To: Craig S.
From: Geoffrey Klempner
Subject: Socrates' slave boy experiment in Plato's Meno
Date: 16th February 2010 13:30

Dear Craig,

Thank you for your email of 7 February, with your essay for the University of London BA Plato and the Presocratics module, in response to the question, 'What exactly is Socrates' examination of the slave boy in Meno meant to show?'

According to you, the examination of the slave boy shows four things:

1. The Socratic 'elenchus' can really work, provided the interlocutor is sufficiently willing (!).

2. The method can lead to positive knowledge, not merely to the recognition that one didn't know as much as one thought.

3. That 'learning is recollection' (whatever that means).

4. The special status of geometrical truths as 'paradigmatically certain knowledge'.

I like the fact that you cite 1. which generally gets skipped over in treatments of this example, in particular, the difference between the lazy Meno and the keen slave boy. It's not a big point, but it's worth mentioning.

With regard to 2. I think that you make things harder for yourself by the fact that you consider the paradox of inquiry to be a 'poor argument', which it admittedly is on the face of it, but it has in fact survived 2500 years to take the form of G.E. Moore's 'paradox of analysis'.

There is a problem here, to which the subsequent point about a priori knowledge and 'recollection' is relevant. In order to define 'virtue', one must have some prior notion of what virtue is, so that one can reject proposed definitions as inadequate. But if you already know, why can't you say?

You say that the paradox 'evaporates if we recognize that partial knowledge allows us to proceed with acquiring more'. This is fine for empirical inquiry, say the hunt for the Holy Grail. We might have very wrong ideas about what the Holy Grail is (cf. the Da Vinci Code), but the knowledge we have is a start, and enables us to 'acquire more'. It gives us an idea where to look, or at least how to conduct the research. In the case of knowing what virtue is, by contrast, the reader of the dialogue is expected to know *all* that it is necessary to know in order to critique proposed definitions. It's all there. But how can that be?

Plato's answer: it's 'all there' in the same way as the truths of geometry are 'all there', because it is (in some sense) a priori.

I have my doubts about the success of the experiment. The case of geometry clearly looks very different from the procedure of Socratic inquiry, and Plato's saying that they are basically the same amounts to little better than a pious hope.

But let's move on to recollection. Here, I think, your reference to contemporary innatism and evolution muddies the waters to some extent, although we have to blame Plato for this. It is probably true that the slave boy does so well because of his evolutionary innate knowledge. However, if the truths of Euclidean geometry are provable, then surely someone who lacked this innate knowledge would still be able to prove them, albeit more slowly and with greater effort. (Human beings differ with respect to their capacity for mathematical 'vision', hence talent in mathematics is not just a matter of IQ or brute brain processing power.)

How does the idea of recollection or a priori knowledge carry over to the definition of virtue, or knowledge, or justice? Surely, there is a huge gap here which needs to be filled -- but Plato does say a lot about this elsewhere. For Plato, the universe itself has a teleological structure which is mirrored in the structure of the human mind or soul (cf. the similes of the sun, line and cave in the Republic). That's the cash value of the idea of 'recollection'. Virtue and justice are definable, through Socratic inquiry because the universe itself, ultimately, is 'virtuous' and 'just', or, rather, because aligning our human purposes to the 'order' of the universe is what virtue and justice ultimately are.

As we don't share Plato's view about the ultimate nature of things, there are limits to what one can carry over to the contemporary view of the nature of philosophical analysis. The idea that one is bringing out the innate knowledge of the average English (or French or etc.) speaker is these days regarded as naive and passe, especially since Quine's attack on the analytic-synthetic distinction. Yet a philosopher these days analysing the notion of 'knowledge' or 'causation', or offering a 'theory of justice' proceeds in an a priori fashion, considering thought experiments, putting forward 'hypotheses' for the reader's consideration. Isn't it remarkable that one can do that?

My view of this would be basically Humean: appearances to the contrary, all this is basically logic. One puts forward hypothesis (just as Plato said) and considers what would follow from those hypotheses. Of course, a fair bit of creativity is also required along the way, in inventing/ discovering ways to 'regiment' (Quine) ordinary language.

With regard to 4. your point about non-Euclidean geometry is well taken. But what does it show? I referred above to Plato's metaphysical description of the universe and our relation to it. If human minds 'naturally' reason in Euclidean terms, and the universe is non-Euclidean, then that is indeed a body blow for Plato's perfect picture.

All the best,

Geoffrey