¹ Unless otherwise indicated, all references to the Republic are to the translation by G.M.A. Grube, revised by C.D.C. Reeve (Indianapolis, 1992).

² Cross, R.C. and Woozley, A.D., Plato's Republic: A Philosophical Commentary (London, 1964), 264–5CrossRefGoogle Scholar. Similarly, Annas, J., An Introduction to Plato's Republic (Oxford, 1981)Google Scholar notes the absence of an argument in the first ‘proof’ of Book 9 (305). On the other hand, against this verdict one could argue for the superiority of Book 9's first argument on the ground that it alone of the three seems to commend the just life as conventionally understood, whereas both of its successors undertake explicitly to prove the superiority of the philosophical life. Of Book 9's arguments it is thus the most straightforward in answering the fundamental challenge in Book 2, in which Glaucon and Adeimantus seem clearly to be lobbying for a defence of the conventional conception of the just life. Cf., for example, Sachs, D., ‘A fallacy in Plato's Republic’, Philosophical Review 72 (1963), 141–58CrossRefGoogle Scholar and Jang, I.H., ‘The problematic character of Socrates’ defense of justice in Plato's Republic’, Interpretation 24 (1996), 85–107Google Scholar.

³ Nettleship, R.L., Lectures on the Republic of Plato (London, 1955), 321Google Scholar.

⁴ Cf. Nettleship (n. 3), 328 n. 1; Cross and Woozley (n. 2), 265; Annas (n. 2), 311; Gosling, J.C.B. and Taylor, C.C.W., The Greeks on Pleasure (Oxford, 1982), 104–27CrossRefGoogle Scholar; Santas, G., Understanding Plato's Republic (Oxford, 2010), 216CrossRefGoogle Scholar.

⁵ Here I sidestep some important puzzles concerning Plato's practice of treating a part of the soul as parallel with a person dominated by that part of the soul; on these, see Moline, J., ‘Plato on the complexity of the psyche’, Archiv für Geschichte der Philosophie 60 (1978), 1–26CrossRefGoogle Scholar; Annas (n. 2), 306; and Lorenz, H., The Brute Within (Oxford, 2006), part 1CrossRefGoogle Scholar.

⁶ Cf. Erginel, M.M., ‘Inconsistency and ambiguity in Republic IX’, CQ 61 (2011), 493–520, at 51CrossRefGoogle Scholar, on whose view the pleasures of even the soul's lower parts consist in drawing conclusions from certain arguments.

⁷ The cognitive abilities belonging to the lower parts of soul have been the subject of considerable controversy. Cf. Annas (n. 2), 129–30; Cooper, J.M., ‘Plato's theory of human motivation’, HPhQ 1 (1984), 3–21Google Scholar; Irwin, T., Plato's Ethics (Oxford, 1995), 214–20, 282CrossRefGoogle Scholar; Price, A.W., Mental Conflict (New York, 1995)Google Scholar; Burnyeat, M.F., ‘Culture and society in Plato's Republic’, The Tanner Lectures on Human Values 20 (1999), 217–324Google Scholar; Bobonich, C., Plato's Utopia Recast: His Later Ethics and Politics (Oxford, 2002), especially 244–5CrossRefGoogle Scholar.

⁸ Here and throughout I make no claim that the Republic’s ten-book division is Plato's own. Any discussion of book divisions refers merely to the conventional ten-book division. Cf. Sedley, D.N., ‘Socratic intellectualism in the Republic’s central digression’, in Boys-Stones, G., Murr, D. El and Gill, C. (edd.), The Platonic Art of Philosophy (Cambridge, 2013), 70–89, at 70–4CrossRefGoogle Scholar.

⁹ Glaucon's challenge insists that Socrates compare ‘the most just [person] and the most unjust [person]’ (τόν τε δικαιότατον καὶ τὸν ἀδικώτατον) in order to examine ‘the extreme of injustice’ (ἐσχάτη […] ἀδικία) (360e1–361a3). This would seem to exclude the intermediately unjust types of timocrat, oligarch and democrat described in Book 8, particularly when one remembers that Socrates’ isomorphism between psychological types and political constitutions is yet to be made explicit.

¹⁰ An interpretation along these lines is developed and defended in Shaw, J.C., ‘Poetry and hedonic error in Plato's Republic’, Phronesis 61 (2016), 373–96CrossRefGoogle Scholar.

¹¹ Socrates’ criticism at 505c5–6 of the view that pleasure is identical with the good would seem to deny pleasure to the higher parts of the soul, and to reason in particular.

¹² That pleasure is not restricted to the appetitive part is perhaps implied by Socrates’ account of temperance at 430e–431d. On this picture, the temperate agent's better desires—those ‘directed by calculation in accordance with understanding and correct belief’ (αἳ δὴ μετὰ νοῦ τε καὶ δόξης ὀρθῆς λογισμῷ ἄγονται)—restrain his baser desires. The natural home for these better desires is presumably the rational part, in which case it follows that the rational part has desires. Moving on to civic temperance, Socrates links desire with pleasure when he describes the best city as one ‘in control of itself and of its pleasures and desires’ (κρείττω ἡδονῶν τε καὶ ἐπιθυμιῶν καὶ αὐτὴν αὑτῆς). It is also clear earlier in Socrates’ division of the soul that each psychic part has desires (435c4–441c6). However, it seems that we must wait until Book 9 for Socrates to make explicit the view that each psychic part has distinctive pleasures, and moreover that each part also spurns the pleasures associated with the other parts. My thanks to an anonymous reader for calling my attention to these passages.

¹³ Cf. Scott, D., ‘Metaphysics and the defence of justice in the Republic’, Proceedings of the Boston Area Colloquium in Ancient Philosophy 16 (2000), 1–20, at 18Google Scholar, who refers to Socrates’ strategy in this argument (among others in Book 9) as a ‘psychological revisitation’ of Book 4's tripartition.

¹⁴ Pappas, N., Routledge Philosophy Guidebook to Plato and the Republic (New York, 1995), 161Google Scholar.

¹⁵ Cf. Nettleship (n. 3), 319–20.

¹⁶ Plato makes a point of telling us at 576b8 that Socrates’ main interlocutor shifts from Adeimantus to Glaucon, just before Socrates begins the first argument of Book 9. This is perhaps because the tyrannical pain and suffering that is about to be on display is meant as a counterweight specifically to Glaucon's insistence way back in Book 2 at 361e2–362a2 that it is the truly just man who ‘will be whipped, stretched on a rack, chained, blinded with fire, and, at the end, when he has suffered every kind of evil, he'll be impaled’. Cf. Scott (n. 13), 15, who similarly notes the significance of the shift from Adeimantus to Glaucon at this moment in the dialogue.

¹⁷ For the general observation that Plato in Book 9 is keener to condemn the bad condition than to argue concretely for the good condition, see Riel, G. Van, Pleasure and the Good Life: Plato, Aristotle, and the Neoplatonists (Leiden, 2000), 16CrossRefGoogle Scholar.

¹⁸ The second and third premises of this argument seem to be presuppositions of Book 9's argumentative section generally. Socrates asserts something very much like the second premise in his debate with Callicles in the Gorgias: ‘Tell me: don't you think that those who do well have the opposite experience of those who do badly?’ (εἰπὲ γάρ μοι, τοὺς εὖ πράττοντας τοῖς κακῶς πράττουσιν οὐ τοὐναντίον ἡγῇ πάθος πεπονθέναι;) (495e1–2) (transl. Woodhead, W.D., in Hamilton, E. and Cairns, H. [edd.], Plato: The Collected Dialogues [Princeton, 1961], 229–307Google Scholar). And Thrasymachus’ remark way back in Book 1 seems to imply the third premise: ‘You'll understand this most easily if you turn your thoughts to the most complete injustice, the one that makes the doer of injustice happiest and the sufferers of it, who are unwilling to do injustice, most wretched. This is tyranny’ (344a2–5). This is presumably written into the debate's terms when Glaucon renews Thrasymachus’ case in Book 2; Glaucon declares tyranny and Socrates’ kingly city to be ‘total opposites’ (πᾶν τοὐναντίον) at 576d3. This opposition between tyranny and kingly rule is of course a theme running through Books 5–7.

¹⁹ Throughout I will refer to the profit-lover and to his psychic counterpart, the appetitive part of soul, jointly as ‘the profit-lover’, and likewise to the honour-lover and to his psychic counterpart spirit jointly as ‘the honour-lover’.

²⁰ Thrasymachus identifies consummate injustice with tyranny at 344a4–5. Glaucon characterizes his challenge as a renewal of Thrasymachus’ case at 358b1–d2, and he consistently frames the debate as a competition between the supremely (ἔσχατον) just and unjust lives.

²¹ The story of the ring of Gyges appears at 359c4–360b2, according to which a lowly shepherd uses his newfound power of invisibility to usurp the throne. The crux of Adeimantus’ case that the unjust man outdoes the just man where the gods are concerned appears at 363e4–366b2.

²² Hence I reject the translation of ἐξ ἁπάντων τούτων in causal terms.

²³ Cf. 576c8: ‘And will not the relations between the cities with respect to virtue and happiness be the same as those between the men?’ (οὐκοῦν, ὅτι πόλις πρὸς πόλιν ἀρετῇ καὶ εὐδαιμονίᾳ, τοῦτο καὶ ἀνὴρ πρὸς ἄνδρα;).

²⁴ Such a debate would look suspiciously like the exchange between Socrates and Polus at Grg. 470c1–472d5, where Polus appeals to real tyrants Archelaus of Macedonia and the Great King of Persia to refute Socrates’ claim that tyrants are miserable. Socrates calls Polus’ style of refutation ‘worthless’ (οὐδενὸς ἄξιός ἐστιν), contrasting it with his own.

²⁵ While it is true that σκοπέω is sometimes translated as ‘contemplate’, even in these uses it generally represents a contemplation of particulars rather than universals, for which θεωρέω is more common. The Grube and Reeve (n. 1) translation of σκοπέω as ‘consider’ obscures its connection with sight and the grasp of particulars. The word reappears at 572a1. They translate ἄθρει simply as ‘see’.

²⁶ Grube and Reeve (n. 1) translate ἑστιάσας as ‘feast’, which renders the metaphor gustatory rather than visual, but none the less preserves the link with the sense faculties and their associated pleasures. They supply ‘speculations’ for σκέψεων, which is linked at least etymologically to vision.

²⁷ Cf. Cross and Woozley (n. 2), who deny that the first argument of Book 9 ‘could count as philosophical argument’ at all, but who claim that it nevertheless ‘merits attention as a remarkably graphic and penetrating picture of morbid psychology’ (264).

²⁸ Here I sidestep an ambiguity in the precise hedonic mistake being made: Socrates’ claim is either that i) what the spectator experiences is a genuine pleasure, but she overestimates its size; or that ii) the pleasure experienced is not really a pleasure at all, but is instead just the removal of a pain. My point seems to find support on either reading. Cf. 583b–588a, discussed in Shaw (n. 10), 376–7. See also Shaw (n. 10), 383: ‘It is plausible to think that our experience of tragedy involves some non-pleasures. Watching another endure misfortune can make us aware that we are free from the pain they feel. By contrast with their pain, we then feel our own freedom from pain as if it were pleasure.’

²⁹ At 605a7–b4 Socrates claims that tragic imitation, like painting, gratifies an inferior part of the soul so as to win favour with the multitude, and moreover that it depicts a peevish and multifaceted character (ἀγανακτητικόν τε καὶ ποικίλον ἦθος) because this type is easy to imitate. He then compares the effect of this imitation on the soul's better part to the effect of putting a scoundrel in charge of the city. Cf. 603c3–5, where Socrates characterizes imitative poetry generally as imitating ‘human beings acting under compulsion or voluntarily, who, as a result of these actions, believe they are doing either well or badly, and so experience either pain or enjoyment in all these situations’. Moreover, at 603e6–604ab4 Socrates seems to propose that pity and grief, key emotions in our experience of tragedy, belong to appetite but are opposed by spirit and reason.

³⁰ On the view Socrates airs in Book 10, tragedy works its magic only on an audience who admires the tragic hero; otherwise, it is comedy. As our argument explains, the tyrant exploits the appetites let loose in a democracy by setting himself up as the people's champion. Socrates calls attention to the public's admiration for the tyrant at 576c1–3. That this characterizes the first argument's target audience explains its rhetoric of overturning a favourable opinion of the tyrant with a clearer vision of the tyrannical soul.

³¹ Cf. 505c5–6.

³² Others who seem to read the argument this way are Irwin, T., Plato's Moral Theory (Oxford, 1977), 338 n. 62Google Scholar and Annas (n. 2), 307–9.

³³ Cf. 441a1–2 and 441e3–4.

³⁴ Cf. Gosling and Taylor (n. 4), who characterize the second argument of Book 9 straightforwardly as an ‘argument from authority’ (at 100). Similarly, Russell, D.C., Plato on Pleasure and the Good Life (Oxford, 2005)CrossRefGoogle Scholar argues that this argument takes an ‘authoritative perspective on the sort of life that is most worth living’ (at 126).

³⁵ While the first argument also refers to competent judges (ἀξιῶν κρίνειν) (577a1), this appears to be an allusion to the judging of choruses. Glaucon makes this connection explicit a little later at 580b4–5. Add this to the last section's evidence for the dramatic form of Book 9's first argument.

³⁶ For a close reading of the argument, see Murphy, N.R., The Interpretation of Plato's Republic (Oxford, 1951), 207Google Scholar; Cross and Woozley (n. 2), 266–9; Gosling and Taylor (n. 4), 97–128; Frede, D., ‘Rumpelstiltskin's pleasures: true and false pleasures in Plato's Philebus’, Phronesis 30 (1985), 151–80CrossRefGoogle Scholar; Frede, D., ‘Disintegration and restoration: pleasure and pain in Plato's Philebus’, in Kraut, R. (ed.), The Cambridge Companion to Plato (Cambridge, 1992), 425–63, at 435–7 n. 6, 493–520; and Shaw (n. 10)CrossRefGoogle Scholar.

³⁷ For alternative explanations of Socrates’ ‘greatest and most decisive’ remark, see Gosling and Taylor (n. 4), 104 and Scott (n. 13), 16–20.

³⁸ I would like to thank Rachel Barney, Jennifer Whiting, Philip Clark, Brad Inwood and Raphael Woolf for helpful comments on an earlier draft of this paper; Tad Brennan, Jacob Klein and the other participants of the 2014 Ancient Philosophy Workshop at Colgate University for generous feedback on an earlier draft; and an anonymous reviewer, whose report improved this article on several fronts.