To improve the method of Socrates we must divorce it from Plato’s epistemology, and we must replace it with something that allows investigation of justice to proceed in basically the same way, otherwise we won’t be modernizing the method of Socrates, but simply creating something new of our own and attributing it to him. With that in mind a suitable replacement must answer the four questions that were raised by our initial examination of the method, which will allow us to leave the method, as applied, completely unchanged, so let us start there.
All four questions are resolved by Plato’s epistemology because it explains how considering an ideal city gives us better access to the concept of justice. And so we must replace it with something that plays essentially the same function. While there are a number of possibilities I would also like to take my cue from the methods of Thrasymachus and Glaucon and ground the method in a claim about how meaning works. With those constraints in place we are forced to adopt as a foundation a claim about meaning that entails that there is an essential meaning of words, and that we are better able to use this meaning, and thus identify it, in fictional situations where there are fewer complications, fewer distractions that prevent us from applying the essential meaning without being swayed by the other associations and connotations we have come to possess in addition to it. Since it is not my intention to seriously defend this claim about meaning there is no need to provide more details, for it has already served its purpose. With this theory of meaning in hand we can answer the four questions raised by the method in basically the same way as Plato’s epistemology answered them: considering the ideal city makes us better able to understand justice, as well as the other virtues, because it allows us to concentrate on their essential meanings. The only difference is that the method is no longer “infallible”, the essential meanings are part of our mind, not a non-physical world, and so there is no guarantee that the essential meanings are “right” in the same way that grasping the forms was supposed to be.
Now it is time to critically evaluate these methods. With this revision all three are on a relatively even footing, although they all require commitments to a specific view about how meaning works none of the proposals are particularly metaphysically outlandish. And certainly we can’t investigate one on the grounds of the others. The problem with philosophical methods is that there is no easy way to judge them. Obviously we can’t judge them by their results, meaning the positions they lead us to take on various philosophical claims, because that would require us to have some superior method that allows us to definitively determine what positions we should be led to adopt. Similarly, most philosophical methods will endorse themselves and thereby reject the others. For example, if we were working under ordinary language philosophy we might set out to investigate how reference works by observing what people claim that they are referring to (because we can’t directly get at the internal meaning of words empirically), and this will support the conclusion that what words refer to, or mean, is the things that they are most often used to signify. And if we believe that we are to search for the essential meanings of words then it is easy to believe that our meditations on meaning itself, in its most abstract forms, will lead us to believe that we are correct to investigate them in this way.
But, while it is impossible to determine which method is “correct” in a positive fashion, by investigating and coming to a definitive answer about what the method of philosophical investigation is, we can certainly take a page from Socrates and investigate it in a negative fashion, and demonstrate the shortcomings of the methods. While I cannot raise every possible objection I can produce a single strong objection that throws all three methods into question. First let us consider whether we want our philosophical method to produce the truth or mere opinion. It is so easy to create opinions that it seems unlikely that we would rigorously and philosophically investigate issues just to produce one more. And so it seems that we must be after the truth, the true answer to “what is justice?” But it is not clear that any of the methods under consideration can claim to produce a true answer to this question instead of a conventional answer. Certainly how words are used may change over time, in fact we know that they do, and so the methods of Glaucon and Thrasymachus, having a basis in ordinary language philosophy, do not produce an eternal truth about what justice is, but only what justice is currently considered to be. The revised method of Socrates also faces this problem, we have no reason to believe that people reflecting on ideal scenarios will come to the same conclusions about what the essential meaning of the word is. Indeed I myself, upon reflecting on Plato’s ideal city, see in it not justice but injustice, in the way individuals are allowed to do only what they are deemed best at, and not given the freedom to choose for themselves. And so the essential meaning produced by the method seems as much a matter of opinion as the definitions empirically discovered by the previous two methods. Now this may seem like an argument to simply keep Plato’s original epistemology, where it was at least claimed that we could grasp the truth itself, but that version of the method is not any better position to deal with this objection. Even when considering abstract scenarios, where they are supposedly closer to the forms, people may disagree about what justice is, as I am an example of. And when that happens there is no way to decide who is right, since we may both claim that the other is in error, and that we are closer to the forms, and hence the truth. And thus which answer to accept is a matter of opinion, and not truth.
And so, upon critical examination, I would conclude that we should reject all of the methods for discovering what justice is used in the Republic. Which is not to say whether their conclusions are true or false because, as they say, even a stopped clock is right twice a day. If we can trust our intuitions to some extent and if the conclusions in the Republic seem intuitive then we might tentatively accept them, but we should keep in mind that we are accepting them on the basis of intuition and not because of a method that makes them more or less likely to be correct then any other intuitive claim.