In defending the view that justice is the advantage of the stronger, Thrasymachus puzzlingly claims that rulers never err and that any practitioner of a skill or expertise (τέχνη) is infallible. In what follows, Socrates offers a number of arguments directed against Thrasymachus’ views concerning the nature of skill, ruling, and justice. However, both Thrasymachus’ views and Socrates’ arguments against Thrasymachus’ views have frequently been misunderstood. In this paper, I clarify Thrasymachus’ views concerning the nature of skill and ability, reconstruct Socrates’ arguments against Thrasymachus’ views concerning skill and justice, and argue that Socrates’ arguments are better than often supposed.
The doctor never hesitates to claim divine omniscience, nor to clamour for laws to punish any scepticism on the part of laymen … On the other hand, when the doctor is in the dock, or is the defendant in an action for malpractice, he has to struggle against the inevitable result of his former pretences to infinite knowledge and unerring skill.George Bernard Shaw The Doctor’s Dilemma
In Republic book 1, Thrasymachus claims that justice is the advantage of the stronger. After being shown by Socrates that several of his views are inconsistent, Thrasymachus evades Socrates’ reductio by claiming that no ruler and no practitioner of a skill (τέχνη) ever errs (Rep. 340e2-3). Socrates then proceeds to offer several arguments directed against Thrasymachus’ views (Rep. 341a5-354c3). There has been an enormous amount of disagreement over the nature and consistency of Thrasymachus’ definitional or semi-definitional remarks concerning the nature of justice, but commentators widely agree that: (a) Thrasymachus’ view that the practitioners of a τέχνη are infallible is groundless and may be rapidly dismissed;1 and (b) that Socrates’ subsequent arguments against Thrasymachus are ‘weak and unconvincing to an amazing degree’.2
In this paper, I argue that claims (a) and (b) are mistaken. I show that Thrasymachus’ claims about τέχνη are not groundless and that Socrates’ arguments against Thrasymachus are significantly stronger than usually thought. To this end, I first (Section 2) offer a brief, critical reconstruction of the initial argument between Thrasymachus and Socrates which leads Thrasymachus to claim that τέχνη is infallible. I then (Section 3) argue that Thrasymachus’ views concerning the infallibility of τέχνη are neither groundless nor should they be rapidly dismissed. By carefully examining what Thrasymachus says and understanding his views as part of a broader intellectual current which took each τέχνη to be a complete and perfected area of rational expertise, I show how Thrasymachus’ views follow from certain more intuitive assumptions concerning what it is to have certain kinds of ability whose success is not to be credited to luck. Plato’s Socrates does not dismiss such views and dialectically appeals to them elsewhere. Finally (Section 4), I turn to Socrates’ arguments against Thrasymachus. These include: an argument that each τέχνη is directed towards the advantage of its object; a discussion of wage-earning; an argument that a genuine practitioner of a τέχνη does not outdo or overreach (πλεονεκτεῖν); and a pair of arguments that justice is required for successful action and proper functioning. I offer a much-needed clarification of these arguments and argue that appreciating the dialectical nature of Socrates’ arguments and how they appeal to Thrasymachus’ claims about τέχνη (while also responding to Thrasymachus’ objections to Socrates’ own arguments) reveals that they are significantly stronger than typically thought.
2 The Initial Argument between Socrates and Thrasymachus
In Republic 1, Socrates initially discusses the nature of justice with Cephalus and Polemarchus and argues that it is not the function (ἔργον) of the just person to harm either a friend or anyone else, but of his opposite, the unjust person (Rep. 335d12-13) because ‘in no case is it just to harm anyone’ (335e5-6). At this point, Thrasymachus interrupts the conversation and begins expounding his views about justice, claiming that ‘justice is nothing other than the advantage of the stronger’ (τὸ τοῦ κρείττονος συµφέρον, 338c2-3). In what follows, Thrasymachus puts forward several claims about justice and, by means of making explicit Thrasymachus’ commitment to several additional claims, Socrates offers a reductio which lands Thrasymachus in a contradiction.
It is difficult to determine Thrasymachus’ views about justice precisely.3 As a result, it is also difficult to offer an entirely uncontroversial reconstruction of Socrates’ reductio. However, the argument begins with Socrates being willing to grant that justice is some kind of advantage, but being unsure about whether it is the advantage of the stronger (339b5-7). Upon being challenged by Socrates to explain what he means by ‘the stronger’, Thrasymachus says that he has in mind political strength (Rep. 338c5-d10) and that by ‘the stronger’ he means whoever has political power over one or whoever is in charge and has all the political power (and is thus the strongest), i.e. whoever in actual fact rules the πόλις (Rep. 338d9, 339a1-2; cf. Leg. 714c6-d7; Grg. 488b2 ff.).
Thrasymachus proceeds to articulate the view that in each πόλις a person’s actions may be called ‘just’ insofar as they are advantageous to the rulers of that person’s πόλις or to those who are stronger than that person, and that their actions may be called ‘unjust’ insofar as they are disadvantageous (Rep. 338e1-339a4). With only slight simplification, we may say that Thrasymachus’ first relevant claim is that: (1) an action is just if and only if it is advantageous to the ruler of the πόλις in which the action was performed.4 Socrates replies that he will attempt to determine whether (1) is in fact true (Rep. 339a5-6) and proceeds to secure Thrasymachus’ explicit agreement to the claim that obedience to rulers is just (339b9-11, c10-12, d5-10, e4). This is the second substantive claim relevant for the reductio: (2) if an action is or involves obeying a ruler, then that action is just.
Having confirmed that Thrasymachus is committed to this second claim, Socrates asks Thrasymachus whether rulers are incapable of erring (ἀναµάρτητοι) or whether they might err (ἁµαρτεῖν, Rep. 339c1-2). Thrasymachus allows that rulers might make mistakes (ἁµαρτεῖν, 339c3; διαµαρτάνειν, 339d7). The third relevant claim then is that: (3) rulers may err. Presumably this means that, in attempting to perform an action or bring about a certain result, a ruler may nonetheless fail to perform that action or bring about the relevant result. For instance, rulers might fail to establish laws correctly (ὀρθῶς, 339c4-5) by attempting to enact laws which benefit themselves but in fact enacting laws which do not benefit themselves (339c7-8, d5-9).
To recap, Socrates has established that Thrasymachus accepts the following three claims:
An action is just iff it is advantageous to the ruler(s) of the πόλις in which the action is performed.
If an action is or involves obeying a ruler, then that action is just.
Rulers may err (e.g. in attempting to enact laws which benefit themselves they may enact laws which do not benefit themselves).
Socrates then proceeds to put the finishing touch on his reductio: ‘then according to your account, it is not only just to act to the advantage of the stronger, but also to the opposite, to what is not to his advantage’ (Rep. 339d1-3). Simply put, (3) allows that rulers may err and so may enact laws which are not beneficial to themselves. Accordingly, suppose that the rulers do err by enacting a law the following of which is not beneficial to themselves. Instances of obeying the rulers by following that law will—per (2)—be just and yet, not being beneficial to the rulers, will—per (1)—not be just (cf. 339e1-5).
Socrates emphasises that the contradiction emerges from what Thrasymachus has himself accepted and briefly restates the argument to the applause of Polemarchus (Rep. 339d5-340b5). Cleitophon is less impressed (340a3-4) and suggests that when Thrasymachus said ‘the advantage of the stronger’ he had in mind those things which the stronger believed to be to their advantage (340b6-8). Some commentators take Cleitophon’s suggestion to be a good one,5 but Thrasymachus dismisses it. Instead, Thrasymachus rejects (3) and offers his most precise account (e.g. 340e1-341a4, 341b8-c1, 342b6-7), in which he claims that ‘no craftsman ever errs’ (οὐδεὶς τῶν δηµιουργῶν ἁµαρτάνει, 340e2-3). Every τέχνη is such that practitioners of that τέχνη do not commit errors in practising their craft. Assuming that ruling is a craft (τέχνη), rulers are also thereby incapable of errors in ruling or enacting laws to their own advantage (340e8-341a4). Thus, Thrasymachus rejects (3) and instead embraces:
(3*) Rulers may not err.
By accepting (3*) instead of (3), Thrasymachus escapes Socrates’ reductio.
Although interpretations of Socrates’ initial encounter with Thrasymachus differ significantly, commentators are almost unanimously united in regarding Thrasymachus’ claim that no craftsman errs as deeply problematic. Many think that Thrasymachus’ so-called ‘idealisation’ of rulers and practitioners runs counter to his ‘realistic’ views concerning justice or else simply runs afoul of reality,6 and several take it to manifest a broader incoherence.7 Julia Annas accurately captures several of the main worries when she writes (1981, 43):
This is a very counterintuitive position, and Thrasymachus is probably only forced into saying this about skills in general because he finds it plausible as a position to hold about the stronger in any situation. He is thinking of the obviously true point that the man who has the upper hand cannot afford to make mistakes, or he will soon cease to have the upper hand. He saves the consistency of his position by a verbal move that makes this true of all rulers and all practitioners of any skill. But this flouts our beliefs about doctors, rulers, etc.
Thrasymachus’ claims about the infallibility of practitioners are thought to be either highly implausible and groundless (when construed as the claim that practitioners of a skill do not mistakes) or else tautologically true (when construed as the claim that perfect practitioners of a skill do not mistakes).8 Either way, Thrasymachus’ claims about the infallibility of τέχνη have typically been rapidly dismissed as a feeble response to Socrates’ criticisms.
3 Thrasymachus’ Unerring Skill
Thrasymachus’ claims are neither ad hoc nor groundless, but they do require an explanation (which they have hitherto not received).9 In order to better understand Thrasymachus’ views, we should carefully examine what Thrasymachus says (Rep. 340d2-e5, trans. Reeve):
 When someone makes an error in the treatment of patients, do you call him a doctor in regard to that very error (κατ’ αὐτὸ τοῦτο ὃ ἐξαµαρτάνει)? Or when someone makes an error in calculation, do you call him a calculator in regard to that very error in calculation (ἢ λογιστικόν, ὃς ἂν ἐν λογισµῷ ἁµαρτάνῃ, τότε ὅταν ἁµαρτάνῃ, κατὰ ταύτην τὴν ἁµαρτίαν)?  I think that we express ourselves in words that, taken literally, do say that a doctor or a calculator, or a grammarian errs.  However, I think that each of these, insofar as he is what we call him, never errs (τὸ δ’ οἶµαι ἕκαστος τούτων, καθ’ ὅσον τοῦτ’ ἔστιν ὃ προσαγορεύοµεν αὐτόν, οὐδέποτε ἁµαρτάνει). Accordingly, according to the precise account (κατὰ τὸν ἀκριβῆ λόγον)—and you are a stickler for precise accounts—no craftsman ever errs (οὐδεὶς τῶν δηµιουργῶν ἁµαρτάνει).  For it is when his knowledge abandons him that he who goes wrong goes wrong—when he is not a craftsman (ἐπιλειπούσης γὰρ ἐπιστήµης ὁ ἁµαρτάνων ἁµαρτάνει, ἐν ᾧ οὐκ ἔστι δηµιουργός). So that no craftsman, wise man, or ruler makes a mistake then when he is a ruler (ὥστε δηµιουργὸς ἢ σοφὸς ἢ ἄρχων οὐδεὶς ἁµαρτάνει τότε ὅταν ἄρχων).
In , Thrasymachus suggests that, when a practitioner of a τέχνη, e.g. a doctor, makes a mistake, the practitioner is not a practitioner with respect to that mistake. This concerns what an ability or capacity (δύναµις) is responsible for and what is constitutive of the actions produced by a practitioner’s ability or capacity.10 That is to say, a τέχνη of φ-ing—or the ability (δύναµις) constitutive of such a τέχνη—manifests itself only in φ-ing. If one’s action does not amount to φ-ing, then that action does not count as a manifestation of one’s τέχνη or ability (cf. Rep. 341c10-d4). Just as my cooking is not a manifestation of my jumping ability, neither is my tripping in a failed attempt to perform a jump (assuming that the tripping in question does not amount to jumping).11
In , Thrasymachus recognises that we do often say things like ‘the doctor made a mistake’, but in  he claims that every practitioner insofar as he or she is a practitioner never errs. ‘Qua’ locutions and (in the relevant contexts) ‘insofar as’ locutions are not straightforward to interpret, but on one possible reading, the claim that an individual may nor err qua practitioner simply repeats the claim made in  so that (e.g.) Dr Smith may botch a surgery but may not do so qua doctor.12 What has been said thus far might allow that while erring, Dr Smith may nonetheless retain their medical τέχνη and remain a genuine doctor (even though the relevant failed actions are not a manifestation of the relevant τέχνη).
However, in  Thrasymachus says: ‘For it is when his knowledge abandons him that he who goes wrong goes wrong—when he is not a craftsman. So that no craftsman, wise man, or ruler makes a mistake then when he is a ruler.’ Here, Thrasymachus is not merely saying that errors are not to be considered manifestations of the relevant capacity or τέχνη. Instead, Thrasymachus is claiming that so long as the practitioner has their knowledge13 then the practitioner will not err.14
Whereas  and  suggest that a practitioner’s use of a τέχνη is incompatible with error (but do not seem to rule out that the possession of a τέχνη is incompatible with error),  suggests that a practitioner’s possession of a τέχνη is incompatible with error and that errors indicate that the relevant τέχνη has abandoned the practitioner at the time of their error.15 That is to say, if a person errs (i.e. makes an attempt to φ which does not result in successfully φ-ing), then they did not possess the relevant τέχνη or the relevant ability (constitutive of a τέχνη) at the time of their error (cf. ὅταν, Rep. 340c7).16 Thus, with regard to those abilities which constitute a τέχνη, it seems that Thrasymachus holds or assumes the following view:
(TECHNICAL ABILITY): if S has the ability (constitutive of a τέχνη) to φ at t, then if S were to attempt to φ at t, then S would φ at t.17
Such a view—which claims that it is a necessary condition of an agent having an ability (constitutive of a τέχνη) that the agent can successfully perform the relevant action when they attempt to—is not entirely without some intuitive appeal,18 but what matters for our purposes is how it explains Thrasymachus’ views.
Imagine a case of the following sort. Suppose that Diana is a skilled archer and that the ability to shoot arrows is constitutive of her relevant τέχνη. Suppose also that Diana takes part in an archery competition and that, in the competition, Diana attempts to shoot an arrow but fails to shoot and instead drops the arrow, perhaps as a result of someone hitting her arm just as she was preparing to loose the arrow, or as a result of a momentary bout of fever, or as a result of being distracted by a sudden altercation in the audience. Regardless of the circumstance, Diana had attempted to shoot an arrow and yet her action did not amount to shooting an arrow. The action should not—as per  to —be considered as a manifestation of her archery ability. Moreover, according to  and (TECHNICAL ABILITY), Diana had thereby revealed that she did not—at the time of her failure—have the ability to shoot arrows. Whatever ability Diana might have had at the time at which she won archery competitions in the past, and whatever ability she might have at some point in the future, this ability had, to echo Thrasymachus, abandoned her at the time of her failure.19 Since the ability is constitutive of the skill, Diana is not strictly speaking an archer when she fails her shot.
Thrasymachus’ views should not be rapidly dismissed for two closely related reasons. On the one hand, (TECHNICAL ABILITY) is assumed by several ancient thinkers and admits of explanation. On the other hand, Plato’s Socrates appeals to (TECHNICAL ABILITY) elsewhere, notably in the Charmides and the Euthydemus,20 in a manner which suggests that his audience (in the relevant contexts) would have found the view attractive. Let us consider these two points in turn.
First, it deserves greater attention that one finds views of τέχνη among some of the Hippocratic authors which are very similar to those of Thrasymachus in the Republic.21 According to these thinkers, a τέχνη was a complete and perfected area of rational expertise which guarantees success independently of luck or circumstance in such a way that a genuine doctor (arithmetician, etc.) infallibly brings about certain results (and anyone who botches a diagnosis or miscalculates does not deserve the title). Thus, for instance, in contrast to thinkers like Isocrates—who argued that τέχναι do not grant the abilities (δυνάµεις) often promised by the sophists (In Sophistas 10-11, 19; Antidosis 147-8) and that τέχναι were fallible, imperfect and vulnerable to luck and circumstance (In Soph. 3, 11, 13: cf. Hippocrates, De Vetere Medicina [VM] 1, 7, 9; Antid. 184-5, 193-4, 271-6)—22 the authors of the Hippocratic texts De Arte and On Places in Man reflect an ‘infallibilist’ strand within ancient medical thought of precisely the sort Isocrates criticised.23 Such ‘infallibilist’ thinkers took medicine to be a genuine τέχνη because it was perfect, completely discovered, free from error, independent from luck, and because it guaranteed success to its practitioners. In discussing whether medicine depends upon luck (cf. Isocrates, Antid. 197), and emphasising that medical successes cannot be credited to luck, the author of On Places in Man offers one of the more explicit surviving articulations of these views (Loc. Hom. 46, trans. Craik 1998):
 In my view, medicine has been completely discovered (ἤδη ἀνευρῆσθαι ὅλη), medicine of this kind which teaches in each case both its inherent character and proper treatment (καιρός). The man who has this understanding of medicine least depends on luck (τύχη); but whether with or without luck his actions would succeed (ἀλλὰ καὶ ἄνευ τύχης καὶ ξὺν τύχῃ εὖ ποιηθείη ἄν). For the whole of medicine has advanced, and its finest established techniques seem to have very little need of luck.  For luck rules itself and is ungovernable, and it is not its way to come in response to one’s wish. But knowledge is governable and successful when the one with knowledge wishes to use it (ἡ γὰρ τύχη αὐτοκρατὴς καὶ οὐκ ἄρχεται, οὐδ᾿ ἐπ᾿ εὐχῇ ἐστιν αὐτὴν ἐλθεῖν· ἡ δ᾿ ἐπιστήµη ἄρχεταί τε καὶ εὐτυχής ἐστιν, ὁπόταν βούληται ὁ ἐπιστάµενος χρῆσθαι).
In , the author claims that medicine is a complete and perfect τέχνη (in much the same way that moderns might speak of an ideal or perfect physics). It requires no further discoveries, it is free of errors, and its success is independent of—or at least highly resistant to—luck (τύχη). Thus, even if luck is against him, the practitioner of medicine nonetheless acts well and succeeds (εὖ ποιέω). In , the author contrasts luck with the medical τέχνη or ἐπιστήµη. Whereas luck is outside of human control, knowledge is within human control. The doctor’s knowledge is always successful (εὐτυχής).24 That is to say, because medicine is a complete and perfect science, it guarantees success. Whenever (ὁπόταν) the possessor of knowledge wishes or decides (βούληται) to act or put his knowledge to effect, it will indeed successfully come into effect.25 Because medical science is completely discovered, it will prescribe the correct treatment. Thus, if a doctor were to fail in his endeavour, that would reveal that he lacked the medical τέχνη (or the ability constitutive of it) at that moment. In emphasising that τέχνη is independent from luck and circumstance and that its successes cannot be credited to luck, the author of On Places in Man makes τέχνη infallible in the same manner as Thrasymachus and he assumes—and in fact comes close to articulating—(TECHNICAL ABILITY).26
The author of De Arte provides a similar account. He attempts to defend medicine against the accusation that it has no efficacy or that any successes claimed by medical practitioners are in fact due to luck.27 In so doing, he argues that spontaneity (τὸ αὐτόµατον) does not exist, that no medical successes should be credited to luck (De Arte 6),28 and—like the author of On Places in Man—claims that medicine’s cure of diseases is infallible or free from error (ἀναµάρτητος, De Arte 9, 13; cf. Rep. 339c1, 340d8-e1). We might think that such views are utterly implausible, but in discussing the success guaranteed to genuine doctors and attempting to explain away apparent failures,29 the author of De Arte (very likely a sophist)30 stresses that one must give attention to the doctor’s proper task (ἐργασία) and its perfection or end (De Arte 8).31 He claims that it is foolish to equate the patient not recovering with an error on behalf of the doctor because, even in the case of curable diseases, the doctor’s activity is not constituted by the patient recovering, but by correctly diagnosing the illness and ‘by giving proper orders’, i.e. prescribing the correct regimen for the patient to follow (De Arte 7; cf. Plato, Pol. 260a4-7). Faultless diagnoses and prescriptions can thus be judged successful even if the patient does not recover because the patient’s recovery is not constitutive of the doctor’s successful action.
The author supposes that success requires only ability and not also luck or favourable circumstance (and shapes his construal of ‘success’ accordingly), and the view that τέχνη is infallible seems to stem largely from these assumptions about the perfection and completeness of τέχνη and its independence from luck or circumstance. The extent to which the historical Thrasymachus of Chalcedon should be associated with this infallibilist strand of ancient thought is not clear, but such views were evidently in the air and are very similar to those of Plato’s Thrasymachus in the Republic.32
Secondly, it deserves attention that Plato’s Socrates does not rapidly dismiss such views about τέχνη but instead assumes them or at least dialectically appeals to them elsewhere in a manner which suggests that at least some of his interlocutors would have found such views attractive. Thus, for instance, σωφροσύνη and τέχνη are assumed to have an unerring (ἀναµάρτητος) nature in the Charmides (171d1-172a5),33 and Socrates also speaks of ἐπιστήµη (often spoken of interchangeably with τέχνη)34 in a similar manner later on in the Republic (477e5-7).35 However, perhaps the most significant appeal to such views is found in the Euthydemus. There, in his exhortation to wisdom (σοφία), Socrates claims (280a6-b3).:
 Wisdom (σοφία) makes men succeed (εὐτυχεῖν) in every case, since I don’t suppose she would ever make any sort of mistake but must necessarily act correctly and succeed—otherwise she would no longer be wisdom (οὐ γὰρ δήπου ἁµαρτάνοι γ’ ἄν ποτέ τι σοφία, ἀλλ’ ἀνάγκη ὀρθῶς πράττειν καὶ τυγχάνειν· ἦ γὰρ ἂν οὐκέτι σοφία εἴη).  We finally agreed (I don’t know quite how) that, in sum, the situation was this: if a man had wisdom, he had no need of good fortune in addition (Συνωµολογησάµεθα τελευτῶντες οὐκ οἶδ’ ὅπως ἐν κεφαλαίῳ οὕτω τοῦτο ἔχειν, σοφίας παρούσης, ᾧ ἂν παρῇ, µηδὲν προσδεῖσθαι εὐτυχίας).
In , εὐτυχεῖν is equated simply with successful and error-free action (even if this successful and error-free action is entirely down to the agent) and in , Socrates is claiming—much as Thrasymachus did above—that so long as wisdom is present it guarantees successful action in a manner which is completely free from mistakes. (Socrates makes the same claim regarding various τέχναι in 279d8-e6). In contrast, in , εὐτυχία is equated with factors outside an agent’s control.36 The thought here is that agents who possess wisdom require only wisdom. The absence of luck or favourable circumstances will not impugn their ability because, even if luck or good fortune is not present, they are nonetheless guaranteed to succeed. In what follows, Socrates repeats this thought and, adverting to various τέχναι (such as carpentry, musicianship and so on), he claims: ‘knowledge (ἐπιστήµη) seems to provide men not only with good fortune (εὐτυχία) but also with success (εὐπραγία), in every case of possession (κτῆσις) or action (πρᾶξις)’ (Euthydemus 281a6-b4).37
As has been noticed by readers, Socrates’ claims about σοφία guaranteeing success receive little explicit argumentative support in the dialogue.38 However, Panos Dimas and Daniel Russell have suggested that one may explain Socrates’ claims by taking him to have in mind so-called ‘internal-successes’: actions which are successful purely in virtue of their internal features (rather than their results).39 Thus, for instance, while the expert striker will not always score when he shoots (this would be an example of external-success), he does—the thought goes—always hit the ball well (e.g. with good aim). Dimas and Russell do not put forward any textual support for this suggestion, but it is attractive and—as we have seen above—something very much like this view was suggested by the author of De Arte in attempting to defend medicine from apparent failures (De Arte 7-8). However, even if we suppose that the agent’s actions are internal-successes, a worry remains. Putting to one side whether (e.g.) the archer will hit her target or not, why suppose that she will always make a good shot?40 What motivates this view is presumably (TECHNICAL ABILITY) or a view very much like it. The relevant assumption is that the ability constitutive of τέχνη guarantees success independently of luck or circumstance because if one possesses ability (constitutive of τέχνη or σοφία), then one can perform the relevant action(s) when one attempts to.
Whether Socrates puts forward these views about the infallibility of τέχνη or σοφία merely dialectically, or whether Socrates genuinely accepts these views is not a primary concern.41 Although the sufficiency of τέχνη or σοφία for practical success has often been deemed a Socratic thesis (since at least the Stoics),42 the thesis is not distinctively Socratic and seems to have Sophistic origins. Such claims about the infallibility of τέχνη were deemed attractive or at least plausible by the interlocutors of Plato’s Socrates (in the relevant contexts) and could be effectively dialectically appealed to in at least certain contexts. Accordingly, Thrasymachus’ claims about the infallibility of τέχνη are not a purely ad hoc or ungrounded response to Socrates’ criticisms, but instead articulate a serious existing view (which Plato’s Socrates does not rapidly dismiss) according to which a τέχνη is a complete and perfected area of rational expertise whose success cannot be credited to luck and which guarantees that a skilled practitioner will act successfully when they attempt to.
4 The Nature of τέχνη and the Arguments of Republic I
Thrasymachus evaded Socrates’ reductio and supported his initial claim that justice is the advantage of the stronger by assuming that the practitioners of a τέχνη are infallible and claiming that rulers infallibly decree what is best for themselves (340e8-341a4). Socrates’ response to Thrasymachus consists in several connected arguments which have typically been deemed to be extremely weak and somewhat disconnected.43 If they are allowed any value, it is taken to consist primarily in raising (but unsatisfactorily discussing) issues which are examined later on in the Republic.44 However, Socrates’ arguments have generally not been well understood.45 In what follows, I will reconstruct and clarify the arguments while arguing that appreciating their dialectical nature enables us to see that they are significantly stronger than often supposed.
4.1 The Altruism of τέχνη
Socrates’ first argument against Thrasymachus is that a genuine τέχνη is directed not simply towards some good or other (as is elsewhere often assumed by Plato’s Socrates),46 but towards the good of the object of the τέχνη (Rep. 341b3-343a4). Such a view may be suggested in some other dialogues and may represent the views of the historical Socrates (Xenophon, Mem. 1.2.32),47 but it only receives sustained attention and defence in Republic 1 and clearly has a strongly dialectical purpose.48 Most saliently, the argument is regarded by Socrates and several of the other witnesses as decisively showing the deficiencies of Thrasymachus’ account of justice and ‘turning it into its opposite’ (343a1-4).
Despite its importance, Socrates’ argument has often been misunderstood. Readers often suppose that Socrates argues that τέχναι are altruistic (i.e. directed towards the advantage of their objects) on the basis of observing one or two existing τέχναι like sailing and medicine, judging that they are directed towards the good of their objects, and then inductively inferring that every τέχνη is directed towards the good of its object.49 However, such readings misrepresent Socrates’ argument to its detriment. They make the argument question-begging (because no support is offered on behalf of the judgements that such-and-such τέχνη is altruistic and Thrasymachus denies these claims) and inductively weak (for the conclusion is arrived at on the basis of very few observations).50
A better interpretation is possible. We should note that Socrates begins the argument by attempting to get clear on precisely what Thrasymachus thinks about the nature of τέχνη. Taking a doctor as his first example, Socrates suggests that the person who is really a doctor (τὸν τῷ ὄντι ἰατρὸν ὄντα, 341c6-7; cf. 341b3-8) is a healer (θεραπευτής, 341c8) rather than, for instance, a money-maker (341c5-7). Socrates then turns the talk to what is expedient. Thrasymachus rapidly agrees that every τέχνη is directed towards discovering something advantageous (341d5-8),51 and that what is advantageous for each τέχνη is for it to be as complete or perfect (τέλειος) as possible (341d11-12; cf. Hippocrates, Loc. Hom. 46). Socrates elucidates this claim by comparing the art of medicine to the human body (341e2-8). The human body is not self-sufficient and requires something (i.e. medicine) to provide what is advantageous to the body and to heal the body when it is defective (πονηρός, 341e5).52 Socrates then discusses τέχνη more abstractly and, after some preliminary questions,53 the following exchange takes place (Rep. 342b1-342c2, trans. Reeve):
[Socrates:] Or does it need neither itself nor another craft to consider what—in light of its own deficiency—is advantageous for it (ἢ οὔτε αὑτῆς οὔτε ἄλλης προσδεῖται ἐπὶ τὴν αὑτῆς πονηρίαν τὸ συµφέρον σκοπεῖν)? Indeed, is there no deficiency or error in any craft and is it inappropriate for any craft to consider what is advantageous for anything besides that of which it is the craft (οὔτε γὰρ πονηρία οὔτε ἁµαρτία οὐδεµία οὐδεµιᾷ τέχνῃ πάρεστιν, οὐδὲ προσήκει τέχνῃ ἄλλῳ τὸ συµφέρον ζητεῖν ἢ ’κείνῳ οὗ τέχνη ἐστίν)? And since it is itself correct, is it without fault or impurity (αὐτὴ δὲ ἀβλαβὴς καὶ ἀκέραιός ἐστιν ὀρθὴ οὖσα) so long as it is wholly and precisely the craft it is? Consider this with regard to that precise account. Is it so or not?
[Thrasymachus:] It appears to be so.
[Socrates:] Doesn’t it follow that medicine does not consider what is advantageous for medicine, but for the body?
Socrates here appeals to Thrasymachus’ earlier claims (and those of other infallibilist thinkers) and states that each τέχνη is perfect, i.e. it has no defect (πονηρία) and is free from error (ἁµαρτία) (342b2-4). Given that each τέχνη is directed at or provides something advantageous, and that each τέχνη is perfect, Socrates thinks that it follows that each τέχνη is directed not towards what is advantageous to itself (for it is already perfect and cannot be improved) but is instead directed towards what is advantageous for the objects over which it is set (342c4-6; cf. 345d1-5). Having established this conclusion, Socrates then proceeds to observe that medicine is directed not towards its own advantage, but that of the body (342c1-2), and that the equestrian τέχνη is directed towards the advantage of horses (342c4-6, d1-3; cf. Euthyphr. 13a2-8).
Far from baldly claiming that some particular τέχναι are directed towards the good of their objects and then inferring that every τέχνη is directed towards the good of its object (as is often supposed), Socrates has instead offered a more abstract argument of the following form:
(1) Each τέχνη provides something advantageous to itself or something advantageous to its object;
(2) each τέχνη is perfect;
(3) if a τέχνη is perfect, it does not provide something advantageous to itself;
∴ (4) each τέχνη does not provide something advantageous to itself;
∴ (5) each τέχνη provides something advantageous to its object.
The conclusion of the argument is established with an eye towards showing that Thrasymachus’ claims about the infallibility of τέχνη do not safeguard his views concerning the nature of justice and of ruling. Accordingly, after offering the argument, Socrates proceeds to criticise Thrasymachus and claims that no doctor seeks or orders what is advantageous to himself, but what is advantageous to his patient (342d3-5; cf. Hippocrates, Vict. 1.2). The same applies to ruling on the assumption that ruling is a craft (or simply on the assumption that τέχναι rule over or are stronger than their objects, 342c8-d2).
Socrates’ argument may be criticised (and we shall examine Thrasymachus’ objections in a moment),54 but the dialectical context is important. It is because Thrasymachus takes a τέχνη to be a complete and perfected area of rational expertise which guarantees success independently of luck or circumstance that Socrates is able to establish (2). Supposing that τέχναι are indeed infallible and perfect in the manner Thrasymachus takes them to be, and that ruling is a τέχνη, rulers are directed towards the advantage of those over whom they rule (342e7-11; cf. 346e3-7, 347a1-3). Thrasymachus is thereby wrong to think that rulers seek their own advantage at the expense of (i.e. the disadvantage) of those over whom they rule (343a2).55
4.2 Thrasymachus’ Objections
Although Socrates’ argument on behalf of the altruistic nature of τέχνη is valid and stronger than often supposed, it is nonetheless open to various objections. Thrasymachus’ objection(s) to Socrates are only reported briefly, but it is clear that Thrasymachus does not abandon his claims concerning the perfection and infallibility of τέχνη (i.e. he does not challenge (2)). Instead, Thrasymachus invokes the image of the shepherd (sometimes considered a model for rulers)56 in order to ridicule the notion that shepherds fatten their flocks and take care of them with some aim other than what is good for their masters and themselves (πρὸς ἄλλο τι βλέποντας ἢ τὸ τῶν δεσποτῶν ἀγαθὸν καὶ τὸ αὑτῶν, 343b1-4; cf. Tht. 174d3-e2).57 Thrasymachus thus takes the shepherd to differ from the wolf primarily in his systematicity and goes on to claim that those to whom one would apply the term ‘unjust’ always come out ahead to the disadvantage of those to whom one would apply the term ‘just’ (Rep. 343d2-344a3) and that the person who comes out best of all is the rapacious tyrant.58 Such a person is regarded by ordinary convention as the paradigm of injustice, but Thrasymachus takes the person who maximally outdoes or overreaches (πλεονεκτεῖν, 344a1-2) and takes away all the goods of everyone else (Rep. 344a4-b1; cf. 349c7-9) to be a paradigm of intelligence and practical reason. The salience of some aspects of Thrasymachus’ retort is more readily apparent when we keep in mind that in the ancient world shepherds often did not typically own the flocks which they tended, but instead worked for the benefit of their master, the owner of the sheep, in return for payment (µισθός, e.g. Homer, Il. 21.446-60).59 For our purposes, it is important to note that Thrasymachus’ retort seems to capture three of the stronger possible objections to Socrates’ argument that τέχνη is altruistic.
First, the truth of (3) seems suspect because the fact that a τέχνη is complete and perfect, i.e. error-free, does not indicate that it cannot or does not provide advantages to itself or its practitioners independently of its sphere of application. Thus, one might accept that medicine cannot provide medical discoveries which might make the medical τέχνη more effective (or that medical practitioners cannot make themselves better medical practitioners) because maximal effectiveness has already been attained. However, medicine or medical practitioners might nonetheless seek advantages independent of the sphere of application of that τέχνη (e.g. material advantages) in the same way that the shepherd seeks payment.
Secondly, there is a further feature of Thrasymachus’ retort. Thrasymachus’ discussion of the tyrant and his praise of outdoing or overreaching (πλεονεκτεῖν) and of πλεονεξία (wanting to gain something at the expense of another)60 indicates that Thrasymachus not only thinks that a τέχνη is directed towards the good of its practitioners or the rulers of the πόλις rather than its object, but that a τέχνη—as a paradigm of reasoned and intelligent activity—does so at the expense of and to the disadvantage of its object. This seems to be based upon the assumption that goods are zero-sum.61 For one person to gain benefit or advantage another person must be disadvantaged.
Thirdly, even if we put aside worries concerning how broadly the relevant universal claims are meant to apply,62 one might question why one should accept (1) as true. Even if a τέχνη—since it is perfect—cannot provide something advantageous to itself or its practitioners, it needn’t thereby provide something advantageous to its object. The disjunction in (1) is not exhaustive and a τέχνη which is not directed at its own advantage might nonetheless provide something advantageous to someone or something other than its object, such as the rulers of the πόλις (cf. 345b8-d1). Just as shepherds benefit the owners of their herds, so too medical practitioners might be directed towards benefiting the hospital board, medical insurance companies, or the government.
4.3 Socrates’ Response(s) to Thrasymachus’ Objections: Wage-Earning, Overreaching, and Successful Functioning
In response to Thrasymachus’ objection(s), Socrates offers a series of connected but difficult arguments. Initially, Socrates responds to Thrasymachus by claiming that the compensation offered to rulers indicates that ruling is not itself directed towards its own advantage (345e5-346a1) and proceeds to offer:
(α) a discussion concerning wage-earning (µισθωτική) (Rep. 346a1-347a5).
Then, in the course of examining whether the unjust person’s life is better than the just person’s life (347e2-354c3), Socrates offers:
(β) an argument that a genuine practitioner of a τέχνη does not outdo or overreach (πλεονεκτεῖν) and thus does not act unjustly (349b1-350c11);
(γ) an argument that justice is required for appropriate or successful acting, ruling or functioning (Rep. 351a6-352a10);
(δ) an argument that souls rule, deliberate and live well if and only if souls are just (Rep. 352d2-354a11).
Socrates’ arguments are best understood when we see how Socrates’ arguments constitute a response to Thrasymachus’ objections and appreciate the dialectical nature of Socrates’ arguments.
First, let us consider (α), the discussion of wage-earning. This is best understood as a response to Thrasymachus’ first objection (that a perfect τέχνη may nonetheless seek advantages independent of its sphere of application). In the discussion of wage-earning, it is claimed that τέχναι are distinct because their constitutive δυνάµεις are distinct (Rep. 346a1-3) and that each distinct τέχνη brings about (παρέχειν, ποιεῖν) or is directed towards (παρασκευάζειν, ἐπιτάττειν) some particular benefit (Rep. 346a6-8, c2-3), i.e. a benefit which is unique to it. Medicine brings about health, navigation brings about safety at sea (346a7-8) and wage-earning (µισθωτική) brings about wages (346b1).
Now, even if navigation at sea regularly brings about health, this does not thereby indicate that navigation is directed towards bringing about health (346b2-6). An activity may regularly bring about something even if the activity is not directed towards bringing about that thing (346b8). By the same line of reasoning, Socrates continues, neither should medicine be thought to be directed towards bringing about wages (346b11-12). Since each τέχνη is directed towards bringing about some particular benefit, and the practitioners of several distinct τέχναι gain wages, this suggests that there is a distinct τέχνη practised by the doctors, navigators, and the like, which brings about wages: wage-earning (346c2-7). Rep. 346d2-8, e3-7:
Then this benefit, receiving wages, doesn’t result from their own craft (ἀπὸ τῆς αὑτοῦ τέχνης), but rather, if we’re to examine this precisely, medicine provides health, and wage-earning provides wages; house-building provides a house, and wage-earning, which accompanies it, provides a wage; and so on with the other crafts. Each of them does its own work and benefits the thing it is set over. So, if a wage isn’t added, is there any benefit that the craftsman gets from his craft? … Then, it is clear now, Thrasymachus, that no skill or rule (ἀρχή) provides for its own advantage, but, as we’ve been saying for some time, it provides and orders for its subject and aims at its advantage, that of the weaker, not of the stronger.
As Socrates here makes clear, with the exception of wage-earning (which brings about wages for its practitioner, 346d3-5, 346c5-11), each τέχνη brings about a particular benefit which benefits the object over which it is set (ὠφελεῖ ἐκεῖνο ἐφ’ ᾧ τέτακται, 346d5-6, e3-7). The same applies to ruling (on the assumption that ruling is a τέχνη) (cf. 346e7-347a5). Unless wages are provided (or some penalty avoided), the practitioner of a τέχνη in fact gains no direct advantage from ruling because the τέχνη is directed towards the good of its object.
In this discussion of wage-earning, Socrates makes two important assumptions:
for any τέχνη A (e.g. medicine), there exists some benefit B (e.g. health) such that only A is suitably directed towards B;
each τέχνη is responsible for only that benefit B which it is suitably directed towards;
Thus, per (a), if medicine is a τέχνη, then there is some benefit (e.g. health) such that only medicine is suitably directed towards it. Equally, per (b), medicine is responsible only for bringing about health, and navigation is responsible only for bringing about safety at sea (Rep. 346b2-6).63 Accordingly, it follows that a τέχνη may regularly bring about a result without being suitably directed towards said result. To argue that wage-earning is a τέχνη, a further assumption seems to be required: that for each kind of benefit suitably (or perhaps merely regularly) brought about, there is a distinct τέχνη which is suitably directed towards it (346c2-7).
It is often thought that the notion of a wage-earning τέχνη is problematic on its own terms (for instance: what, precisely, does a wage-earning τέχνη do?), and that, by allowing or arguing for the existence of an anomalous τέχνη—such as wage-earning—which benefits its practitioner(s), Socrates is fatally undermining his own claim that τέχναι are altruistic,64 and that the discussion does nothing to provide an effective response to Thrasymachus’ objections.65 However, such readings misunderstand the nature and function of the argument (and perhaps also the nature of Thrasymachus’ objections). In the first instance, even if Socrates genuinely accepted the premises, reasoning, and conclusion of the argument, the existence of an anomalous τέχνη which is not directed towards the advantage of its object does not discredit Socrates’ claims about the altruistic nature of τέχνη. At most it merely restricts the domain of quantification of his claims (as was noted above, these might require some restriction anyway).
More importantly, even if a wage-earning τέχνη is problematic, it is important to notice is that this is a problem for Thrasymachus. Once again, we must be sensitive to the dialectical nature of Socrates’ argument. Socrates is not aiming to establish that in his own view there is a wage-earning τέχνη. Instead, he aims to show that, if Thrasymachus stands by his claims about shepherds and rapacious rulers, then by Thrasymachus’ own lights he should accept that there is a wage-earning τέχνη (and that this τέχνη is responsible for the relevant wages). The existence of a wage-earning τέχνη requires the assumption that each distinct τέχνη produces or is responsible for only that unique benefit B which it is suitably directed towards and that wages are a benefit. However, as Socrates’ needling about accuracy and precision makes clear (e.g. 346b2-6, d2), it was Thrasymachus who claimed that a τέχνη of φ-ing manifests itself only in φ-ing (340d2-341a3, 341c10-d4; see Section 3 above). Equally, it is Thrasymachus (not Socrates)66 who assumes that wages are a benefit.67
Far from ‘establishing nothing’ (as per Beversluis 2000, 235), Socrates’ discussion of wage-earning is an effective response to Thrasymachus’ first objection. Pace Thrasymachus, the fact that a shepherd earns a wage for his labour does not show that shepherding is directed towards said wage(s). Socrates appeals to Thrasymachus’ claims about τέχνη (and what a τέχνη is responsible for) to show that—by Thrasymachus’ lights—shepherding cannot be directed towards such a wage (with something similar applying to ruling and the other τέχναι).
Socrates’ argument (β) that a genuine practitioner of a τέχνη does not overreach, outdo, or take advantage (πλεονεκτεῖν) and thus does not act unjustly (349b1-350c11) is somewhat abstruse.68 It has probably attracted greater castigation than the other arguments,69 but is best understood as a response to Thrasymachus’ second objection (which adverts to a τέχνη as a model of intelligent activity to argue that its practice leads to the disadvantage of its object). In broad outline, Socrates argues as follows. Thrasymachus thinks that the just person aims to outdo (πλεονεκτεῖν) only the unjust person, whereas the unjust person ‘strives to get the most he can for himself from everyone’ (349c7-9) and thus aims to outdo (πλεονεκτεῖν) everyone: both the just and the unjust (349b1-d3). That is to say, the unjust person seeks to outdo those like himself and those unlike himself, whereas the just person seeks to outdo only those unlike himself (349c11-d2). Socrates gets Thrasymachus to agree that the practitioner of a τέχνη (i.e. the one who is good and clever)70 does not wish to outdo (πλεονεκτεῖν) his fellow practitioners, but does wish to outdo those who are non-practitioners (i.e. those who are bad and ignorant, 349e10-350a10) (Rep. 350a6-9):
In any branch of knowledge or ignorance, do you think that a knowledgeable person would intentionally try to outdo other knowledgeable people or say something better or different than they do, rather than doing or saying the very same thing as those like him?
In contrast, the non-practitioner of a τέχνη seeks to outdo both practitioners and non-practitioners. The non-practitioner thus seeks to outdo (πλεονεκτεῖν) those who are like himself (i.e. non-practitioners, who are ignorant and bad) and those who are unlike himself (i.e. practitioners, who are clever and good) (350a11-c11). Accordingly, the non-practitioner of a τέχνη precisely resembles—and shares the same qualities as (349c11-d12, 350c7-8)—71 the unjust person, and the unjust person is neither clever nor good (as Thrasymachus supposed). The exploitative person who seeks to outdo, overreach, or take advantage of (πλεονεκτεῖν) everyone is thus not a practitioner of a τέχνη.
Socrates’ argument seems to face two principal worries. First, although one might try to consistently translate ‘πλεονεκτεῖν’ as ‘outdo’ (or perhaps ‘do better than’ or something similar), the term seems to vary between having the same sense as: (i) ‘performing an activity better than others’ (cf. Leg. 683a2-4); and (ii) ‘taking advantage of (or gaining advantages at the expense of) others’ (cf. Rep. 362b7; Grg. 490d11-e8).72 Owing to the fact that in English, and several other languages, senses (i) and (ii) of ‘πλεονεκτεῖν’ are typically reproduced in different and unrelated expressions, it seems that the senses are distinct.73 There thus seems to be equivocation (either ambiguity or polysemy).
Secondly, there is the worry that the controversial assumption in the argument—that practitioners of a τέχνη do not seek to outdo (πλεονεκτεῖν) fellow-practitioners—is ungrounded (Socrates gives no reasons for it) and that it is either obviously false (when construed as the claim that practitioners of a τέχνη do not seek to do better than their fellow practitioners), or else (when construed as the claim that practitioners of a τέχνη do not seek to take advantage of their fellow practitioners) should never have been granted by Thrasymachus.74
The first worry seems justified, but it is not easy to entirely rule out (or, for that matter, to establish) that there is a uniform sense here.75 However, this second worry may be addressed if we—once again—attend to the dialectical nature of Socrates’ argument and how Socrates is appealing to Thrasymachus’ earlier claims about the perfection and infallibility of τέχνη. Thus, although several readers complain that the argument is flawed because practitioners of a τέχνη are often competitive and obviously seek to do better than each other,76 it is important to notice that Thrasymachus cannot readily allow this. This is because, according to Thrasymachus, each τέχνη is maximally perfect and its practitioners act unerringly. A genuine practitioner presumably knows that the actions of fellow genuine practitioners cannot be improved upon and thus cannot seek to do better than his fellow (genuine) practitioners (cf. Rep. 380e3 ff.). Thus, for instance, a true doctor (musician, etc.) cannot seek to perform his task better than another true doctor (musician, etc.), but merely, as Socrates says (350a6-9, cited above), seeks to do the same thing as him (and to outdo non-practitioners).
Once we appreciate that the claim that practitioners of a τέχνη cannot seek to outdo each other is something that Thrasymachus seems to be committed to (rather than something Socrates has unwarrantedly plucked from thin air), we can see that, although argument (β) may be flawed due to the possible ambiguity or polysemy of ‘πλεονεκτεῖν’, Socrates offers a plausible dialectical response to Thrasymachus’ objection(s). Given Thrasymachus’ assumptions, a genuine practitioner of a τέχνη does not universally πλεονεκτεῖν (as Thrasymachus had earlier claimed, e.g. 344a1-2). Thrasymachus’ claim that a genuine practitioner of the ruling τέχνη is an intelligent, rapacious tyrant driven by a universal desire to πλεονεκτεῖν (e.g. Rep. 362a2-c6, 365d2-6; 574a6-10, 586a1-b4) is thus incorrect and the person who does universally πλεονεκτεῖν resembles (and shares the same qualities as) not the clever and the skilled, but the ignorant and the bad.
Finally, arguments (γ) and (δ) directly address whether the life of injustice is superior to the life of justice. Socrates initially aims to show that justice is stronger than injustice (cf. Rep. 350e11-351b2) and that the just are more capable of acting (δυνατώτεροι πράττειν οἱ δίκαιοι, 352b8-9). To this end, he initially offers a reductio, arguing that, even if it is true that, if x enslaves y, then x is stronger than y (cf. Rep. 351b7-8), there is nonetheless some absurdity which results from supposing that x unjustly enslaves y. This is justified by a subsidiary argument, (γ), to the conclusion that, if x is unjust, then x is unable to act successfully (Rep. 351c7-9; cf. Prot. 322b6-8, 324d7-325a1, 333d1 ff.). Simplified, the argument is as follows:
(1) If x is composed of elements acting unjustly towards each other,77 then x is conflicted and unable to act successfully (351c7-d5);
(2) x is unjust iff x is composed of elements acting unjustly towards each other (Rep. 351e4-8, 352a6-7);
∴ (3) if x is unjust, then x is conflicted and unable to act successfully (Rep. 351e10-352a3).
This assumes that injustice in an agent—whether an individual agent or group agent—is a matter of that agent’s parts or elements acting unjustly towards each other (an issue which receives greater attention later on in the Republic).78 Injustice prevents individuals and groups from functioning successfully (Rep. 352a6-9; cf. Aristotle, EN 1167b9-16).
In argument (δ), Socrates invokes considerations about the ἔργον (‘function’) of things to argue that souls rule, deliberate and live well if and only if souls are just (Rep. 352d2-354a11). A simplified version of the argument runs thus:79
(1) If x has a function (ἔργον) φ then there exists some appropriate excellence (ἀρετή) A such that x φs well (εὖ ἐργάσεται, etc.) iff x has A (Rep. 353b14-d2, e1-2);
(2) souls have the function of managing, ruling, deliberating, and living (Rep. 359d9-10);
∴ (3) there exists some appropriate excellence A such that souls manage, rule, deliberate, and live well iff souls have A;
∴ (4) this excellence is justice (Rep. 353e7-8);80
∴ (5) souls rule, deliberate and live well iff souls have justice (i.e. are just).
Justice is thus an excellence of the soul which enables it to function well and—contrary to Thrasymachus’ claims—injustice is not a sign of strength or ability (344c5-7), but something which cripples and renders things unable (Rep. 351e10-352a3).
Arguments (γ) and (δ) do not offer a direct response to Thrasymachus’ objections (they primarily address broader concerns), but they do suggest that Thrasymachus should reconsider his assumptions about the nature of advantage and his view that goods are zero-sum (i.e. that for one person to gain advantage, another must be disadvantaged) while offering additional grounds for criticising Thrasymachus’ views about justice. Thus, for instance, argument (γ) prompts us to reconsider the relation between part and whole, between individuals and the societies they are members of, and what is beneficial to each. It suggests that what benefits a part benefits the whole of which that part is a constituent and that, in benefiting some other part, one might thereby benefit the whole of which one is also a part. This clearly anticipates some of the later central concerns of the Republic (cf. Rep. 420b3-c4; 519c8-521b11; Leg. 715a8-d6), but for our purposes it suffices to notice that the argument challenges Thrasymachus’ assumption that in each benefit-producing action there is only one locus for advantage (and that benefit or advantage in one locus comes at the price of a disadvantage in another). Although Plato’s Socrates does not offer an explicit response to Thrasymachus’ third objection,81 he does address the grounds of Thrasymachus’ second objection and makes salient the possibility that, in the first premise of the argument that τέχνη is altruistic (i.e. the premise that each τέχνη provides something advantageous to itself or something advantageous to its object), the disjunction is an inclusive disjunction. The practitioners of a τέχνη may benefit their objects and themselves.
Argument (δ) provides a similar moral. Beyond simply claiming that acting justly benefits oneself, reflecting upon the ἔργον of things prompts an additional response to Thrasymachus’ objection(s). Thus, suppose that a τέχνη benefits its objects, and that one of the ways it does so is by helping them attain excellence (ἀρετή). For instance, a shepherd might benefit his sheep by helping them attain ἀρετή and thus enable them to fulfil their function (ἔργον).82 If the ἔργον of sheep is directed even in part towards benefiting humans, then sheep are benefited by shepherding but shepherds (and others) are benefited by the sheep in turn (and so too the equestrian τέχνη benefits horses, but horses benefit humans, and so on). Accordingly, reflecting upon the ἔργον of things provides additional grounds for doubting that in every action there needs to be a winner and a loser, or that working for the advantage of another results in one’s disadvantage (or vice versa). Instead, in benefiting another one might also thereby benefit (be it directly or indirectly) oneself (cf. Rep. 369a1 ff.). One of the major aims of the rest of the Republic is to provide further warrant for how and why this should be so.
Readers of the Republic usually rapidly dismiss Thrasymachus’ claim that a τέχνη is infallible as an ad hoc and ungrounded attempt to defend his account of justice from Socrates’ criticisms. Readers also typically take Socrates’ subsequent arguments against Thrasymachus’ views to be extremely weak. I have here sought to clarify and explain Thrasymachus’ views and have emphasised their place within a broader tradition which took a genuine τέχνη to be a complete and perfected area of rational expertise whose success is not to be credited to luck. Plato’s Socrates, I showed, does not dismiss such views, but appeals to them elsewhere in a manner which suggests they were found attractive or at least plausible by Socrates’ interlocutors.
I then clarified Socrates’ response to Thrasymachus’ views about justice, Socrates’ argument that a τέχνη is directed towards the advantage of its object, Thrasymachus’ objections to this argument, and Socrates’ subsequent arguments (α)-(δ). On the reading I have offered, Socrates’ arguments are not ‘weak and unconvincing to an amazing degree’ (as per Annas 1981, 50), but instead form a coherent and interesting series of arguments. Socrates dialectically appeals to Thrasymachus’ claims about the perfection and infallibility of τέχνη to argue that each τέχνη is directed towards the advantage of its object and that genuine ruling cannot be as Thrasymachus says. Thrasymachus’ objections advert to the idea of intelligent, exploitative τέχναι which seek their own advantage or that of the rulers of the πόλις at the expense of their objects. In response, Socrates offers a connected series of dialectical arguments which show that by Thrasymachus’ own lights there cannot be intelligent, rapacious rulers who practice a ruling τέχνη at the expense of those they rule. The first book of the Republic thus serves as a προοίµιον (357a2) to the Republic as a whole, but Socrates’ arguments do more than merely raise issues which are to be tackled later on in the Republic. Once their true form and dialectical nature is understood, Socrates’ arguments in Republic 1 offer a more effective response to Thrasymachus than often supposed. In keeping with Plato’s own remarks about the function of προοίµια, Socrates’ arguments in Republic 1 are ‘an exercise in skilled dialectical reasoning (ἔντεχνον ἐπιχείρησιν) which is useful for what will subsequently be accomplished’ (Leg. 722d3-6).83
Algra, K. (1996), ‘Observations on Plato’s Thrasymachus: The Case for Pleonexia’ in K. Algra, P. W. van der Horst, and D. Runia (eds.), Polyhistor: Studies in the History and Historiography of Ancient Philosophy (Leiden), 41-60.
Barney, R. (2006), ‘Socrates’ Refutation of Thrasymachus’ in G. Santas (ed.), The Blackwell Guide to Plato’s Republic (Malden), 44-62.
Fine, K. (1982), ‘Acts, Events, and Things’ in W. Leinfellner, E. Kraemer, and J. Schank (eds.), Language and Ontology (Vienna), 97-105.
Grice, P. (1989), ‘Metaphysics, Philosophical Eschatology, and Plato’s Republic’ in P. Grice (ed.), Studies in the Ways of Words (Cambridge), 304-38.
Hutchinson, D. (1988), ‘Doctrines of the Mean and the Debate Concerning Skills in Fourth-Century Medicine, Rhetoric and Ethics’, Apeiron 21: 17-52.
Joseph, H. W. B. (1935), ‘Plato’s Republic: The Argument with Thrasymachus’ in H. Joseph (ed.), Essays in Ancient and Modern Philosophy (Oxford), 15-40.
Kahn, C. (1993), ‘Proleptic Composition in the Republic, or Why Book I was Never a Separate Dialogue’, Classical Quarterly 43: 141-2.
Nawar, T. (forthcoming a), ‘Relativism in Ancient Greek Philosophy’ in M. Kusch (ed.), The Routledge Handbook to the Philosophy of Relativism (London).
Striker, G. (1996), ‘Plato’s Socrates and the Stoics’ in her Essays on Hellenistic Epistemology and Ethics (Cambridge), 316-24.
Vlastos, G. (1984), ‘Happiness and Virtue in Socrates’ Moral Theory’, Proceedings of the Cambridge Philological Society 210: 181-213.
Adam 1902, 33; Joseph 1935, 18; Allan 1940, 27; Cross and Woozley 1964, 46-7; Annas 1981, 43; Klosko 1984, 14-15; Pappas 2003, 30; Santas 2010, 20-1.
Annas 1981, 50. See also Cross and Woozley 1964, 52, 58; White 1979, 8, 61-73; Annas 1981, 49-58; Reeve 1988, 19-21; Grice 1989, 312; Beversluis 2000, 228-42.
Thrasymachus claims that: (i) justice is the advantage of the stronger (338c2-3, c6, 339a3-4, 339b5-7, 341a3-4, 343c3-4, 344c7-8, 347e1-2); (ii) justice is the advantage of the ruler (338e1-3, 338e6-339a4); and (iii) justice is the advantage of another (343c3-4; cf. 392b3-4). Thrasymachus’ remarks are difficult to render entirely consistent. For discussion over how to construe Thrasymachus’ definition (if it is a definition) of justice, see Kerferd 1947; Cross and Woozley 1964, 23-41; Nicholson 1974; Annas 1981; Reeve 1985; 2008, 86-98; Boter 1986, Chappell 1993; Irwin 1995, 174-5; Williams 1997; Barney 2006; Wedgwood 2017. To my mind, if we examine what these claims jointly amount to, it seems that Thrasymachus might mean that: (α) x acts justly iff there is a y such that y is stronger than (i.e. rules over) x and x acts to y’s advantage; or (β) x acts justly iff there is a y such that y is strongest (i.e. rules x’s πόλις) and x acts to y’s advantage; or (γ) x acts justly iff there is a y such that x ≠ y and y is strongest (i.e. rules x’s πόλις) and x acts to y’s advantage.
We may further note the following points. First, assuming that being stronger than and ruling over are asymmetric relations, then neither (α) nor (γ) seem to allow that the rulers of a πόλις may act justly, whereas (β) does. Secondly, it seems that (α) more easily allows for a hierarchy of advantage such that (e.g.) the actions of a cobbler’s slave benefiting his master are just, and the cobbler’s actions benefiting his landlord are just, and the landlord’s actions benefiting the city’s rulers are just (for (β) and (γ) to allow for such hierarchies, simultaneous membership of multiple πόλεις is required, cf. 422e5-b3). Thirdly, it seems that (β) and (γ) are favoured by Rep. 338d7-339a4 (which equates ‘the stronger’ with those who are in charge) while (iii)—i.e. the claim that justice is the advantage of another (343c3)—prima facie favours readings (α) and (γ). Fourthly, (iii) (i.e. Rep. 343c3) should be read with caution and in context because it is intertwined and seemingly glossed as the advantage of the stronger (and, in any case, it does not rule out (β) because 343c2-4 might be taken to claim that for everyone but the rulers, justice is the good of another). Finally, at 343c1-344c9 it is not clear whether Thrasymachus is keeping to his own definitions or else adverting to popular definitions of ‘injustice’ and ‘justice’. If he is adhering to his own definition(s), then 344a4-c4 favours (α) and (γ), which do not allow that rulers may act justly.
Since an action-type may be beneficial to the rulers in one πόλις but not another (e.g. lying may be beneficial to one’s rulers when one is in Cnossos but not when one is in Athens), this can lead to a kind of relativism about justice. See Nawar forthcoming a.
Joseph 1935, 17; Cross and Woozley 1964, 46; Harrison 1967, 30-1; Maguire 1971, 145-6.
The view has been common since at least Adam 1902, 33. Cf. Joseph 1935, 18; Allan 1940, 27; Harrison 1967, 30-1; Maguire 1971, 145-146; Pappas 2003, 30; Dorter 2005, 37-9; Barney 2006, 48; Sheppard 2009, 35. However, it is often not entirely clear what terms like ‘realism’ or ‘idealism’ mean in these contexts.
E.g. ‘a hasty and confused thinker’ (Annas 1981, 38); ‘an absence of rational agency’ (Blondell 2002, 181). Cf. Maguire 1971; Klosko 1984; Grice 1989, 309; Everson 1998.
Cf. Cross and Woozley 1964, 17; Nicholson 1974, 222-5; Everson 1998, 121.
There have been no detailed attempts to explain Thrasymachus’ claims about τέχνη. Even sustained treatments of τέχνη in Republic 1—such as those offered by Cambiano 1971, Lycos 1987, Parry 1996 and 2003, Roochnik 1996 and Vegetti 1998, 193-207 (‘techne’) and 233-56 (‘Trasimaco’)—give it little attention. Reeve 1985, 250-1 and 1988, 12-13, 276-7 is virtually alone in signalling that Thrasymachus’ views concerning τέχνη may not be deeply wrongheaded. He briefly suggests that the semantics of dispositional ascriptions may explain Thrasymachus’ views—which is partly correct—but does not discuss the issue in detail.
In turn, it is assumed that a δύναµις is or constitutes the relevant τέχνη (Grg. 447c1-3, 455d6-457c3; Soph. 219a4-6; Pol. 304e3-11; cf. Isocrates, Antid. 50-1, 178-9, 197-8, 253-4, 270-2).
For further discussion, see below and Nawar 2017.
Fine 1986 offers an influential semantics for ‘x qua F ’ locutions which posits qua-objects consisting of an individual (x, the basis) and a property (F, the gloss) such that: (i) x-qua-F = y-qua-G iff (x = y & F = G); and (ii) x ≠ x-qua-F. Several philosophers are inclined to resist Kit Fine’s analysis by rejecting (ii), but all that matters for my purposes is that it is generally accepted that if x-qua-F exists at t, then x is F at t. I will assume this throughout the rest of this paper.
Here, ‘ἐπιστήµη’ (‘knowledge’) is either synonymous with ‘τέχνη’ (cf. Lyons 1963, 96, 139-176) or else signifies the knowledge one’s τέχνη is grounded in.
Cf. Prot. 345b5: αὕτη γὰρ µόνη ἐστὶ κακὴ πρᾶξις, ἐπιστήµης στερηθῆναι.
Slings notes that even though the aorist is not used at Rep. 340e3, ‘ἐπιλειπούσης’ is here almost certainly used to indicate that ‘the exhaustion is now complete’ (Slings 2005, 9) and has parallels elsewhere (e.g. Slings 2005, 8 notes that at Rep. 574d1-2, ‘the most logical interpretation [of the sentence involving “ἐπιλείπῃ”] is that the young tyrannical man has already squandered his parents’ resources’).
Cf. Rep. 340c6-7: ἀλλὰ κρείττω µε οἴει καλεῖν τὸν ἐξαµαρτάνοντα ὅταν ἐξαµαρτάνῃ; In the Protagoras it is assumed that ‘doing badly is nothing other than being deprived of knowledge’ (αὕτη γὰρ µόνη ἐστὶ κακὴ πρᾶξις, ἐπιστήµης στερηθῆναι, Prot. 345b5).
One might worry that ἐν ᾧ (340e4) need not be read temporally. However, temporal considerations are elsewhere present (ὅταν, Rep. 340e5; cf. 340c7) and it is clear that what is at issue in 340e3-5 is possession of τέχνη (e.g. ἐπιλειπούσης γὰρ ἐπιστήµης…) and the complete loss of said possession (see above). Moreover, ‘at t’ simply makes explicit something which would otherwise be implicit (e.g. if the gloss were: ‘if S has the ability (constitutive of a τέχνη) to φ, then if S were to attempt to φ, then S would φ’).
Very similar views (or views which entail very similar views about abilities in general) were sometimes assumed in twentieth-century in discussions of free will or the semantics of ‘can’. See Moore 1907, 196-222; Ryle 1949, 116-153; Austin 1956; Wolf 1990, 94-116.
Such a view is distinct from and should not be conflated with the view attributed by Aristotle to the Megarics according to which x has the ability to φ at t iff x is φ-ing at t (Metaph. 1146b29-32).
Cf. Pol. 297a5-b3.
Vegetti 1998, 238-9 notices that Thrasymachus’ talk of the accuracy of τέχνη may derive from medical thinkers, but says little about the nature of τέχνη (and nothing about its modal profile).
Isocrates takes ability (δύναµις) to be grounded in natural aptitude and practical experience (ἐµπειρία) (In soph. 14-15; Antid. 184-92, 200-1). He criticises others for neglecting practical experience (In soph. 10) and thinks there exist no general and universally applicable truths of the kind required by τέχνη (as it is conceived of by others) (Antid. 184). Although the author of On Ancient Medicine is also critical of those who think that medical τέχνη is infallible (e.g. VM 9), in contrast to Isocrates he allows that medicine is accurate (VM 12) and attributes its success to reasoning rather than luck (VM 4-6).
Galen spoke of ancient rivalries between groups or bands (χοροί) of medical thinkers from Cos, Cnidos, and Italy (De Methodo Medendi 10.5-6 Kühn; cf. Hippocrates, Acut. 1-3). There have been numerous attempts to taxonomise the views of ancient medical thinkers since then (e.g. Hutchinson 1988; Mann 2008). However, it is misleading to speak of ‘schools’ (Langholf 1990, 12-36), and it is often difficult to even securely distinguish between different ‘currents’ of thought in the ancient medical debates. The label ‘infallibilist’ is peculiar to the concerns of this paper.
In the passage cited, ‘εὐτυχής’ has roughly the same meaning as ‘successful’ while ‘τύχη’—somewhat like ‘luck’—denotes a situation or outcome due to factors outside the agent’s control. In what follows (Loc. Hom. 46), the author moves between this sense of ‘τύχη’, and another sense wherein it means positive outcome or successful actions even if due to the agent. In the Euthydemus, ‘εὐτυχία’ is similarly ambiguous or polysemous (see below).
The author makes these claims despite elsewhere emphasising how sensitive medical practitioners must be to the particulars of the situation (e.g. Loc. Hom. 41). Others adverted to particularist concerns to suggest that medicine could not be complete or perfect but was nonetheless a genuine and successful τέχνη (e.g. Hippocrates, VM 9; Vict. 1.2, 3.67; Isocrates, Antid. 184; cf. Plato, Phlb. 56b1-2).
E.g. ‘knowledge is governable and successful when the one with knowledge wishes to use it’ (Loc. Hom. 46).
The identity of these detractors is unclear. See Schiefsky 2005, 55-62.
Cf. Hippocrates, VM 1, 12; Morb. I 7.
The author ascribes infallibility not to those who simply have a desire to heal, but to those who are able or have the ability (ἐξεύρηνταί γε µὴν οὐ τοῖσι βουληθεῖσιν, ἀλλὰ τουτέων τοῖσι δυνηθεῖσι, De Arte 9), i.e. genuine doctors (as opposed to those who are merely called doctors; cf. Rep. 340d6-e3, 341b3-8, 341c5-7, 343b5, 345c2-3, 345e1-3, 347d4-6; Euthyd. 280a7-8). Moreover, he claims that (real) doctors only treat treatable diseases (De Arte 13). ‘Those who encourage such things [the taking on of incurable cases] are admired by those who are doctors in name, but are ridiculed by those who are in fact doctors by virtue of their skill’ (παρακελευόµενοι δὲ ταῦτα, ὑπὸ µὲν τῶν οὐνόµατι ἰητρῶν θαυµάζονται, ὑπὸ δὲ τῶν καὶ τέχνῃ καταγελῶνται, De Arte 8).
Cf. Mann 2012, 8-20, 44-9, 77-8.
The true doctor will only care about the opinion of those who have rationally considered what the task of a craftsman is directed towards and in relation to which thing(s) it may assessed as perfect (πρὸς ὅ τι αἱ ἐργασίαι τῶν δηµιουργῶν τελευτώµεναι πλήρεις εἰσί, De Arte 8).
Isocrates may have engaged with the thought of the historical Thrasymachus, but the details of the debates (which were seemingly at least partly over style, Cicero, Orator 30) are lost to us. The historical Thrasymachus seems to have been interested in discussions of τέχνη (Dionysus of Halicarnassus, De Isaeo 20) and in certain passages it seems Isocrates has in mind someone very much like Plato’s Thrasymachus (e.g. Isocrates, Antid. 275-6, 281-5).
‘For those of us who had temperance would live lives free from error and so would all those who were under our rule (ἀναµάρτητοι γὰρ ἂν τὸν βίον διεζῶµεν αὐτοί τε [καὶ] οἱ τὴν σωφροσύνην ἔχοντες καὶ οἱ ἄλλοι πάντες ὅσοι ὑφ’ ἡµῶν ἤρχοντο). Neither would we ourselves be attempting to do things we did not understand (οὔτε γὰρ ἂν αὐτοὶ ἐπεχειροῦµεν πράττειν ἃ µὴ ἠπιστάµεθα) … nor would we trust those over whom we ruled to do anything except what they would do correctly, and this would be that of which they possessed the knowledge (οὔτε τοῖς ἄλλοις ἐπετρέποµεν, ὧν ἤρχοµεν, ἄλλο τι πράττειν ἢ ὅτι πράττοντες ὀρθῶς ἔµελλον πράξειν—τοῦτο δ’ ἦν ἄν, οὗ ἐπιστήµην εἶχον) … And with error rooted out (ἁµαρτίας γὰρ ἐξῃρηµένης) and rightness in control, men so circumstanced would necessarily fare admirably and well in all their doings and, faring well, they would be happy’ (Chrm. 171d6-172a3; cf. Pol. 297a5-b3).
Lyons 1963 influentially argued that Plato uses the terms ‘τέχνη’, ‘ἐπιστήµη’, and ‘σοφία’ interchangeably. This often seems correct (e.g. Rep. 342c4-d2, 350a1-9; 428a11-e9, 438c6-d7; cf. Pol. 258b1-e11) and many readers of Plato follow Lyons on this matter (cf. Nawar 2013), but Balansard 2001 gives reasons to be cautious. For uses of ‘τέχνη’ in Plato, see also Roochnik 1996, 253-64.
Reeve 1985, 251 notices this point and adduces several other examples (from the Charmides, Laches and Protagoras) but he does not distinguish between the view which I am here discussing and the view that a genuine cause of F (e.g. heat) always produces F instances (e.g. hot things).
As often noticed—e.g. Hawtrey 1981, 80; Gifford 1905, 20-22; Roochnik 1996, 161-4—‘εὐτυχία’ is ambiguous between: (i) a positive outcome or successful action even if this is due to the agent (cf. ‘the harder I practice, the luckier I get’); and (ii) a positive situation, outcome, or action which is due to factors outside of the agent’s control.
For further discussion of ἐπιστήµη in the Euthydemus and its relation to ability, see Nawar 2017.
E.g. Irwin 1995, 56-60.
Dimas 2002, 17-21, 25; Russell 2005, 34-5.
This problem is noted by Jones 2013, 9-10 and is also discussed in Nawar 2017 and forthcoming b.
In the Hippias Minor, Socrates articulates a similar but more modest view about abilities. There, Socrates claims: ‘But each person who can do what he wishes when he wishes is able (δυνατὸς δέ γ’ ἐστὶν ἕκαστος ἄρα, ὃς ἂν ποιῇ τότε ὃ ἂν βούληται, ὅταν βούληται). I mean someone who is not prevented by disease or other such things, just as I might say you are able to write my name whenever you wish (οὐχ ὑπὸ νόσου λέγω ἐξειργόµενον οὐδὲ τῶν τοιούτων, ἀλλὰ ὥσπερ σὺ δυνατὸς εἶ γράψαι τοὐµὸν ὄνοµα ὅταν βούλῃ). Or don’t you say that the person in such a condition is able?’ (Hp. Mi. 366b7-c4). Socrates thus seems to claim:
(ABILITY*): If S has the ability to φ at t, then if S were to attempt to φ at t, and S were not prevented from φ-ing, then S would φ.
In explicitly making allowances for preventative factors, (ABILITY*) is more modest (and more plausible) than (TECHNICAL ABILITY). For further discussion, see Nawar forthcoming b.
Vlastos 1984; Irwin 1995, 52-76; Striker 1996; Annas 1999, 31-51, 83-88; Russell 2005, 16-47. For discussion of the Stoic epistemic notions (which differ from those of Plato), see Nawar 2014.
‘Almost embarrassingly bad’ (Cross and Woozley 1964, 52); ‘weak and unconvincing to an amazing degree’ (Annas 1981, 50); ‘grossly fallacious’ (Reeve 1988, 20); ‘weak to the point of feebleness’ (Grice 1989, 312). Some of these remarks apply to particular arguments and others to the whole batch. See Cross and Woozley 1964, 52, 58; White 1979, 8, 61-73; Annas 1981, 49-58; Reeve 1988, 19-21; Grice 1989, 312; Beversluis 2000, 228-42.
E.g. White 1979, 7-8; Kahn 1993; Algra 1996. Lycos 1987 and Barney 2006 aim to offer correctives to these readings.
This also holds of those who take more positive views towards Republic 1. Thus, for instance, Lycos 1987 sees Republic 1 as a successful examination of social and political power, but he ‘is less concerned to assess for validity the arguments Socrates uses … [than] in establishing the need for his contemporaries to rethink their attitude to justice’ (Lycos 1987, 6). Barney 2006 offers the clearest existing treatment of the arguments, but my understanding of the arguments and the points they raise against Thrasymachus differs significantly from hers.
E.g. Lach. 195c7-d2; Chrm. 165c10-e2, 171d1-2; Grg. 512b1-2; Euthyd. 288b3-293a6; Pol. 293a6-e5, 296c4-297b3; Aristotle, EN 1094a1-2. This is partly why rhetoric and cookery—which merely aim at what is pleasant (Grg. 464e2-465a2, 500b3-5)—are not considered genuine τέχναι in the Gorgias (500a7-b5, 501a3-4, e1-3; cf. Rep. 493a6-c8). Of course, rhetoric and cookery also suffer epistemic deficiencies (Grg. 465a2-7, 500e4-501b1; Phlb. 55e1-56c6).
‘Socrates said somewhere that it would seem amazing to him if someone who became a herdsman (νοµεύς) of a herd of cattle and made the cattle fewer and worse (τὰς βοῦς ἐλάττους τε καὶ χείρους ποιῶν) did not agree that he was a bad cowherd. It would be more amazing still if someone who became the presiding ruler (προστάτης) and made the citizens fewer and worse were not ashamed and did not think that he was a bad ruler of the city’ (Xenophon, Mem. 1.2.32).
That a τέχνη is directed towards the good of its object is assumed in the Gorgias (e.g. 502e2-7, 504d5-e4, 511c7-512b2, 513e2-3, 514d3-516d3) and is sometimes suggested elsewhere (e.g. Euthyphr. 13a4-c2; Lach. 195c7-d2; Pol. 293a6-e5, 296c4-297b3; cf. Soph. 219a10-b2), but—to take one example—in the Charmides it is assumed that a τέχνη will produce something which is good or beneficial partly or principally for the practitioner of the τέχνη (e.g. Chrm. 164a9-b9). That ruling is or should be directed towards the good of the ruled is fairly consistently maintained, but determining the degree to which Plato’s Socrates (or a speaker such as the Eleatic visitor in the Sophist and the Statesman) is sympathetic to the view that each τέχνη is directed towards the good of its object on a particular occasion often requires clarifying the nature of πολιτική, its relation to knowing the good (which, in the Euthydemus is said to lead into a labyrinth, 291b7) or knowing things just and unjust (cf. Grg. 459c6 ff.), the nature of rearing (τροφή) or providing care (θεραπεύειν) (e.g. Pol. 275d8-e1; cf. Grg. 500a1, 513d1-5, 521a2 ff.) and their relation to τέχναι, the relation of subsidiary τέχναι to overseeing τέχναι (cf. Grg. 517d6-518e1), the relation of the various τέχναι to architectonic πολιτική, and the relation of τέχναι to knowledge of good and evil (as in the Charmides 174a10 ff.). Cf. Nawar forthcoming b.
‘It is from particular instances which favour his thesis that Socrates has reached his generalisation that no art (or practitioner of it) pursues its own interests’ (Cross and Woozley 1964, 48). Barney 2006, 49-50, 56 takes the same view. Roochnik 1996, 140-1 also seems to take the same view.
Even Barney 2006, 50, 56—who attempts to present the argument charitably—reads the argument this way and admits these faults.
It is not entirely clear what the antecedent of ‘ἑκάστῳ’ is at 341d6. Presumably Thrasymachus agrees (341d7, 10) because he takes the advantage not to be for the object of the τέχνη (341d11-12). This is clarified by the subsequent argument. See below.
Cf. Prot. 321c1 ff.; Rep. 369b7-d9; Pol. 274b5-e4; Grg. 477e7-478b2.
Socrates rhetorically asks: (i) whether the medical τέχνη is defective (πονηρός, 342a2); (ii) whether the medical τέχνη has need of some excellence (ἀρετή, 341a2) in order to secure its advantage (341a2-4; cf. Lysis 217a7-b4); and (iii) whether the medical τέχνη (and other τέχναι) require the assistance of some τέχνη concerned with advantage (342a3-8). The answer to each question is negative.
There is some fluidity between talk of a τέχνη being directed towards its own advantage and talk of the practitioners of a τέχνη being directed towards their own advantage. However, this is not unusual in Plato (e.g. Grg. 464c-e) or Aristotle (e.g. Phys. 195a4-8, 32-5). Presumably, it directs our attention to what the τέχνη is directed towards in such a way that we consider it independently of the particular motivations which led some individual who practises the τέχνη to take up the τέχνη or employ the τέχνη (or other concerns incidental to the practice of the τέχνη).
Cf. Rep. 343d2-344c4; 392b3-4.
The imagery is ubiquitous in Homer, e.g. ποιµένα λαῶν (Il. 1.263; 2.243; 4.296; 10.3, 73, 406); cf. Aristotle, EN 1161a12-15; Haubold 2014, 197.
The vocabulary shifts between discussing what is good (ἀγαθόν, 343b2, 4), what is best (τὸ βέλτιστον, 345c5, d3, 7), and benefit (ὠφέλεια, 346a6, c2, 5, d1; cf. ὠφελέω, 346c5, 7, 9, d7, e1; τὸ ὠφέλιµον, 346e4). These are treated as equivalent to what is advantageous (e.g. 346b1-6, e3-7, 347d6-e2). I assume the shift in vocabulary does not affect the arguments (cf. Prot. 333d8-334c6).
On whether the tyrant is unjust or merely not-just, see Wedgwood 2017. For diachronic considerations (e.g. that the person becoming a tyrant is unjust, but is not unjust when he is a tyrant), see Reeve 1985, 254-9.
Cf. Haubold 2014, 17-19.
Vlastos 1969, 507 n. 8 correctly notes that πλεονεξία is not merely greed, but the desire to have more than others at the expense of others.
Barney 2006, 46, 53 and Wedgwood 2017, 40, 42-3 note that Thrasymachus seems to conceive of goods as being zero-sum. I take this to be correct (cf. Rep. 343a2, 343d2-344c4; 392b3-4) and return to the point below.
It is not clear how the objects of some τέχναι—especially inanimate objects—can be benefited (cf. Euthphr. 13a1 ff.; Pol. 261b7-8; Aristotle, EN 1155b27-31). Accordingly, one might think that the claims should thus be circumscribed in some way, perhaps as applying only to τέχναι which have living beings as their objects (cf. Soph. 219a10-b2).
Perhaps this should read as ‘health or health-related outcomes’, but I shall pass over the fact that Socrates might allow that medicine is responsible for diminishing health (Rep. 333e2-334b6) and that sometimes benefiting the patient involves killing him or her (Grg. 512b1-2; Lach. 195c7-d2). Cf. Nawar forthcoming b.
Reeve 1988, 19; Beversluis 2000, 235; Barney 2006, 52, 56.
‘The wage-earner argument—perhaps the weakest in the early dialogues—establishes nothing’ (Beversluis 2000, 235); ‘the argument does not really do anything to disarm Thrasymachus’ counterexample of the shepherd. Worse, the introduction of “wage- earning” as a distinct craft creates more problem than it solves’ (Barney 2006, 52).
Socrates provisionally treats wages as a genuine benefit, but this is questioned in the discussion of avoiding penalties (Rep. 347a5-e6) and the best kind of people do not in fact earn wages (347a5-e6). That wages have disvalue seems to be motivated by: broader concerns about how wage-earning affects one’s psychological character (cf. Schofield 2006, 250-281); whether a person can be proficient in more than one τέχνη; the notion that justice is doing one’s own thing; and other issues which recur throughout the rest of the Republic (e.g. Rep. 369e3-372c2, 394e1-395c8, 397d1-398b9, 402b5-e3, 406c1-417b9, 419a1-422a3, 428b10-429a3, 433a1-434e2, 438c6-e9, 441d7-e5, 443b1-444b8, 453b1-455a7, 459c9-d3, 493a6-d8, 510c1-511e4, 518b7-521b11).
Thrasymachus should also accept that for each kind of benefit suitably brought about, there exists a distinct τέχνη which is suitably directed towards it (Rep. 346c2-7) for otherwise wages would probably be deemed merely an accidental or lucky by-product of many τέχναι and this would impugn the need for τέχναι while also calling into question their independence from luck.
Cf. ‘very subtle’ (Adam 1902, 48).
E.g. Cross and Woozley 1964, 51-3; Annas 1981, 51-2. Barney 2006, 53 regards it as ‘probably the most confusing and least satisfactory of the series’.
Thrasymachus initially claims that the unjust person is clever (φρόνιµος) and good (ἀγαθός, 349d4-5) and he accepts that the practitioner of a τέχνη is good with respect to those things he clever in, while the non-practitioner of a τέχνη is neither clever nor good in said respects (349e4-9).
This assumes that objective resemblance requires sharing of qualities. Cf. Aristotle, Metaph. 1018a15-18; 1054b3-13.
Cross and Woozley 1964, 52; Annas 1981, 51-2; Reeve 1988, 20.
This is often used by philosophers as a test for ambiguity—see, for instance, Kripke 1977—but it is not especially reliable.
Lycos 1987, 129-30; Reeve 1988, 20; Barney 2006, 53.
Irwin 1977, 181-2 suggests that this is not precisely a matter of equivocation. Perhaps both (i) and (ii) can be captured by one (non-disjunctive) definition and Lycos 1987, 122-3 and Barney 2006, 53 suggest that what is at issue is something like overshooting the mark or going beyond a measure or limit. Weber 1967 offers the most detailed study of the word-group πλεονεκτεῖν, πλεονεξία and πλεονέκτης, but does not clearly resolve this issue.
Reeve 1988, 20; Barney 2006, 53.
Cf. Rep. 351c8, d9-e1; Prot. 322b7.
It is not entirely clear whether (2) supports the view that (in)justice has the same effect in individuals and in groups or whether (2) is supported by this last claim.
I do not here attempt to capture the thought that the ἔργον of a (kind of) thing is that which is done best by that (kind of) thing or with that (kind of) thing (Rep. 352e3-4).
(4) seems vulnerable, but Socrates thinks it is adequately established by the earlier discussion (and argument (γ) in particular). Cf. Prot. 324d7-a1, 326e8-327a2, 329c2-d2.
Socrates does not successfully rule out that a τέχνη might provide something which is neither advantageous for itself nor for its practitioners (but instead, e.g., for some other person) and thus does not entirely succeed in defending premise (1) of his argument for the altruistic nature of τέχνη.
Benefiting the sheep need not, of course, be pleasant to the sheep (Grg. 478b7-9, 521d6-522a7; Pol. 293a9-e5).
For comments on this paper or earlier versions (or perhaps ancestors or cousins) of this paper, I owe thanks to Matthew Duncombe, the audience at the Oxford Workshop in Ancient Philosophy, and the Grundlegung in Groningen.