blog.kennypearce.net

Speaking Loosely

Philosophers often use such phrases as 'strictly speaking' or 'in metaphysical rigor' before saying things that might sound outrageous. For instance, many philosophers have denied the existence of entities which everyone 'knows' to exist, such as chairs, or minds, or numbers. The philosopher will almost always prefix such a denial with this sort of modifier.

The opposite of speaking strictly is speaking loosely. In early modern philosophy, the 'strict and philosophical' mode of speech was often contrasted with the 'loose and popular' mode. Other philosophers might use the modifier 'strictly and literally.'

What is the point of making these qualifications?

One of the main points, it would seem, is that the philosopher who denies the existence of chairs is going to continue speaking 'loosely' when not doing philosophy, and so will sometimes claim to see a chair, or to sit in one. Is this practice defensible? Does it amount to a 'performative contradiction' - that is, does this philosopher contradict with her actions what she says in her philosophy?

I've never seen any literature devoted specifically to this question; as far as I know, it only comes up in the context of defending some form of eliminativism or other. Here, I want to propose a definition of 'loose' speech: to speak 'loosely' is to (attempt to) express a truth by uttering a falsehood. (I insert the 'attempt to' modifier because we should still say someone is speaking loosely when he says something he believes to be false in order to communicate something he believes to be true, even if he's wrong.)

First, we should ask whether this definition is coherent. If the meaning of a sentence is just the proposition it expresses, and every meaningful sentence expresses exactly one proposition, and no proposition is both true and false, then it would seem to be impossible to speak loosely, under my definition. However, I do think a principled distinction can be drawn between the meaning of a sentence and what a speaker means by it. The reason this distinction seems odd is that the two ordinarily coincide; the cases where they do not coincide are mostly cases of loose speech.

The meaning of a sentence, in this view, is the meaning that would be constructed by the methods of formal linguistics. One can tag the parts of speech, identify the antecedents of any pronouns, parse for syntactic structure, figure out the lexicography, etc., picking the sentence apart bit by bit according to formal rules, and arrive at a meaning. However, this meaning does not always line up with what the speaker means by the sentence.

Perhaps the least problematic cases of loose speech are metaphor and sarcasm. When the Bible says that God led the Israelites out of Egypt "with a mighty hand and an outstretched arm," it doesn't imply that God has a hand or an arm. This is because, in metaphor, a falsehood is used to express a truth. Similarly in sarcasm, one says just the opposite of what one believes to be the case, but this is not done in order to deceive. When sarcasm comes off correctly, one communicates a truth by uttering a falsehood.

The cases the philosophers are concerned with are, however, more complicated than these. Eliminativists about macrophysical objects, like Peter van Inwagen and Trenton Merricks, are not speaking metaphorically or sarcastically when they ask whether there are enough chairs for everyone. Rather, they express the content they wish to convey in terms of the (according to them) false presuppositions of their hearers.

Is this a form of dishonesty? I don't think so. It seems to me that honesty permits this in cases where the false presuppositions are irrelevant to the matter at hand. In these cases the false presuppositions are not part of the content of one's speech, but merely the mode of expression.

My claim, then, is that loose speech involves the expression of a truth by uttering a falsehood. I claim, further, that it is possible to do this, and that there are at least three cases in which we actually do it: metaphor, sarcasm, and expression in terms of false presuppositions.