Strange loops 'R' Us

I got a heap of books for my birthday, which I want to get my thinking straight on. So I'll be writing about them here. The first to arrive was a book by Douglas Hofstadter (of 'Gödel, Escher, Bach' fame) called 'I am a strange loop'. In this book, Hofstadter revisits his previous musings on selves, souls and strange loops.

The jumping-off point is a paradox. Careful examination of the brains of human beings has shown no violations of the laws of physics, and all the physical evidence we have points to the idea that the matter of which we are formed follows the same rules as all the other matter we come across. So it looks like we could, in principle, explain all the behaviour of any given person in standard physical terms. So, for example, the answer to the question 'why is this person angry' would be 'well, you see, these various atoms are banging about in such-and-such a way'.

But an odd thing happens; our explanations very often go the other way. We ask 'why is it that this collection of atoms is moving about in such-and-such a way?', and reply 'well, because this person is angry/sad/happy/thankful'. The explanation here looks like it is going the wrong way; the lower levels (in a reductionist sense) are being explained in terms of the higher.

Hofstadter summarises this paradoxical aspect of our selves by saying that we are 'strange loops'. The loopiness is clear; our thoughts can be explained in terms of physical events, and physical events can be explained in terms of thoughts. It is this second kind of explanation which makes the loop strange; it goes the wrong way - events at a lower level are explained in terms of those at a higher level.

As an analogy, he points to the proof of Gödel's theorem. In this proof, a statement G of number theory is concocted which cannot be proved or disproved. The (lower level) fact that G cannot be proved is explained in terms of higher-level properties relating to a subtle scheme for coding statements of number theory as numbers. Here, too, the explanation seems to go the wrong way.

Hofstadter goes no further than this in defining just what a strange loop is, and he certainly does not give a precise definition. Instead, the idea serves as a powerful analogy bringing out an aspect of the strangeness we often overlook in our all-too-familiar selves. The puzzle has been sitting under our noses for a long time, and there have been many proposed resolutions.

For example, it has been proposed that it is impossible in principle to reduce the soul to the usual behaviour of matter. Instead, our souls are made up of a different, more ethereal stuff which interacts with the matter in our brains in a deeply mysterious way. This removes the strangeness and sees our consciousness as involved in a much more mundane causal loop.

The trouble with this proposal is that if it is right we ought to see some violation of the laws of physics happening inside people's brains. The fact that we see no such thing has led to the suggestion that the link may not be causal; that the matter may in fact be carrying on in the usual way of matter and yet still have a mysterious 'self', made of spirit-stuff, associated to it. The trouble with this idea that the soul doesn't affect the body is brought out in a beautiful though experiment by Raymond Smullyan.

Another proposal is that there is no such thing as a soul; that such things as minds, ideas, consciousness etc. are pure illusions, and there is really nothing but physical stuff. Now, given how obvious it is that we are, for example, conscious beings, it is hard to argue for the idea that this is just an illusion. The main argument presented here is that, after careful investigation, we have not found the soul anywhere. But things need not be localised to particular collections of matter in order to exist. Think of the colour red, for example.

Another argument sometimes given here is as follows: Carefully examine the many things you associate with yourself. You find that your body is not yourself. Nor is any particular thought or feeling or sensation. In fact, close introspection reveals that in our multitudinous mental life there is not one simple thing we can point to and cry 'There it is: The self!'. The absurdity of this argument can be seen by applying it to chairs. Examine a chair closely, and study its parts. The legs are not the chair. The flat surface for sitting on is not a chair, and nor is the back. In fact, close examination shows that no simple part of the chair is the chair itself. Are we forced to conclude that there are no such things as chairs?

The option I think is correct works like this: We can (in principle) describe a person in terms of the physical stuff of which they are formed (though such descriptions are likely to be tedious). We could alternatively describe them in terms of their personality, and of what they are thinking and feeling. These are just two different descriptions of the same thing, with the second description much more useful to us because of its coarser grain. Anger does not cause the mental processes we associate with it, nor is it caused by them. They are two descriptions of the same process. The explanation of each in terms of the other is not causal; it is a matter of translation.

A helpful analogy is that of temperature and pressure. We can (in principle) describe a gas by specifying just where all the particles are and how fast they are moving (though such descriptions are likely to be tedious). We could alternatively describe it in terms of its temperature and pressure. Once more, these are just two different descriptions of the same thing, with the second description much more useful to us because of its coarser grain. The temperature does not cause the particles to move at the speeds they do; it is a redescription.

Scientifically, we have a pretty good idea of what temperature and pressure mean in terms of the behaviour of atoms. At one time, though, there was no understanding of what temperature and pressure meant in terms of more basic physics. We are in a similar position today with regard to the soul. This does not mean that there are no such things as souls, any more than their ignorance meant that, for those long-gone scientists, there was no such thing as pressure.

This gives a useful perspective on the second resolution. It may be that, one day, we will understand that some of the vocabulary we use to talk about ourselves (such as the word 'ourselves') is misleading, and reject it, just as the theory of phlogiston was rejected. At that time we will be able to say with confidence that the notions we reject are illusions. We are currently very far from that stage: We may never reach it, and almost certainly won't in my lifetime. For the forseeable future, we won't not know what is illusory and what isn't.

Hofstadter's next point is that our idea of the soul is a little blurred around the edges (just as all our ideas are). For example, each person has models in their own head of the other people they know well. For people in strong relationships these models can become intertwined in feedback loops with the self-models of the other people (Hofstadter repeatedly confuses these self-models with the selves themselves). Accordingly, it isn't possible to completely localise a person within their own head.

The punchline is that, when a person's body dies, the person lives on to a small extent in the heads of those who loved that person. The trouble is that the extent to which the person lives on is negligible by comparison to what is lost. It is analogous to the extent to which the person lives on via the atoms from their body which have been incorporated into other living organisms. When a person dies, the loss to us is irreparable and complete. This is why we grieve so strongly, and why we reach out for comfort in the smallest scrap of hope that the person still, in some sense, lives on.