοι : see on 349E and 350D

34. ἐκεῖ , i.e. in the case of θάρρος based on σοφία , illustrated in 350A Sauppe (after Schöne) reads οἳ σοφώτατοι, οὗτοι κτλ. : but οἱ σοφώτατοι need not go closely with οὗτοι (which is resumptive) any more than οἱ οὕτω θαρραλέοι ὄντες with the preceding οὗτοι . There is a kind of chiasmus in the order οὗτοι— οἱ—θαρραλέοι ) ( οἱ σοφώτατοι οὗτοι .

36. οὐ καλῶς—μνημονεύεις κτλ . Protagoras sees now that ‘All brave are bold’ is not equivalent to ‘All bold are brave’, and rectifies his καὶ νῦν γε by pointing out that he originally said only ‘All brave are bold’ but he confines his attention to this point, without touching on the fundamental flaws in Socrates' argument from 349E to 350B

39. ὡμολόγησα : in 349E

40. τότε ἤρου . So B and T. In τότε there is perhaps a latent contrast to καὶ νῦν γε of l. 31. Protagoras is correcting his recent slip—had you asked me then (viz. at 349E , I should have answered rightly. τοῦτο has inferior MSS. authority.

41. τοὺς δὲ ἀνδρείους ὡς οὐ θαρραλέοι εἰσίν . The οὐ after ὡς is due to a confusion between (1) τοὺς δὲ ἀνδρείους ὡς οὐ θαρραλέοι εἰσίν, οὐδαμοῦ ἐπέδειξας and (2) οἱ δὲ ἀνδρεῖοι ὡς θαρραλέοι εἰσίν, τὸ ἐμὸν ὁμολόγημα, οὐδαμοῦ κτλ. The insertion of οὐ is the more natural because after verbs of refuting and the like the object clause gives what is maintained and not what is refuted, whence ἐλέγχειν ὡς οὐ, ἀντιλέγειν ὡς οὐ etc.

45. καὶ ἐν τούτῳ οἴει κτλ. Protagoras ignores Socrates' third proposition in 350B (25-30), where it is shown that θάρρος without ἐπιστήμη is not ἀνδρεία ; see next note.

48. πρῶτον μὲν γὰρ κτλ. , i.e. you might as well argue (1) οἱ ἰσχυροί are δυνατοί , (2) οἱ ἐπιστήμονες are δυνατοί , therefore οἱ ἐπιστήμονες are ἰσχυροί . This would only be correct if for (1) we substituted οἱ δυνατοί are ἰσχυροί . To make Protagoras' picture of Socrates' argument complete, we should have to add (3) none who are δυνατοί without ἐπιστήμη are ἰσχυροί —but this is not true, whereas Socrates' third proposition is. The completed picture therefore fails to represent correctly Socrates' reasoning in each of its steps, but none the less are Protagoras' objections strictly relevant, and indeed fatal to Socrates' conclusion as expressed in 350C (33-6), and that is why Socrates makes no reply.

58. ἀπὸ ἐπιστήμης—τὴν δύναμιν κτλ. Protagoras says in effect: δύναμις and ἰσχύς are of the body, θάρσος and ἀνδρεία of the soul. δύναμις may come from knowledge (as when one has learnt how to wrestle: above 350E , or from madness or rage (as in the feats of madmen, or men inspired with the thirst for vengeance; such men have no real physical strength or ἰσχύς as it is presently defined, but excel themselves by virtue of δύναμις ). ἰσχύς comes from natural constitution ( φύσις ) and proper nurture of the body (by gymnastics in the widest sense). Similarly θάρσος may result from art ( τέχνη is substituted for ἐπιστήμη in view of the illustrations in 350A , from rage or madness (as when Empedocles leapt into Etna); ἀνδρεία comes from the native character and proper nurture of the soul (cf. III. 410D τὸ θυμοειδὲς—τῆς φύσεως—ὀρθῶς μὲν τραφὲν ἀνδπεῖον ἂν εἴη, μᾶλλον δ’ ἐπιταθὲν τοῦ δέοντος σκληπόν τε καὶ χαλερὸν γίγνοιτ’ ἄν, ὡς τὸ εἰκός).

3. ἐλέγομεν is more humble than λέγομεν —the idea is ‘used to call’ till you said it was wrong.

4. ἀκούετε δή : Socrates bespeaks ‘Attention!’ like a herald. ἀκούετε λεῴ (e.g. Ar. 551) was the usual way of beginning a proclamation at Athens. Cf. 20D, 32A, 201D.

5. ἄλλο τι . This, the shorter form of this particle of interrogation, is probably not a deliberate abbreviation for ἄλλο τι ἤ. Thus ἄλλο τι τοῦτο ἀληθές ἐστιν = ‘this is true—anything else?’ i.e. isn't it?, the words being thrown in parenthetically like nicht wahr and n'est ce pas.

6. ἐν τοῖσδε—οί=ον· τοῖσδεwould lead us to expect an enumeration of the cases: as it is, only examples are given. The effect of the Greek may be brought out by ‘in the following cases—that often for example, etc.’ With this punctuation it is needless to write (as Kroschel does) τοιοῖσδε for τοῖσδε .

8. γιγνώσκοντες ὅτι πονηρά ἐστιν : cf. 77C ἦ γὰρ δοκεῖ σοι, ὦ Μένων, γιγνώσκων τὰ κακὰ ὅτι κακά ἐστιν ὅμως ἐριθυμεῖν αὐτῶν; μάλιστα.

14. κἂν εἴ : see note on 328A above.

16. ὅμως δ’ ἂν κακὰ ἦν . So the MSS.: most recent editors (except Kral) read εἴη . The imperfect is used because the answer ‘No’ is expected and desired: would they be evil just because they give us pleasure? Certainly not. See Goodwin, p. 190, § 503. Schleiermacher wrongly rejected δέ after ὅμως : the ‘apodotic δέ ’ is frequent after ὅμως, ὡσαύτως, οὕτω and the like.

16. ὅ τι μαθόντα ποιεῖ κτλ . So the MSS. A variety of emendations have been proposed for μαθόντα , such as ἁμαρτόντα (Schleiermacher, followed by Heindorf), ἀμαθαίνοντα (Orelli), παρόντα (Hermann), παθόντα (Sauppe and others), while Kroschel doubts the entire passage ἦν—ὁπῃοῦν , saying ‘Totus autem locus ita comparatus est ut de eius emendatione desperandum esse videatur’. It would be natural to regard this particular passage as corrupt if this were an isolated example of the idiom in question; but—not to mention other authors—it occurs also in the following passages of Plato: 36B τί ἄξιός εἰμι παθεῖν ἢ ἀποτῖσαι, ὅ τι μαθὼν ἐν τῷ βίῳ οὐχ ἡσυχίαν ἦγον ; Euthyd.283E εἰ μὴ ἀγροικότερον—ἦν εἰπεῖν, εἶπον ἄν, σοὶ εἰς κεφαλήν, ὅ τι μαθὼν ἐμοῦ—καταψεύδει κτλ. ; ibid. 299A πολὺ μέντοι—δικαιότεπον ἂν τὸν ὑμέτεπον ρατέπα τύρτοιμι, ὅ τι μαθὼν σοφοὺς υἱεῖς οὕτως ἔφυσεν . In each of these cases the MSS. reading μαθών (not παθών ) should be retained. In the direct speech τί μαθών is used in an impatient question: τί μαθὼν τοῦτο ἐποίησας is ‘what ever made you think of doing this?’ (e.g. Ar. 1506 , 251, 826): cf. the idiom τί ληρεῖς ἔχων ; in which τί depends on ἔχων (see Kühner, II, 624). τί μαθών becomes in the indirect ὅ τι μαθών , but in every case in which this phrase occurs in Plato ὅ τι μαθών means not ‘why’ but ‘because’. The transition in meaning is sometimes explained as parallel to the use of οἶος, ὅσος, ὡς etc. for ὅτι τοιοῦτος, ὅτι τοσοῦτος, ὅτι οὕτως , so that ὅ τι μαθών = ὅτι τοῦτο μαθών ; but a simpler explanation is perhaps possible (see my edition of the Appendix II, p. 123). ὅ τι μαθών is an impatient ὅ τι just as τί μαθών is an impatient τί ; but as ὅτι and ὅ τι are in reality the same word (see on 333B , and ὅτι means ‘because’, ὅ τι μαθών comes also to mean ‘because’ (impatiently), the original interrogative force being lost. Similarly, in sentences like ληρεῖς ἔχων , the interrogation is dropped, and the participle alone survives, conveying the same sense of blame or impatience as in τί ληρεῖς ἔχων ; which is probably the form of sentence in which this idiom originated. Here μαθόντα is of course the nom. plur.; if (as we suppose) the original force of the participle is no longer felt in this idiom, there is nothing harsh in finding it in agreement with a neuter subject.

29. οὐκοῦν πάλιν ἂν—ἐροίμεθα . There is no sufficient reason for changing ἄν to αὖ or for omitting ἄν : ἄν points forward to the apodosis in l. 35 φαῖεν ἄν and is just as idiomatic as ἄν in ὥσπερ ἂν εἰ , e.g. in 311B above.

31. οί=ον τά τε γυμνάσια κτλ. II. 357C τρίτον δὲ ὁρᾷς τι—εἶδος ἀγαθοῦ, ἐν ᾧ τὸ γυμνάζεσθαι καὶ τὸ κάμνοντα ἰατπεύεσθαι καὶ ἰάτπευσίς τε καὶ ὁ ἄλλος χπηματισμός; ταῦτα γὰπ ἐρίρονα φαῖμεν ἄν, ὠφελεῖν δὲ ἡμᾶς, καὶ αὐτὰ μὲν ἑαυτῶν ἕνεκα οὐκ ἂν δεξαίμεθα ἔχειν, τῶν δὲ μισθῶν τε χάριν καὶ τῶν ἄλλων ὅσα γίγνεται ἀπ’ αὐτῶν .

32. τὰς στρατείας . Ar. X. 7. 1177 b . 9 οὐδεὶς γὰρ αἱρεῖται τὸ πολεμεῖν τοῦ πολεμεῖν ἕνεκα, οὐδὲ παρασκευάζει πόλεμον· δόξαι γὰρ ἂν παντελῶς μιαιφόνος τις εἶναι, εἰ τοὺς φίλους πολεμίους ποιοῖτο, ἵνα μάχαι καὶ φόνοι γιγνοιντο .

33. καύσεών τε—λιμοκτονιῶν . 521E πολλὰ ὑμᾶς καὶ κακὰ ὅδε εἴργασται ἁνήρ(sc. ὁ ἰατρός )— τέμνων τε καὶ κάων, καὶ ἰσχναίνων καὶ ρνίγων—ρικπότατα ρώματα διδοὺς καὶ ρεινῆν καὶ διψῆν ἀναγκάζων. λιμοκτονία (‘Hungerkur’, says Stallbaum) is part of the μικρὰ δίαιτα of III. 406D.

40. ἄλλων ἀρχαί : ‘dominions over others’—not ἄλλων πόλεων : cf. 452D αἴτιον ἅμα μὲν ἐλευθερίας αὐτοῖς τοῖς ἀνθρώποις, ἄμα δὲ τοῦ ἄλλων ἄρχειν ἐν τῇ αὑτοῦ πόλει ἑκάστῳ. Heindorf's conjecture ἄλλαι (i.e. ‘besides’) is not necessary, in view of 514A(quoted by Sauppe) ἢ χρήματα πολλὰ λαμβάνειν ἢ ἀρχήν τινων ἢ ἄλλην δύναμιν ἡντινοῦν .

43. τέλος . The philosophical sense of this word—as an end of action—is beginning to appear in Plato: cf. 499E ἆρα καὶ σοὶ συνδοκεῖ οὕτω, τέλος εἶναι ἁπασῶν τῶν πράξεων τὸ ἀγαθὸν κτλ. ;

44. ἀλλ’ ἢ ἡδονάς is Stephanus' emendation for ἀλλ’ ἡδονάς .

47. ὡς ἀγαθὸν ὄν : the copula is attracted to the predicate, as regularly: cf. below, 359D τὸ ἥττω εἶvαι ἑαυτοῦ ηὑρέθη ἀμαθία οὖσα .

64. εἴ με ἀνέροισθε : without the φαίην ἄν which we should expect. Socrates apostrophises the ἄνθρωποι as if they stood before him; the φαίην ἂν ἔγωγε of l. 66 is part of what Socrates says to the ἄνθρωποι .

68. ἐν τούτῳ : see on 310D above. πᾶσαι αἱ ἀποδείξεις means all the proofs that pleasure is good and pain evil.

69. ἀναθέσθαι suggests (as often) draughts-playing: a comparison which is worked up into a simile by Plato in VI. 487B-C.

71. ἢ τὸ κακὸν ἄλλο τι . B and T have ἤ before ἄλλο τι by a natural mistake: it is corrected in T by a later hand.

77. ἄνθρωπος : so the MSS., but ὁ ἄνθρωπος in line 80. Kroschel refers to the variation in 321D ἄνθρωπον—τῷ ἀνθρώπῳ —ἀνθρώπω?—ἄνθρωπος . Cf. below 355E

79. καὶ αὖθις αὖ λέγετε . We should of course expect λέγητε , but the second part of a logically subordinate clause in Greek tends to become an independent clause: see on 313A

10. ὑπὸ μὲν ἡδονῆς . The antithesis is not expressed by a δέ clause but in ἐκείνῳ δὴ—τοῦ ἀγαθοῦ below.

11. μετείληφεν : the subject is ἡδονή and τὸ ἀγαθόν is in apposition to ἄλλο ὄνομα. ἀντὶ τῆς ἡδονῆς means ‘in place of pleasure’, i.e. the name pleasure.

18. ἐν ὑμῖν : not ‘before your tribunal’, but ‘in you’, i.e. inside you, in your souls. The idea is of a conflict between the good and evil in the soul before you do the evil. ἐν ὑμῖν might have been ἐν αὐτῶ? , but the ὑβριστής now selects his interlocutors as his examples. The subtle reasoning which follows may be put thus: We do evil, knowing it to be evil, because we are overcome by good. But—since that which we do is evil—the good which overcomes is less worthy than the evil in us which it overcomes. ‘Less worthy’ (to overcome) means that ‘there is less of it’: to be overcome by good is therefore to choose less good than evil. The argument is extraordinarily ingenious but hardly sound— the flaw lies in substituting ‘the evil in us’ for ‘us’; it was not ‘the evil in us’, but ‘we’ who were overcome by good. See also Introduction, p. xxvi. The usual way of taking ἐν ὑμῖν as ‘before your tribunal’ or the like (cf. 464D) makes the false step much more serious—since it substitutes not ‘the evil in us’ but simply ‘the evil’ for ‘us’.

23. πλείω, τὰ δὲ ἐλάττω ᾖ . See on 330A

26. μεταλάβωμεν , i.e. ‘let us change and take the names pleasant and painful and apply them to, etc.’ Cf. below 356D ἐπί follows ὀνόματα as in τίθεσθαι ὄνομα ἐπί τινι and the like.

31. καὶ τίς ἄλλη ἀναξία . This—the MSS. reading—is undoubtedly right. Plato coins the word ἀναξία for ‘unworthiness’ to complete the parallel with D above: cf. ll. 17-21 οὐκ ἀξίων ὄντων νικᾶν—κατὰ τί δὲ—ἀνάξιά ἐστιν τἀγαθὰ τῶν κακῶν κτλ. Similarly in 105E, Plato coins ἀνάρτιος , in 106A ἄθερμος (rightly accepted, in spite of MS. authority, by editors) and ἄψυκτος (see MSS., but Wyttenbach's ἄψυχρος is—we think rightly—accepted by some editors) as pointed and convenient negatives to ἄρτιος, θερμός, ψυχρός . Cicero's translation ‘quae igitur potest esse indignitas voluptatis ad molestiam, nisi in magnitudine aut in longitudine alterius utrius posita?’ (Nobbe's p. 1313) shows that ἀναξία was read in his time: so in Ficinus' ‘quae vero alia inest ad dolorem indignitas voluptati quam excessus inter se atque defectus?’ The word ἀναξία hardly took root in Greek, though occasionally found in the writings of the Stoics, whose regular word for this notion is ἀπαξία. ἡδονῇ (now found to be the reading of some MSS.) for ἡδονή is due to Heindorf.

33. ταῦτα δ’ ἐστί : as Sauppe remarks, we should expect ταῦτα δ’ ἐστὶ τὸ μείζω—γίγνεσθαι , but the words are said as if for καὶ τίς ἄλλη ἀναξία κτλ. had been written καὶ πῶς ἄλλως ἀνάξια τὰ ἡδέα ρπὸς τὰ λυρηπά, ἀλλ’ ἢ ὑρεπβάλλοντα ἀλλήλων καὶ ἐλλείποντα ;

35. μᾶλλον καὶ ἧττον : this case is not given above (D-E) because the notion of intensity is more applicable to pleasure (and pain) than to good (and evil): cf. 24A ff. with 27E ἡδονὴ καὶ λύπη πέρας ἔχετον, ἢ τῶν τὸ μᾶλλόν τε καὶ ἧττον δεχομένων ἐοτόν; ναί, τῶν τὸ μᾶλλον, ὦ Σώκρατες .

εἰ γάρ τις λέγοι . So far we have reached this point. To be overcome by pleasure is to choose in place of lesser pleasures greater pains and the like. But here the element of time comes in: a man might fairly say—yes, but the pleasures are now, the pains remote. Socrates tries to show that this makes no real difference.

36. τὸ παραχρῆμα ἡδύ . When we are overcome by pleasure, we do the pleasant, not the painful: therefore the contrast is between the present pleasure and the remote consequences, whether pleasure or pain. For this reason it would be wrong to supply καὶ λυπηρόν after ἡδύ .

39. ἀλλ’ ὥσπερ ἀγαθὸς ἱστάναι κτλ. Cf. X. 602D ἆρ’ οὖν οὐ τὸ μετρεῖν καὶ ἀριθμεῖν καὶ ἱστάναι βοήθειαι χαριέσταται ρπὸς αὐτὰ ἐφάνησαν, ὥστε μὴ ἄπχειν ἐν ἡμῖν τὸ φαινόμενον μεῖζον ἢ ἔλαττον ἢ πλέον ἢ βαρύτερον κτλ. ;

40. καὶ τὸ ἐλλὺς κτλ. : in order that a near pleasure might count for more than a remote.

42. ἐὰν μὲν γὰρ ἡδέα κτλ. Cf. v. 733B ἡδονὴν βουλόμεθα ἡμῖν εἶναι, λύρην δὲ οὔθ’ αἱπούμεθα οὔτε βουλόμεθα, τὸ δὲ μηδέτεπον ἀντὶ μὲν ἡδονῆς οὐ βουλόμεθα, λύρης δὲ ἀλλάττεσθαι βουλόμεθα· λύρην δὲ ἐλάττω μετὰ μείζονος ἡδονῆς βουλόμεθα, ἡδονὴν δὲ ἐλάττω μετὰ μείζονος λύπης οὐ βουλόμεθα κτλ.

45. ἐάν τε τὰ ἐγγὺς κτλ. When once you have equated ‘near’ and ‘far’ (see above), then (but not till then) your final choice is not affected by the question of proximity in time. Theoretically, no doubt, this is right; but no man is so ἀγαθὸς ἱστάναι as to weigh ἐγγύς and πόρρω correctly; whence the saying ‘Let us eat and drink; for tomorrow we die’. ‘Near’ and ‘far’ can only be weighed aright on the supposition that man is immortal and will live hereafter under the same moral laws as rule us here: but of immortality there is no hint in this dialogue.

47. ταῦτα : i.e. τὰ ἡδέα .

51. ὅτε : the notion of time readily passes into that of cause, as in quoniam and cum.

52. φαίνεται ὑμῖν κτλ. Two examples are given: (1) size— of which παχέα and φωναὶ μεγάλαι (according to the Greek and Roman way of viewing sound) are special varieties, (2) number, i.e. τὰ πολλά . The specific varieties παχέα and φωναὶ μεγάλαι are not dealt with in the sequel. With what follows compare (besides X. 602Dcited above) 7Bff. and 126C ff., and with the present passage 42A and VII. 523C524C. The resolution by the intellect of such contradictory sense-perceptions is made the basis of Plato's scheme of higher education in the

55. αἱ ἴσαι is Heindorf's correction for ἴσαι .

57. μήκη—πράττειν : as we ἡδέα πράττομεν. πράττειν (with μήκη ) is explained by λαμβάνειν : see above on 314A

59. ἡ μετρητικὴ τέχνη . See Introduction, p. xxix.

60. ἢ ἡ τοῦ φαινομένου δύναμις. ἤis omitted in B and T. ‘The power of that which appears’ is its power to affect us.

61. ἐπλάνα καὶ ἐποίει κτλ. ἄνis not to be supplied. Plato is reverting to the illustration in C. The imperfect is idiomatic for ‘causes us, as we saw, to wander’, etc. (viz. because τὰ αὐτὰ μεγέθη seem to us both μείζω and ἐλάττω ): it is what Goodwin calls the ‘philosophic imperfect’: cf. 47D διαφθεροῦμεν ἐκεῖνο καὶ λωβησόμεθα, ὃ τῷ μὲν δικαίῳ βέλτιον ἐγίγνετο(becomes as we saw), τῶ? δὲ ἀδίκω? ἀπώλλυτο . By adding καὶ ἐν ταῖς πράξεσιν —σμικρῶν Plato treats his hypothesis that εὖ πράττειν is to select μεγάλα μήκη as a fact; he does so also in ἂν ἐποίησε κτλ. i.e. ‘would have made’, viz. if we had made use of it, but we did not. If the hypothesis were still treated as a hypothesis, we should have had ἂν ἐποίει (and not ἂν ἐποίησε ) to correspond to what would then have been ἂν (expressed or understood) ἐπλάνα καὶ ἐποίει .

ἄνω τε καὶ κάτω. ἄνω κάτω( 195C) and ἄνω καὶ κάτω ( 495A) are other forms of this proverbial phrase.

77. ἀριθμητική . In 451Bff. ἀριθμητική is said to be a science concerned with τὸ ἄρτιόν τε καὶ περιττόν, ὅσα ἑκάτερα τυγχάνει ὄντα , whereas the function which is here assigned to ἀριθμητική is there given to λογιστική : τὰ μὲν ἄλλα καθάπερ ἡ ἀπιθμητικὴ ἡ λογιστικὴ ἔχει· ρεπὶ τὸ αὐτὸ γάπ ἐστιν τό τε ἄπτιον καὶ τὸ ρεπιττόν· διαφέπει δὲ τοσοῦτον, ὅτι καὶ ρπὸς αὑτὰ καὶ ρπὸς ἄλληλα πῶς ἔχει πλήθονς (cf. ll. 71 ff. τὸ πλέον—τὸ ἔλαττον ἢ αὐτὸ πρὸς ἑαυτὸ ἢ τὸ ἕτερον πρὸς τὸ ἕτερον ) ἐπισκοπεῖ τὸ περιττὸν καὶ τὸ ἄρτιον ἡ λογιστική .

79. εἶεν—ἐπεὶ δὲ δή . The MSS. read ἐπειδὴ δέ , but after εἶεν Plato regularly uses δὲ δή in coming to the application of a train of reasoning or illustrations: cf. 312E εἶεν· ὁ δὲ δὴ σοφιστὴς περὶ τίνος δεινὸν ποιεῖ λέγειν ; and 311C εἶεν—παρὰ δὲ δὴ Πρωταγόραν κτλ.

87. εἰσαῦθις σκεψόμεθα . Siebeck, who asserts ( pp. 121 ff.) that this and similar formulas were intended by Plato to be a distinct promise of future dialogues (if not a reference to a later part of the same dialogue), finds here a reference to 283D ff., where the ‘measuring art’ is discussed and described as the βασιλικὴ τέχνη .

90. ἡνίκα ἡμεῖς—ὡμολογοῦμεν : 352B ff.

93. ἔφατε : in 352D The original ἤρεσθε has to be repeated in consequence of the introduction of this independent clause.

95. ἤρεσθε : in 353A

98. εἴπατε : but εἴπατον in 353A otherwise there is no change in the quotation. B and T have εἴπετε , but Stobaeus (who cites from τοσοῦτον in 357B to μεγίστη 357E in his ) reads εἴπατε . According to Rutherford ( p. 219) εἴπετε is not Attic, but this is doubtful, if any faith can be placed in MS. authority.

104. ἧς τὸ πρόσθεν . This—the reading of the best MSS.— has been unjustly rejected in favour of εἰς τὸ πρόσθεν ; but εἰς τὸ πρόσθεν can hardly be justified here. Sauppe refers to 339D ὀλίγον δὲ τοῦ ποιήματος εἰς τὸ πρόσθεν προελθών , where, however, εἰς τὸ πρόσθεν goes with προελθών , and in all the cases which he quotes there is a verb containing the idea of progress to go with the phrase, and so in Ar. 751 ἀλλ’ ἐς τὸ πρόσθε χρὴ παρεῖν’ ἐς τὴν πύκνα(where probably we should read παριέν’ for παρεῖν’ as in 43 πάριτ’ ἐς τὸ πρόσθεν ). Where there is no such verb (which is very rarely the case), the phrase means ‘with a view to what lies in front’, e.g. 132B οὐκοῦν τοσοῦτον μὲν ἡμῖν εἰς τὸ πρόσθεν πεπέρανται . It need hardly be said that the phrase cannot mean simply ἔμπροσθεν . On the other hand ἧς seems to be just what is needed: the construction in full would be καὶ οὐ μόνον ἐπιστήμης ( ἐνδείᾳ ), ἀλλὰ καὶ ( ταύτης τῆς ἐπιστήμης ἐνδείᾳ ) ἣν τὸ πρόσθεν ἔτι ὡμολογήκατε ὅτι μετρητική (sc. ἐστι : for the omission of ἐστί after ὅτι cf. below 359D ἀπεδείχθη ἄρτι ὅτι ἀδύνατον ): ‘and from lack, not merely of knowledge, but of the knowledge which you have further admitted above to be measuring knowledge’. The attraction of the relative extends to μετρητική also in spite of ὅτι : cf. 37B ὧν εὖ οἶδ’ ὅτι κακῶν ὄντωνfor τούτων ἃ εὖ οἶδ’ ὅτι κακὰ ὄντα ( ὅτι being used with the participle as in 481E: see note on ).

108. ἀμαθία ἡ μεγίστη : cf. III. 689A τίς οὖν ἡ μεγίστη δικαίως ἂν λέγοιτο ἀμαθία;—ταύτην τὴν διαφωνίαν λύρης τε καὶ ἡδονῆς ρπὸς τὴν κατὰ λόγον δόξαν ἀμαθίαν φημὶ εἶναι τὴν ἐσχάτην, μεγίστην δέ, ὅτι τοῦ πλήθους ἐστὶ τῆς ψυχῆς.

110. οὔτε αὐτοί : the verb hangs fire: we should expect Plato to continue οὔτε οἱ ὑμέτεροι παῖδες ἔρχονται κτλ. Stephanus thought προσέρχεσθε had fallen out after αὐτοί , Madvig ἴτε , while Kroschel suggests οὔτ’ ἐφοιτᾶτε αὐτοί . It seems probable that the text is corrupt; at all events no parallel has yet been adduced. The two cases quoted by Sauppe (Dem. πρὸς Ἄφοβον , § 54, and pseudo-Dem. πρὸς Τιμόθεον , § 52) are more than doubtful in respect of their text, which (as Sauppe quotes it) is not according to the MSS. Madvig's ἴτε is simplest, but the future seems out of place: perhaps ᾖτε should be read.

112. ὡς οὐ διδακτοῦ ὄντος : i.e. τοῦ μὴ ἡδονῆς ἡττᾶσθαι .

3. ὦ Ἱππία τε καὶ Πρόδικε . Here and in 358E and 359A ὦ has dropped out in the MSS. after preceding ‐ω . The effect of its omission would be to render the address unduly abrupt.

3. ὑμῖν . So the MSS. Sauppe's ἡμῖν is not necessary. ὑμῖν means all three sophists: and ὁ λόγος is not the discussion as a whole, but the speaking, viz. in answer (cf. δοῦναι καὶ δέξασθαι λόγον ). Throughout this chapter all the sophists reply: in the next Socrates makes use of their united admissions to destroy Protagoras alone.

7. διαἰρεσιν τῶν ὀνομάτων : see note on 314C

9. εἴτε ὁπόθεν καὶ ὅπως—ὀνομάζων . This mode of address insinuates that Prodicus was a θεῖος ἀνήρ , since a god was addressed in a similar way. See 400E ὥσπερ ἐν ταῖς εὐχαῖς νόμος ἐστὶν ἡμῖν εὔχεσθαι, οἵτινές τε καὶ ὁρόθεν χαίπουσιν ὀνομαζόμενοιand cf. 288A θαυμάσιά γε λέγετ’, ὦ ἄνδρες Θούριοι, εἴτε Χῖοι εἴθ’ ὁπόθεν καὶ ὅπῃ χαίρετον ὀνομαζόμενοι and 212C ὅ τι καὶ ὅπῃ χαίρεις ὀνομάζων, τοῦτο ὀνόμαζε. A similar effect is produced in 176Cby the words Σωκράτη δ’ ἐξαιρῶ λόγου , since there was a proverb τὸ θεῖον ἐξαιρῶ λόγου : cf. VI. 492E θεῖον μέντοι κατὰ τὴν παροιμίαν ἐξαιρῶμεν λόγου. The multitude of epithets given to gods is best illustrated by such a hymn as the Homeric Hymn to Ares, ll. 1 ff. Ἆρες ὑρεπμενέτα, βπισάπματε, χπυσεορήληξ, ὀβπιμόθυμε, φέπασρι. πολισσόε, χαλκοκορυστὰ κτλ. In the multitude of names the suppliant hopes to include the acceptable one.

11. πρὸς ὃ βούλομαι : emphatic, ‘with a view to my meaning’ )( πρὸς ὃ ὀνομάζω .

13. ἐπὶ τούτου . This usage can only be on the analogy of ἡ ἐπὶ Σάρδεων ὁδός and the like: actions are looked on as ways leading to a goal or destination or τέλος : cf. above 354C and D. To take the words (with Sauppe) as ‘in the case of this’, ‘in this domain’, would give a wrong meaning; the moral character of an act being determined by its end, it is καλόν only if its end is καλόν . The expression is however very curious, and perhaps unique. I once thought that ἅπασαι might conceal some present participle such as ἄγουσαι —agreeing with πράξεις : but the occurrence of ἁπάσας below in 359E shows ἅπασαι to be probably genuine here. The reasoning is this. Pleasure, the end, is good, pain evil; consequently all actions aiming at the end are καλαί , and therefore good (as usual ὠφέλιμος is but a synonym for ‘good’); therefore— as no one willingly selects evil rather than good—no one willingly does the worse (i.e. selects evil actions) when he might act better. As it is with πράξεις that bravery and cowardice are concerned, it is necessary for Plato to establish that ‘no one willingly seeks the worse’ is true of individual acts as well as of ultimate ends.

14. ἆρ’ οὐ καλαί . The words καὶ ὠφέλιμοι which follow οὐ καλαί in the MSS. are rightly rejected because they anticipate the identification in the next line.

17. ἢ ἃ ποιεῖ, καὶ δυνατά is Heindorf's emendation of the corrupt ἢ ἃ ἐποίει καὶ δύναται of the MSS.

18. ἥττω εἶναι αὑτοῦ κτλ. The phrases ἥττων αὑτοῦ and κρείττων ἑαυτοῦ are here substituted for ἥττων τῆς ἡδονῆς and κρείττων τῆς ἡδονῆς . Cf. the discussions in IV. 430Eff. and I. 626Eff., where κρείττω and ἥττω εἶναι ἑαυτοῦ are interpreted as the victory of the better part in us over the worse and conversely. As the worse part is the part which seeks ἡδονή , i.e. τὸ ἐπιθυμητικόν , the identification is just.

30. καλεῖτέ τι δέος . Heindorf's emendation for καλεῖτε δέος of the MSS. See note on 330C above.

31. προσδοκίαντινὰ—κακοῦ : a common definition; cf. 198B δέος γὰρ εἶναι προσδοκίαν μέλλοντος κακοῦ. φόβος in 1. 644C is defined as ‘expectation before pain’ ( πρὸ λύπης ἐλπίς ).

46. δέος, φόβος δ’ οὔ . Prodicus' distinction is just, though often dropped in practice: in φόβος ‘the physical agitation due to present danger ( ἡ παραυτίκα πτόησις , says Ammonius) is the leading idea’, in δέος the apprehension of evil to come ( κακοῦ ὑπόνοια ). See note on 12B, where Plato implicitly recognises the difference.

35. ἀλλὰ τόδε : sc. διαφέρει. οὐδὲν διαφέρει does not of course mean ‘there is no difference’, but ‘it doesn't matter’.

4. τότε : 330A ff.

7. τὸ ὕστερον : 349D ff.

17. ἠρόμην : 349E

18. καὶ ἴτας γ’, ἔφη . The recapitulation stops here. ἔφην ἐγώ two lines below is the usual ‘said I’ of narrated dialogue.

20. ἐπὶ τί : so MSS.: the second hand in B reads ἐπὶ τίνα , but cf. below, l. 26, where the question is repeated with ἐπὶ τί . For the common passage from singular to plural the editors compare 58C τί ἦν τὰ λεχθέντα καὶ πραχθέντα ;

28. ἐν οἷς σὺ ἔλεγες τοῖς λόγοις : see on 342B

31. ἐπειδὴ τὸ ἥττω—ἀμαθία οὖσα . The only proof given that we ‘encounter evils believing them to be evils’ was the phrase ἤττω εἶναι ἡδονῆς (for which ἥττω εἶναι ἑαυτοῦ is here substituted: see on 358C . We have demolished this proof by showing that the phrase means ἀμαθία , and we therefore infer that ἐπὶ ἃ δεινὰ ἡγεῖται εἶναι οὐδεὶς ἔρχεται ( δεινά being a subdivision of κακά ).