To: Larry B.
From: Geoffrey Klempner
Subject: Gorgias 'On What Is Not'
Date: 5th November 2008 11:49

Dear Larry,

Thank you for your email of 26 October, with your fifth essay for the Ancient Philosophy program, in response to the question, 'Select one argument from Gorgias' 'On What is Not' and discuss its interpretation and validity.'

This is a splendid essay, which sheds light on two important issues: the role of rhetoric today, by comparison and contrast with its role in Ancient Greece; and the problem of identification in forensics, and in particular the questionable reliability of witness reports.

First of all, however, I want to look at the question of the nature of the 'reality' in question for Gorgias.

An obvious interpretation, which I emphasize in the unit on Gorgias, is that what is incommunicable in language are 'sense data', subjective impressions, to take your example the taste of wine.

There are lots of things a knowledgeable wine taster can say about a particular wine. It might even be possible to identify accurately a wine from its description -- if you are an expert and the person describing the wine is an expert. Yet, even so, this would be a matter of luck, depending on the restricted range of wines from which the selection is made. Given any description, however, detailed and expert, there are two or more possible tastes which would be consistent with that description: for example, two possible wines, A and B, such that the description fits A as accurately as it fits B. To know *how the wine tastes* you have to taste it. Words are never quite enough.

While this is important, it is easy to go too in the direction of a 'phenomenalist' view of reality, according to which (as Bertrand Russell or A.J. Ayer in different ways argued) all we 'really' know are our own sense data. But that interpretation would be somewhat anachronistic applied to Gorgias. He is not a sense datum theorist. In any case, there are serious objections to the sense datum theory. Wittgenstein's argument against a 'private language' challenges the naive idea that 'I know what my incommunicable sense data are like for me'.

A different interpretation would view 'knowing reality' not in terms of sense data or subjective experience but rather in terms of our practical engagement with the world. There are many things which we 'know' in this sense which we cannot express adequately in words. The only way to get someone else to learn is to put them through the same training regime that we have been through. Think of any skill which can be taught using words but which requires something else -- practical manipulation of reality -- in addition to the mere words.

In the unit I didn't emphasize enough the importance of the fact that Gorgias is a rhetorician. He isn't just saying that it is difficult to describe reality using language. He is putting forward the daring hypothesis that language NEVER describes things. All 'describing' is persuading, all argument is rhetoric. There is no other use for language than as a tool to make your audience do things that you want them to do.

On the question of witness reliability, we need to reckon with the fact that the human brain is amazingly good at facial recognition (the area of the brain concerned with facial recognition is highly developed) so that we find it difficult to understand how someone who has suffered brain damage has difficulty recognizing people that he or she knows. It is not surprising, under the circumstances, that we are prepared to put strong credence on identification parades, or identification of the accused by a witness in the court room. When I recognize someone I've met before, I *know* without any doubt that it is them.

It is because of this that we can go so badly wrong. One of the ministers in the current UK government, Peter Hain, was once arrested and prosecuted for bank robbery -- in the days when he was an active Anti-Apartheid activist. He was eventually acquitted. It is possible, although this has never been firmly established, that he was set up by the South African secret service. The bank robbery was committed by someone who looked a lot like Peter Hain, was dressed in similar clothes, at a time when Hain was in the vicinity.

However, it might be claimed that this is a different point from the problem that Gorgias is raising about language. I can let you know how the wine tastes to me by offering you a glass. The witness can describe the perpetrator to the police or point him out. The offering, or the pointing, in some sense bypass language.

Having said that, it could be argued on the other side that human memory is to a large extent linguistic. I remember the robber's face. I have a hazy image in my mind's eye. I also have words to describe his face. Here's a thought experiment: suppose I ask you to imagine a tall, heavy-set man in blue overalls holding a large spanner. What colour is the man's hair? It wasn't necessary to 'colour in' the man's hair in your mental image (although of course you are doing that now). Or, how many pockets are on the overalls? Are they light blue or dark blue, dirty or clean? If imagining was the same as 'painting an image in your mind's eye' then there would automatically be answers to these questions.

A similar thing can be said about remembering. A memory image of something we have seen is not like a photograph, not even like a faded photograph. It has a linguistic aspect. At some point, the strong non-verbal ability to 'know the person if you saw them again' shades into a mere description which you give yourself. The perpetrator was dark and heavy set. The accused in the identity parade is dark and heavy set.

Finally, congratulations on completing your program! Please confirm your current address for your report and Pathways certificate.

All the best,

Geoffrey