blog.kennypearce.net

How Putnam Defeats Descartes' Demon

A little while back, I wrote a post describing Cartesian demon skepticism as a form of 'adversarial epistemology'. The idea is that Descartes' thought experiment can be conceived of as a game with two players: the meditator and the demon. The meditator selects a process for forming beliefs from perceptual experiences, and the demon knows what process the meditator has selected, and controls all of the perceptual experiences. If the meditator ends up with mostly true beliefs, she wins. Otherwise, the demon wins.

Now, I mentioned at the bottom of that post that this way of framing the problem is helpful for understanding some of the responses that have been proposed. In this post, I want to frame Hilary Putnam's response to external world skepticism in terms of this game.

Most readers of Descartes have thought it was pretty obvious that the only way for the meditator to win is not to play. Indeed, as I mentioned before, Descartes himself tried to escape the problem by establishing a priori that this is not what is going on ("God is not a deceiver"). What is the demon's winning strategy? Well, you might think that, if there is an external world and the demon knows the facts about it, he can win by causing a green perception whenever an object is red, causing a red perception whenever an object is yellow, and so forth through all the colors, and then using a similar strategy for all of the other sorts of properties objects can have. This way, the meditator will believe that green objects are red, etc., and the demon will win.

The implicit assumption in this strategy is that the content of our perceptions can vary independently of the facts about the world: that is, that the content of our perceptions could change radically without the world changing, or the world could change radically without the content of our perceptions changing. ('Could' and 'can' are intended to reference logical or metaphysical possibility - of course it is not physically possible for these things to happen.) Berkeley, whose solution I will certainly discuss another day, challenges the first half: he says that the demon can't change our perceptions without changing the world. Putnam, on the other hand, holds to externalism about representation, and therefore argues that if the qualitative nature (i.e. the 'what it's like') of our perceptions were kept the same and the world radically changed, the meaning or content of our perceptions, and of the thoughts formed from them, would necessarily be changed. If the demon followed the strategy above, then the perception which in English I call a perception of green would in fact represent red, and my word 'green' would actually mean red.

The reason for this is that, according to the externalist, what a representation (such as a word or image) represents is determined by its causal associations: because (in the example) I typically have green perceptions in the presence of red objects, it is red objects and not green objects which my green perceptions represent.

Now, take a more radical example: what if either there is no world at all, or the world is completely unrelated to my perceptions? That is, what if the demon invents my perceptual world from whole cloth, as it were? In this case, Putnam believes, my perceptions might actually refer to the demon's ideas or some other aspect of the demon's process for deciding what perceptions to cause. As long as some causal explanation can, in principle, be given for my perceptions, they are going to come out mostly true. The demon, therefore, has no winning strategy.