Aim is to help Plato against two charges:
1) his defense of justice commits a 'fallacy of irrelevance': the defense of 'Justice' that Plato offers does not defend anything that is recognizable as conventional justice.
2) he gives people the wrong reason to be just. They should be just for egotistical self-centred reasons, not for its own sake.
Dahl maintains Plato is committed to a form of 'intuitionism' (the belief that (at least some) moral judgements are known to be true by intuition.)
Plato is not agent-centred, but act-centred, Dahl will conclude.
Dahl has a good way of formulating the three classes of goods from 357-8:
1) Goods which are good "in virtue of what is due to those goods themselves, and not for what arises from them together with circumstances that normally accompany them."
2) goods which are good "both in virtue of what is due to those goods themselves and for what arises from them together with circumstances that normally accompany them."
3) goods which are burdensome in themselves but are good for what arises from them together with circumstances that normally accompany them.
That formulation has the advantage of pointing out what Irwin wanted in his article: namely the claim that Plato can claim that justice is good for its own sake and for the sake of happiness, and yet may still belong to the first class. That is because " circumstances that normally accompany them" can be taken to refer to circumstances external to the agent (I assume: note that that leaves a gray area insofar as the distinction between agent and circumstances may be unclear).
Plato wants to show that justice belongs to class 2, but to do that, he has to show that justice is worth choosing even if one gets the usual rewards and reputation of INjustice, and even if the unjust person gets the rewards and reputation for justice.
For Plato, virtue (arete) belongs to a thing that has a characteristic function (ergon): virtue makes that thing perform its function well. So human virtue makes humans perform characteristically human activity (or activities) well. A 'characteristic activity' is one that a given thing alone does, or that the given thing does better than other things do it.
So virtue enables a person to live a good life, a life of eudaemonia.
(Those are all meant to be analytic statements)
Virtue is supposed to make one act well and be well off (as Thrasymachus' claim that injustice is a virtue because it makes one well off shows (348-9)).
Plato is 'internalist':full understanding of virtue provides one with a MOTIVE to be virtuous.
So Plato wants to give people a MOTIVE to be just and to claim that just people are BETTER OFF.
In order to do so, Plato 1) shows what justice is, and 2) argues that the just person is better off than the unjust person.
1) What is justice? First, Plato says there are three parts to a state, each having its function, and justice is when they all perform their functions well. Then he argues that states are like individuals. The soul has three parts: the appetite provides for bodily needs; reason rules; and spirit enforces reason's rules when they conflict with appetite. Justice is harmony of these functions. After exploring the various possibilities for lack of harmony and harmony, Plato argues that the just person, whose parts are in harmony, is happier than the unjust person. The least just person is a slave to his or her appetites (576-7) and is never able to satisfy them (579). The just person prefers the pleasures which the just life brings, and is the most happy (580-83). Pleasure is defined and the just person has the most.
So, you can question:
1) that the individual and the soul are relevantly similar
2) that pleasure is what Plato says it is
3) whether philosophers are the best judges of pleasures
David Sachs, 'A Fallacy in Plato's Republic' (Phil. Rev. 1963, 141-58) claims that Plato set out to talk about conventional justice, which is act-centred, and includes what you would expect (don't lie, tell the truth, take care of your parents), but winds up talking about Platonic justice, which is agent-centred (justice is whatever the person with the harmonious soul does: acts are just derivatively). Then Sachs says that unless Plato shows that conventional justice and Platonic justice are appropriately related, Plato has not done what he set out to do: all his argument is IRRELEVANT.
Plato must show that a conventionally just person will be Platonically just and vice versa. Plato claims the vice versa part of that wrongly, says Sachs, and fails to show the other part: someone may be conventionally just for all the wrong reasons, and Plato does not want to claim that they are just.
Sachs further claims that the MOTIVE for a Platonically just person to be just is self-centred, and so it is not a praiseworthy reason, and is therefore the WRONG REASON. It is egotistical.
Dahl next goes through the versions of justice in Book 1: justice is telling the truth and paying one's debts (Cephalus at 331): justice is giving everyone their due, helping one's friends and harming one's enemies (Polemarchus at 331-2); justice is doing what is in the interest of the ruling class (Thrasymachus); justice is following a social contract, a compromise between the best state of affairs (doing what you want?) and the worst (not doing anything you want) (Glaucon).
Dahl claims that the initial understandings of what justice is are INITIAL: Plato never intended to explain conventional justice nor did he ever claim to. Thus Sachs' objection is misplaced.
Furthermore, Socrates never claims to be going to defend justice in terms of any particular definition: he can reasonably take his first task to be to figure out what justice is.
Plato thinks conventional justice is at most partially correct, because he uses an instance of it (don't return a knife to a crazy man) against Cephalus (331). He claims that harming anyone is unjust to refute Polemarchus (335): thus he rejects part of conventional justice (harming enemies).
Dahl's reformulation of what Plato needs in order for the argument of the R. to work:
1) Platonically just people must to a significant extent be conventionally just, and
2) performing actions that are to a significant extent conventionally just must be a necessary part of the means needed to be Platonically just.
3)the class of conventionally just actions should turn out, to a significant extent, to promote or preserve soul-harmony, Platonic justice.
4) He must do all of this without committing himself to a theory of justice that involves reference to sensible things.
Plato must show 1) in order to have a plausible account of justice.
Plato must show 2) in order to provide people with a motive for acting justly.
Plato must show 3) in order to show that genuinely just actions are called just by Plato's theory of justice.
Plato must show 4) in order to prevent his theory of justice from being unnecessary or mistaken.
Plato wants to avoid using sensible actions in his definition of justice, because any such definition would be subject to the argument he uses against Cephalus: any given type of just sensible action will turn out to be unjust in some instance. That, Sachs thinks, explains why Platonic justice is agent-centred rather than act-centred.
Dahl thinks that we can find a description of Platonically just ACTS: they are those acts which have the features that make them desireable to people with harmonious souls. But if Plato can argue that such a class of actions exists, he will have rendered the theory of justice which he propounds unnecessary, and perhaps mistaken, for his agent-centred theory will have a rival counterpart that is act-centred.
Sachs claims that soul harmony only leads to courage, intelligence, and self-control, and that is not sufficient to meet condition 1 above, because they are compatible with widespread (conventional) injustice.
Three parts of the soul (as seen in Book IV):
the appetitive part desires things that are desired independently of what one thinks is good (although they MAY be good)
reason desires things that are required by one's overall conception of the good
spirit desires things that involve one's conception of the good but perhaps not one's overall conception of the good.
Given only that, reason could rule, and one could still be unjust: one's overall conception of the good might be off.
But Plato thinks each part has its own objects of desire:
appetite desires food, drink, sex, and wealth
spirit desires honor and political success
reason desires knowledge and truth (580-81)
That implies another aspect to reason's rule: one's conception of the good must include truth and knowledge as pre-eminent goods if reason really rules.
So reason rules in that 1) one acts in accord with one's conception of the good, AND 2) one's conception of the good values knowledge and truth preeminently.
The second aspect of reason's rule is Platonic justice, says Dahl, relying on 442 (and the fact that a) justice in the state, as discussed in Book VIII, relies on this sort of reason's rule; and b) this kind of reason's rule is what makes a just person happier and better off in Book IX).
The first aspect is temperance.
So the just person, whose reason rules in the second way, will do conventionally just things to a significant extent, and their motive will be to promote soul harmony. Plato can argue that without begging the question of whether the Platonically just person will do conventionally just things, because his argument never relies on conventionally just things (in other words, he fulfills condition 1 and 2 by arguing to them without assuming they are true).
The Platonically just person might still commit conventional injustice (e.g. steal), but Plato argues that doing so habitually will strengthen an irrational desire, and thereby eventually disturb soul harmony by disturbing the person's conception of the good (558-61 and 588-9). Thus one NEEDS to act in a conventionally just way to preserve soul harmony (condition 2 above).
There might be problems for someone incapable of being a philosopher: how could they be just? They are presumably capable of some kind of soul harmony, and so of justice.
Condition 3 is met too-if one needs to act conventionally justly in order to preserve soul harmony, then it will preserve soul harmony to do so.
Condition 4 is met as well: knowledge- and truth-related actions are not to be identified by sensible characteristics (such actions are specified via the Forms), and so Platonic justice does not commit him to a rival theory of act-centred justice which threatens his agent-centred theory.
What if a philosopher is broke? She needs money to have the leisure to pursue the greatest good, knowledge and truth. Won't she steal to do so? Then conventionally just actions will not significantly overlap with Platonically just ones, and Platonic justice will not give her a motive to be conventionally just.
What of the philosopher who refuses to rule because she wants to contemplate the Forms? Plato says she would be unjust to do so! (520-1)
The desires of the rational part are not structured in such a way that they lead one to value only the intellectual life and nothing else.
The contemplation of Forms leads one to want to make sensible things as much like the Forms as possible, and so it leads the philosopher to DESIRE to instantiate the Forms, including Justice. Such a desire leads the philosopher to conventionally just actions when faced with the situations in section VI, and gives her a motive to do them.
Further problems with the relation of conventional to Platonic justice.
Platonic justice is found in souls, not acts.
Conventionally just actions do not necessarily promote the soul harmony (Platonic justice) of others: stealing money might in fact lead someone to the intellectual life, and so conventional INjustice could lead to Platonic justice.
BUT there is something in the preceding objection which we need to notice: in stealing to promote Platonic justice, the person's motive is an impartial desire to instantiate Platonic justice, the Form Justice. That looks like a desire to be Just for its own sake, which is not egotistical: thus we can meet one of Sachs' objections to Plato's theory (that it gives people the wrong reason to want to be just: it gives them a self-centred motive to want to be just (see section III)). If people SHOULD want to be just for its own sake, then Dahl has all he needs to answer Sachs' objection.
In this section Dahl engages in some fancy argument that I cannot follow: the Form of Goodness, he argues, leads people to instantiate goodness, and that promotes harmony in the soul. Platonically good actions are Platonically just actions, and neither involves Plato in giving an account of actions that includes sensibles.
I begin to understand Dahl at the top of Page 228 again.
Platonically Just people want to instantiate the Good, which is a desire to do something good in itself. Doing things because they are good in themselves is a praiseworthy motive, and so Plato is once again arguing that people should be just FOR THE RIGHT REASONS, not, as Sachs maintains, for the wrong reasons, such as egotistical reasons.
Plato's theories are similar to the 'intuitionist' views of early 20th century philosophers. 'Intuitionism' can be briefly defined as the theory that (at least some) ethical judgements are known to be true by intuition (Antony Flew's definition in "A Dictionary of Philosophy").
The similarity is in that Platonic Justice involves things being just because they exhibit an inarticulable goodness that is like harmony and order. To see that this is so, one needs to understand the Form of Goodness, and one can do that after training and education. Once one is trained, one can use one's intution to see if an action is Good. Such actions overlap with conventionally just actions.
AHA, I SEE IT: Dahl is saying that Platonic Justice is only found in souls, but Platonic Goodness is also found in actions, and so if Platonically just people's actions are Platonically Just only because the Platonically Just person did them, but they are the same as Platonically Good actions, we can say that Plato in fact has an act-centred theory of justice, because Goodness can apply to actions.
So Plato has within his theory of Justice a commitment to Just actions being Good actions. And Good actions are things external to the agent. The chief advantage of agent-centred theories of justice is that they account for complexities better than a list or description of actions could, AND they explain why one should be just by appeal to a person's nature. If Plato is committed to a theory that does not necessarily appeal to a person's nature, he is committed to a theory that is less attractive, because people are less likely to see that they should want to be just.
The danger is that people might see the Good actions that are good independent of them, and so external to them, and yet not want to do them. Plato will reply that Forms are the fullest expression of being, and so the person who knows them will want to come as close to them as possible: they will want to instantiate the form of the Good and the form of Justice.
Without Forms, however, Plato would be on shaky ground, because he would be left arguing that people should be just for egotistical reasons, and he would be left arguing for actions that do not, to a significant extent, exemplify conventional justice.
Another danger is that it is hard to explain how we can resolve instances where people disagree about their evaluation of something as just or unjust. What if people disagree. How can one decide. If the Good is inarticulable, then one cannot compare arguments.
Plato's response again involves the Forms: they are fixed, and there will be no evaluative disagreements among those who understand the Forms.
But that smacks of argument by fiat: Plato has created Forms to do this task, and so they must do it. It does not really tell us how they do it or guarantee to us that they will always produce the same result in different persons.
A further danger is that the Good is impersonal, and so fails to give individual people their appropriate value and fails to adequately include people's 'rights.' Plato does seem to be inhuman in his politics at times. 'Rights' involve recognition of the unique value of individuals, and Plato has little room for that.