Matthew asked:

In simple terms, could you explain the point that Strawson makes in his article ‘On referring’ against Russell’s theory of descriptions. Who’s right, Strawson or Russell?

Answer by Craig Skinner

Strawson made several points. I’ll deal with the most famous one.

Russell, after his success in marrying logic and mathematics, believed that formal logic could clarify ordinary language, helping solve or dissolve problems and confusions, and was a big advance as a method in philosophy.

Strawson thought formal logic had little to offer over ordinary language: logic took a narrow view – what the words mean (semantics) – ignoring speaker meaning and conversational implicature (pragmatics), and, in any case, got the semantics wrong.

So Strawson sought to show that Russell’s approach gave the wrong answers, whereas ordinary language considerations got it right.

The focus of the dispute was the truth-value of statements about nonexistent objects (statements which fail to refer). Example:

‘The present King of France is wise’.

Russell held this to be false, there being no present King of France.

Strawson felt it had no truth-value i.e. was neither true nor false. If we accept this, it is an affront to classical logic where every statement is either T or F, and would derail Russell’s programme.

Part of Strawson’s case was that nobody would make the statement if she knew that the King of France didn’t exist. To do so would violate a presupposition of statement-making so that no statement was really being made, and there was therefore nothing to be T or F.

Aside: others, mistakenly, said that if ‘The present King of France is wise’ is false because there is no King of France, then its opposite ‘The present King of France is not wise’ is also false for the same reason, so we have a statement and its opposite both F when of course the opposite of a F statement must be T. But this is just a logical blunder: both the statements are indeed F (according to Russell) but they are not opposites. The opposite of ‘The present King of France is wise’ isn’t ‘The present King of France is not wise’, rather it is ‘It is not the case that the present King of France is wise’, and this is clearly T.

In my view Russell is right. His theory gives correct and consistent results without any need to modify classical logic by introducing truth gaps. Also, ordinary language considerations, in my view, don’t favour Strawson. To me, statements such as ‘The present King of France cuts my lawn’ or ‘My sister is dating the present King of France’ are clearly false, not, as Strawson had it, lacking truth-value. Also, we do, knowingly, make statements about nonexistent things all the time (Santa Claus, Hamlet, Vulcan for example) and we do mean something by such talk.

However, Russell’s programme delivered less than he hoped for. His, and the early Wittgenstein’s, foray into logical atomism applied to ordinary language failed. As a logician/mathematician, like Frege before him, Russell concentrated on what the words mean, whereas Strawson, Grice, Austin and the later Wittgenstein saw that what the speaker means is wider than this, and may be different, that language is a public ‘game’ with rules, meaning is related to use, and there can be no ideal logical language.