Saturday, 1 November 2008

Putnam on magical theories of reference

Last week, I was invited by my colleague Peter Lucas to reply to a paper he was giving defending phenomenology from the charge, made against it by Hilary Putnam in Reason, Truth and History, that it subscribes to a merely magical theory of intentionality.

I last looked at Reason, Truth and History twenty years ago with quite different eyes. Then, I took its attack on reference to be an attack on a correspondence theory of truth and an inflated metaphsyical realism (as Putnam seems to want us to take it). But now it seems to be more like a defence of a Davidsonian prioritising of truth over reference as an approach to semantics. If so, whilst the Skolem-Lowenheim thesis, to which he appeals, seems to deliver a familiar result, it seems to presuppose a too intellectual picture of how reference might get off the ground to promise a helpful account of language.

Deriving reference from truth in the absence of a contextualising account of a language game (for want of a better phrase) seems a no hoper. Davidson tempers his truth-based approach to formal semantics with the more general philosophy of language of the field linguist employing the principle of charity, for example. Putnam merely darkly talks of reference within a conceptual scheme:

In an internalist view also, signs do not intrinsically correspond to objects… But a sign that is actually employed in a particular way by a particular community of users can correspond to particular objects within the conceptual scheme of those users. ‘Objects’ do not exist independently of conceptual schemes. We cut the world into objects when we introduce one or another scheme of description. Since the objects and the signs are alike internal to the scheme of description, it is possible to say what matches what. [Putnam 1981: 52]

This might be intended to make a merely negative point. One cannot place words in a relation with extra conceptual reality. Or, taking account of Travis’ writing in this area – he argues that objects such as pieces of meat are themselves extra conceptual – things impact on thought only through our conceptual abilities. They never brutely impact. But Putnam’s language implies that there is a substantial insight into the nature of the world summarised in the label ‘internal realism’. We can only refer within our scheme as though extra-scheme reference is denied to us. We’d aim at that more penetrating gaze but are blocked by the limits of our grasp.

What was equally disappointing was the way in which Putnam recruited Wittgenstein to his initial argument against a magical theory of reference. The set-up is something like this. According to Putnam, ‘mental presentations’ (a technical term), had they existed, would have been essentially representational. They would have had essential intentionality. Mental images and concepts are not mental presentations, so defined. They are like physical images and signs whose intentionality is contingent.

Concepts are not mental presentations that intrinsically refer to external objects for the very decisive reason that they are not mental presentations at all. Concepts are signs used in a certain way… And signs do not themselves intrinsically refer. [Putnam 1981: 18]
The doctrine that there are mental presentations which necessarily refer to external things is not only bad natural science; it is also bad phenomenology and conceptual confusion. [Putnam 1981: 21]
[M]ental representations no more have a necessary connection with what they represent than physical representations do. The contrary supposition is a survival of magical thinking. [Putnam 1981: 3]

He then deploys an argument inspired by Wittgenstein which illustrates the contingent intentionality of pictues and hence, apparently, mental images.

Suppose there is a planet somewhere on which human beings have evolved… Suppose these humans, although otherwise like us, have never seen trees. Suppose they have never imagined trees... Suppose one day a picture of a tree is accidentally dropped on their planet by a spaceship ... Imagine them puzzling over the picture. What in the world is this? All sorts of speculations occur to them: a building, a canopy, even an animal of some kind. But suppose they never come close to the truth.
For us the picture is a representation of a tree. For these humans the picture only represents a strange object, nature and function unknown. Suppose one of them has a mental image which is exactly like one of my mental images of a tree as a result of having seen the picture. His mental image is not a representation of a tree. It is only a representation of the strange object (whatever it is) that the mysterious picture represents…
We can even imagine that the spaceship which dropped the ‘picture’ came from a planet which knew nothing of trees. Then the humans would still have mental images qualitatively identical with my image of a tree, but they would not be images which represented a tree any more than anything else.
[Putnam 1981: 3-4]

Putnam thus drives a wedge between pictures (of a tree) and trees. Pictures are not essentially representational. They need not be individuated by what they represent. And they have other non-intentional intrinsic (eg spatial) properties.

But to count against a magical theory of reference or intentionality (of mental images), the argument needs to sever the connection between the aliens’ mental images and the picture. It does not. Why generalise from the contingent intentionality of pictures to that of mental images?

Note also the point about concepts. Putnam says ‘Concepts are signs used in a certain way… And signs do not themselves intrinsically refer.’ But what of signs used in a certain way? I agree that one might pick out the way in a manner that has no essential connection to the meaning of the sign. But if so, that way will not explain concepts in the way that Putnam has just claimed is possible (when he says that concepts are just signs used in a certain way). Picked out in the way that is needed for that identity claim, then the way signs are used seems, contra Putnam, to have an essential connection to the concept.

I think that the overall problem is this. Wittgenstein criticises the appeal to images (which resemble what they are about) and other free-standing mental items (such as inner signs, like signposts, that just stand there) as an explanation of intentionality. Thus the contingent intentionality of pictures counts against the use of images in explanations of intentionality. Putnam, by contrast, argues that what we would pre-philosophically call ‘mental images’ (rather than explanations of mental images) are not mental presentations as he defines them and have merely contingent intentionality. And this move seems simply wrong.

What seems a little ironic is that Wittgenstein himself sometimes seems to say quite spooky things about intentionality:

“You said, ‘It’ll stop soon’. Were you thinking of the noise or of your pain?” If he answers “I was thinking of the piano-tuning - is he observing that the connexion existed, or is he making it by means of these words? - Can’t I say both? If what he said was true, didn’t the connexion exist - and is he not for all that making one which did not exist? [§682]

I draw a head. You ask “Whom is that supposed to represent?” - I: “It’s supposed to be N.” - You: “But it doesn’t look like him; if anything, it’s rather like M.” - When I said it represented N. - was I establishing a connexion or reporting one? And what connexion did exist? [§683]

I do not mean that Wittgenstein subscribes to a magical theory of intentionality. He subscribes to no theory in this area. But his rejection of subvenient mechanisms of thought verges on the spooky.