Republic Book 1

The Republic starts off in pretty classic Socratic fashion.  A short preamble conversation with old man Cephalus on the benefits of again and wealth soon transitions into the main question: "what is justice?"  In the remainder of the chapter various interlocutor's offer definitions of justice that Socrates investigates with his customary questions.  Some of Socrates' arguments feel unusually sophistical, or rhetorical as we might say now, compared to his normal line of questioning.  So perhaps it's not surprising that Book 1 ends in aporetic fashion, with Socrates announcing his disappointment with the fact that he got sidetracked.  Distracted by the verbal combat, he never got to the main question of what justice is, and instead simply investigated whether we should expect it to be beneficial or not.  

The conversation begins promisingly enough.  The first thesis belongs to Cephalus' son Polemarchus, who argues that justice is, "to give each what is owed him."(331e2)  He has in mind the classic Greek idea that you owe your friends good and your enemies harm.  As you can expect, this soon leads to some odd conclusions about justice.  It seems that this kind of justice would be most useful in a war or some other type of conflict situation (332e4).  In peacetime, it would boil down to honesty in dealing with your business partners (33b7).  But this type of honesty really just knows how to safeguard money and give it back, since using money to make a profit breeding horses or building boats requires, not the skill of justice, but knowledge of horse breeding and boat building.  In short, this type of justice seems pretty useless for daily life (334e1).  

The conception of justice as a particular skill or craft, a techne, runs throughout the first book.  Making justice a type of craft immediately implies that it involves some kind of knowledge. Socrates will make use of this underlying idea a number of times in his arguments.  It appears for the first time when he forces Polemarchus to reckon with what happens when people misjudge who is their true friend or enemy (334c5) -- we wouldn't consider it just to harm someone who is truly our friend even if they don't appear that way at the moment.  Whenever there is knowledge in Plato, there follows a distinction between appearance and reality, between opinion and truth.  Which means that Polemarchus is forced to restate his thesis about justice to include both an apparent or relative relationship (friend versus enemy) as well as a true or absolute one (good versus bad).

So you want us to add something to what we said before about justice, when we said that it is just to treat friends well and enemies badly. You want us to add to this that it is just to treat well a friend who is good and to harm an enemy who is bad? (335a6)

Once we have opened an absolute distinction between good and bad, it's a simple step to ask what makes someone good or bad, that is, that in virtue of which they judged good or bad.  And with this question we complete the Platonic equation of all the good nouns: Justice = Virtue = Knowledge = Wisdom = Benefit = Goodness.  Justice is human virtue (335c3), arete.

Arete is broader than our notion of virtue, which tends to be applied only to human beings, and restricted to good sexual behavior or helpfulness on their part to others. Arete could equally be translated "excellence" or "goodness." Thus if something is a knife (say) its arete or "virtue" as a knife is that state or property of it that makes it a good knife—having a sharp blade, and so on. So with the virtue of a man: this might include being intelligent, well-born, or courageous, as well as being just and sexually well-behaved. (tranlator's note pg. 980)

So really almost by definition, justice is what makes people good.  But in that case, Polemarchus' definition falls apart completely.  Because you can't make someone good or even just better, by doing them harm (335d1).  If just people are good, they possess a peculiar virtue for producing goodness.  In which case they're not going to go around producing badness by harming people, even if those people are an enemy and bad (335e1).  

Judging from the conclusion of the chapter (and the much more confrontational dialog that follows with Thrasymachus) I think we're supposed to be vaguely aware that Socrates is pulling a fast one on us here.  In pulling apart the usual Greek formulation of justice as 'just deserts' he seems to have assumed the conclusion as the starting point -- justice is about some universal good, in fact, it's the very quality that makes us good.  But this subtly shifts the question from what justice is, in itself, to what benefits it has, for us and others.

Thrasymachus, however, was not fooled.  He comes storming into the dialog with the post-Nietzschean contention that justice is really just the name we give to whatever is to the advantage of the stronger (338c1).  We've seen this same type of argument in the mouth of Callicles in Gorgias -- basically, might makes right.  Here, Socrates' refutation of Thrasymachus takes a different form than in Gorgias, but it still revolves around the already introduced idea that justice is some sort of craft that involves the knowledge of what is truly good for us.  At first it seems that Thrasymachus will be caught out by the same problem that plagued Callicles' and Polemarchus' definition of justice -- what if the stronger make a mistake about what is truly good for them?  If the ruler accidentally and unknowingly does something that is not to his advantage, we don't want to call that justice, do we?  But Thrasymachus refuses to be led down this path, and defines the stronger, the ruler, as knowing what is to his advantage.

When someone makes an error in the treatment of patients, do you call him a doctor in regard to that very error? Or when someone makes an error in accounting, do you call him an accountant in regard to that very error in calculation? I think that we express ourselves in words that, taken literally, do say that a doctor is in error, or an accountant, or a grammarian. But each of these, insofar as he is what we call him, never errs, so that, according to the precise account (and you are a stickler for precise accounts), no craftsman ever errs. It's when his knowledge fails him that he makes an error, and in regard to that error he is no craftsman. No craftsman, expert, or ruler makes an error at the moment when he is ruling, even though everyone will say that a physician or a ruler makes errors. It's in this loose way that you must also take the answer I gave earlier. But the most precise answer is this. A ruler, insofar as he is a ruler, never makes errors and unerringly decrees what is best for himself, and this his subject must do. Thus, as I said from the first, it is just to do what is to the advantage of the stronger. (340d)

In a way, Thrasymachus is pulling his own fast one here.  Everyone agrees that justice is something good, something which produces a benefit.  Socrates goes in circles by defining justice as what makes us good, without first defining goodness.  Thrasymachus, meanwhile, defines goodness equally vacuously, as just whatever the ruler does, with the presumption that this will always be to their benefit if they truly know the craft of ruling.  Neither definition really gets at the core of the matter, which is the same Platonic question as always -- how do we know what's good for us?

Perhaps Socrates response to Thrasymachus feels like a bit of sophistry because this main question is not at stake; he's just trying to win the argument now and not investigate the deeper issue yet.  The response falls into two parts.  

In the first, Socrates makes an analogy between the craft of ruling and other crafts like shepherding or medicine.  In themselves, these crafts aim to benefit only the thing that is the object of the craft.  Shepherding benefits sheep.  Doctoring benefits the body healed, etc ... In other words, 

... no craft or rule 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. (436e3)

This line of argument that every craft is, in itself, completely generous and altruistic, seems faintly ridiculous.  Indeed, to account for why we need to pay doctors and shepherds, Socrates is immediately forced to introduce the idea of a craft called 'wage-earning'.  The resulting picture is of a doctor who looks out for his own benefit by employing the craft of wage earning, which in turn focuses solely on maximizing his wages by employing the craft of doctoring, which, in itself, has nothing to do with either wages or the benefit of the doctor, but focuses entirely on healing patients.  These dubious artificial distinctions do allow Socrates to make a real point, however, about the analogous craft of ruling.  Rulers don't rule for their own benefit.  And since it's considered shameful to rule just for the wages or the honor, rulers can only be induced to rule because it allows them to avoid a peculiar type of punishment -- being ruled by some incompetent moron.  

... if they're to be willing to rule, some compulsion or punishment must be brought to bear on them—perhaps that's why it is thought shameful to seek to rule before one is compelled to. Now, the greatest punishment, if one isn't willing to rule, is to be ruled by someone worse than oneself. And I think that it's fear of this that makes decent people rule when they do. They approach ruling not as something good or something to be enjoyed, but as something necessary, since it can't be entrusted to anyone better than—or even as good as—themselves. In a city of good men, if it came into being, the citizens would fight in order not to rule, just as they do now in order to rule.(347c1)

I suspect we'll see this idea of the reluctant ruler again.  If the just ruler has to spend all his time looking after his subject like a shepherd looking after his sheep, we're going to need some explanation of how this benefits him.

The second part of Socrates response to Thrasymachus feels even more like sleight of hand designed only to win the argument.  He preys on Thrasymachus' image of the just person as an altruistic pushover to establish that the just don't compete amongst themselves, but want to be rewarded for their justice by finishing ahead of the unjust.  For the unjust, by contrast, it's every man for himself.  The same type of structure would apply to the knowledgeable and ignorant as well.  Two knowledgeable=clever=good scientists don't compete with one another to find a 'better' truth, instead they aim at agreement and equality in the face of this truth.  Meanwhile, ignorant=bad people are always trying to look better than both the knowledge folks, as well as the other ignorant folks.  You can see where this is going.  Socrates completes the syllogism and corners Thrasymachus into admitting that since they compete against unlike rather than like, the just must be knowledgeable and good, whereas the ignorant, who compete against both like and unlike, must be bad and unjust.  It's a bit of a cheap shot, argumentation wise, whose only real contribution seems to be to reinforce the equation of true=clever=just=good that we've already seen.

In the final pages of the chapter, Socrates changes gears a little and adds one more term  to the equation -- harmony (though he doesn't explicitly use this word yet).  Nietzsche's disciple Thrasymachus believes that injustice is powerful, particularly when it is so thorough-going that it leads one to try and tyrannize a whole city.  Socrates wants to find a way to undermine the equation of injustice with power.  So he argues that even a terribly unjust band of marauders is only made powerful by the internal cohesion of their common purpose, which must be based on some sort of internal justice amongst the members, like say, a rule for the common division of spoils.  If the unjust were completely unjust in every way, they wouldn't ever be powerful enough to cooperate, so they would remain relatively powerless.

We have shown that just people are cleverer and more capable of doing things, while unjust ones aren't even able to act together, for when we speak of a powerful achievement by unjust men acting together, what we say isn't altogether true. They would never have been able to keep their hands off each other if they were completely unjust. But clearly there must have been some sort of justice in them that at least prevented them from doing injustice among themselves at the same time as they were doing it to others. And it was this that enabled them to achieve what they did. (352b8) 

In Plato, it almost goes without saying that harmony is a virtue to be added to our equation.  Just in case though, Socrates makes it explicit that harmoniously ruling over and coordinating things is precisely the function of the soul (353d3).  Already we can almost hear the resonance of Soul and Form we found in Phaedo.