Saturday, 9 May 2015

Putnam on Context-Dependence & the Formalisers

Hilary Putnam offers us some philosophers who've defended the notion of context-dependence when it comes not only to meaning; but also to truth-conditions. Indeed, if we have one, then we must have the other. The contextuality of meanings determines the fact that truth-conditions are also contextual in nature:

“… Wittgenstein and Austin well before us, would argue that sentences do not normally have context-independent truth conditions. It’s the meaning of the sentence or the words plus the context that fixes the truth conditions.” (232)

Putnam means that we can't even begin to ‘fix the truth conditions’ of words or sentences if we haven’t already fixed the meanings of those words or sentences. And that fixing of meaning will be a contextual matter. That is, we must move from the

  1. context-dependent meaning of ‘cat’ or ‘The cat is on the mat’.
to the

  1. context-dependent truth-conditions of the word ‘cat’ or the sentence ‘The cat is on the mat’.
Indeed we can't even have a meaning (or ‘sense’) of the word ‘cat’ unless we also have the 'sense' of the sentence ‘The cat is on the mat’. (Or another sentence with the word ‘cat’ in it; at least according to Frege’s 'context principle'.) As with salient or relevant causal facts, we must also determine which truth-conditions to use. (Think here of Putnam’s ‘cherry on the tree’ example.)

Not only that: even according to Tarski’s Convention T, all truth-conditions really are the named sentences disquoted, as in

(T) S is truth iff p.


(T) The sentence ‘Snow is white’ is true iff snow is white.

If the truth-conditions are the disquotation of the former named sentence, and that sentence requires a meaning that is context-dependent, then it follows that the truth-conditions (or the sentential disquotation) will also be context-dependent: dependent not only on the context-dependence of the meaning of ‘snow is white’; but also on the context-dependence of our choosing the truth-condition snow's being white (which isn't, however, really a choice we can make with Convention T).

Putnam believes that one of the main reasons for the denial of context-dependence (or context-sensitivity, as in Frege, Dummett and others) is the desire to make philosophy (or at least semantics) more scientific and logical in nature. Thus more capable of securing determinacy of sense and the objectivity of language and meaning. As Putnam puts it:

“… philosophy stands almost entirely apart from [linguistics, semantics, and lexicography], giving much too much significance to ideal language, mathematical logic and all that.” (232)

In the first half of the 20th century, most philosophers in the analytic tradition virtually ignored relevant work in linguistics and semantics and most definitely in lexicography. Even Strawson - who was one of the first to reject the ideal of an ideal language (or even its possibility or use) - wouldn’t have spent much time on (pure) linguistics and (pure) semantics, let alone on lexicography!

These philosophers were dealing primarily with abstract entities like senses, meanings and propositions. They did so with the tools of mathematical logic. Thus it's hardly surprising that such philosophers had no time for linguistics, semantics and certainly not for lexicography. To think that a linguist (or semanticist) - let alone a lexicographer - could have told them the true sense or proposition behind an expression would have seemed outrageous or ludicrous to them. These are, after all, empirical disciplines; whereas philosophy, logic and mathematical logic are a priori disciplines which deal with what must be taken as a platonic realm; even in the cases when the philosopher isn't strictly speaking a platonist.

This antipathy towards the ideal of an ideal language is neatly expressed by Putnam:

I think that we still suffer from the idea that formalising a sentence tells you what it 'really' says. Perhaps we are now doing something similar with Chomskian linguistics.” (233)

It's not just arrogant for a philosopher to tell us ‘what we really mean’: it's also wrong. It can't be done. What they tell us (if they tell us anything about our actual meaning) is what we should have said or meant. And that normativity should - or would - have depended on the ontological and logical commitments of the philosopher concerned.

Thus Bertrand Russell is well known for telling us what we really mean when we say that ‘The king of France is bald’. What Russell really meant by that is that we're making philosophical and logical mistakes in such a formulation. Or, alternatively, that we're committed to things which we don’t realise we're committed to. Thus he was telling us what we should say and what we should be logically/philosophically committed to. Not that we are logically/philosophically committed to X. And Russell certainly wasn't telling us what we really said or meant.

In most cases (though not all), if we had wanted to say or mean something else by what we said, we would have said something else.

Would we say, for example, that when a young child says that ‘4 + 2 = 7’ that he really meant to say ‘4 + 2 = 6’? What if someone says that ‘Tony Blair is a shit’ and the philosopher says that he really meant ‘Tony Blair is a bad Prime Minister’? Perhaps that’s what he should have said. But he didn’t mean that; otherwise he would have said it and meant it.

The formalising of a sentence (whether by Russell or Dummett) is thus a little like psychoanalysis. The psychoanalyst tells us, for example, what our dreams or our locutions really mean; even though we don't know what they really mean. Indeed almost by definition the psychoanalyst (in these cases) believes that we mustn't (or can't) really know what we mean by what we say (or what our dreams mean).

Thus does the formalising philosopher also tell people that (almost by definition) we can't know what we really mean because we haven’t got the philosophical and logical skills to fully know (or know at all) what we really mean. Similarly, the psychoanalyst also has skills (or so he thinks) that we don't have and is thus automatically entitled to tell us what we really mean by something or other. Who knows, perhaps their psychoanalyst is being normative in that he's really telling us not what we really mean; but what we should mean by what we say and do.

It's strange, then, that on this interpretation of what the formalising philosopher does, he's essentially dealing in normativity when it comes to formalising our sentences. That is, he's telling what we should mean given the ontological and logical implications of what we do actually say. Not only that: this also means that we should be ontologically and logically committed to things that we aren't at present committed to.

Thus the normative dimension of formalising philosophy appears to be even more normative in nature. Instead of discovering the deep truths about propositions and meanings, the formalisers are really offering deep truths about themselves or about what it is they're logically and ontologically committed to.

If that's the case, perhaps they should have come clean with both others and themselves and told us that this is what in fact they were/are doing.

Another way to put all this is to agree with Wittgenstein and say that there are no deep truths not only about what we say; but also in philosophy as a whole. Everything "is on the surface". Any depths there are belong to the minds of the philosophers who create those depths. If they are only illusory depths (perhaps “cast by the shadows of our language”), then we have good reasons to ignore what the philosophers tell us about what we really mean. If anything is deep, it's their minds. In that case, perhaps psychoanalysts should cast their nets into the deep depths of the minds of formalising philosophers and then they could tell us what they, not us, really mean.

Putnam then offers us an interesting and ironic take on these philosophical formalisers. (Indeed perhaps some of these formalising types would say that Putnam’s analysis of what they do is ‘psychologistic’ or even one large ad hominem.) Putnam says:

I think part of the appeal of mathematical logic is that the formulas look mysterious – you write backward Es!” (233)

Who’s to say that to some the use of mathematical logic will automatically take us nearer to the truth about such matters? (The truth about meaning for a start.) Just because mathematical logic is arcane and an extra-special specialism - indeed just because it's also ‘mysterious’ (with its ‘backward Es’), does that automatically mean that it will be of value – of any value – when it comes to discovering things about the non-formal matters of philosophy? Mathematical logic may help us in certain ways. (It will help us formalise our problems for a start.) Though wouldn’t we reach a point at which it would be of no help whatsoever?

Is mathematical logic seen by some philosophers as the strange and arcane symbols of ancient myths and religions were seen by people like Newton and others – as keys to an otherwise impenetrable - though deeper - world?
Perhaps we should finish here with a mention of an appositely entitled book by Bertrand Russell – Mysticism and Logic!

No comments:

Post a Comment