Saturday, April 9, 2011

On Galen Strawson on Nicholas Humphrey

Thoughts on Galen Strawson's review/criticism of Nicholas Humphrey's 'Soul Dust' (see also Midgely):

Once upon a time, not so long ago, no one thought that there was a mind-body problem. No one thought consciousness was a special mystery and they were right. The sense of difficulty arose only about 400 years ago
Going back all the way to Plato, the soul and the body are described there as entirely separate: The soul is active, immortal, while the body is made of various concrete elements, is passive, and will cease to exist. Arguments for the immortality of the soul focus on how the soul differs from the body.

I would argue that this is not, in fact, that different from the modern conception of the mind-body problem: Both the ancient Greeks and us today have two very different concepts. There is a mind/soul/consciousness, and there is a body/physical matter/material. The same notions about how the two differ appear both today and 2400 years ago.

Clearly there are great differences as well: We know far more about the physical world than Plato, and that has shaped our view of the physical and the mental. It has led to phenomenal consciousness being the main problem left to explain in the mind/body problem, hence the 'hard problem' of consciousness. Despite this difference, both the ancients and us have a distinction between mind and body. For the ancients this was a fact about the world; for us it is a 'problem' and has been made more specific. But the duality exists in both places. Had Plato learned that we can explain all physical action, including that of our own bodies, by means of impersonal mathematical laws, I believe he too would have arrived at the same 'hard problem' that we face today.

In Soul Dust, Humphrey seems to agree with Dennett, at least in general terms, for he begins by introducing a fictional protagonist, a consciousness-lacking alien scientist from Andromeda who arrives on Earth and finds that she needs to postulate consciousness in us to explain our behaviour. The trouble is that she's impossible, even as a fiction, if Humphrey means real consciousness. This is because she won't be able to have any conception of what consciousness is, let alone postulate it, if she's never experienced it, any more than someone who's never had visual experience can have any idea what colour experience is like (Humphrey says she'll need luck, but luck won't be enough).
In my opinion this really misses the hardest part of the mind-body problem, which I can try to state as follows: Our bodies are material, and governed by the laws of physics. We can explain their behavior without needing to say anything about consciousness at all, and with enough knowledge about physics and enough computation power, we can in theory explain all of our actions. There is therefore no need for a visiting alien to postulate consciousness when explaining human behavior. Consciousness may be a useful high-level concept, just like 'inflation' is a useful high-level concept in economics, but it is not a fundamental entity.

This so far argues against Humphrey's thought experiment. But the implications refute Strawson's rejoinder as well, for the entire point is that, to 'have a conception of consciousness' is not something that requires consciousness. Through entirely mechanistic methods - the abstract language of mathematics - we can explain why humans talk about having conscious experiences. In other words, we can explain why we talk about the subjective using entirely objective means. And in particular, having conscious experiences is not a precondition for talking about them in a justified way. This seems to leave no room for any actual subjective experience or qualia.

Strawson claims that "we know the intrinsic nature of consciousness and consciousness is a form of matter." But we can take the same computational processes happening in the human brain, and simulate them in an entirely different physical setting (say, in a computer), and get the same behavior of talking about having conscious experiences. It would seem then that consciousness is more a matter of computation, or perhaps we must give in and apply Turing's test blindly. In either case, equating consciousness with matter seems a redefinition. Perhaps we can redefine one step further, and say that consciousness is matter, and matter is computation? But none of this gets us anywhere at all, and we are still left with a fundamental problem to solve, which has been open not for 400 years as Strawson claims, but for at least 2,400.