domingo, outubro 25, 1992

Icing on the Cake: "Consciousness Explained" by Daniel C. Dennett





(Original Review, 1992-10-25)


I feel uncoupled.

Who knows for certain: their inner experience of sights, smells, emotions, and the rest?

And this is why I often find the discussion frustrating; from my reading of his work, Dennett has never denied the experience of being conscious. What he is saying is that if you create a zombie doppelganger that resembles you in every way then the "zombie" will by necessity be of such complexity that it gives rise to consciousness. And it will do so from normal, physics-obeying, materialistic processes. It is in this way that we are all the zombies of the thought experiment - not that we are all empty machines that experience nothing.

A nice analogy I read somewhere on this is that it’s like the opposite of a personal computer. A processor can do one task at a time, but it works so fast that to the user it looks like it is doing many different things simultaneously. A human brain on the other hand has countless different threads of activity happening concurrently; consciousness is the culmination of (some of) these to give the appearance of a single unified agent.

Although Dennett's "Consciousness Explained" almost certainly doesn't give a complete picture on the subject, it does provide testable hypotheses that are grounded in scientific reality. Mind/brain dualism has always struck me as an intellectual dead-end, giving up and attributing something we don't fully understand to magic. Where does the conversation go from there? On past form as we increase our knowledge that way of thinking doesn't tend to last.

Which is ok - I accept that conscious experience is the result of normal physical processes, no Cartesian I - but while it removes the dualism problem, it gets us no further in understanding how ordinary bits of matter can come to experience anything at all, and nothing in the neuroscientific literature that I've come across gets any closer than saying "it just happens when things get complex". Well, duh!

Schopenhauer who bangs on for pages on will, which is consciousness with direction, which is an essential part of what consciousness is. And behind, or beyond will is the thing in itself which is unknowable. And this is in all things. As Dylan Thomas put it "The force that through the green fuse drives the flowers drives my green age....." Consciousness cannot know itself reflexively because it is that which "knows" we can only experience will directly from the inside.

I do have a theory of my own on "why it evolved":

Consciousness is only the icing on the cake, a thin veneer that pretends to ride the zombie but may be able to sometime influence its actions. That does not mean the zombie is “atom-for-atom the same as humans”, it only means that today’s techniques are not good enough to find the small difference.

Consciousness developed because it is advantageous.

Animals manipulate other inanimate and living things all the time. We know that animals, different to rocks, have an active self, subjectivity. We must make predictions about a rabbit’s future actions in order to catch it. Same goes for other humans. But they also have a view of our subjectivity, our self, make predictions about our future actions. If we want to be able to manipulate their perception of us, we have to have a view, a sense of that self, first. Hence consciousness. This squares with a lot of research – more social animals (with the usual genetic differences of sexual reproduction – i.e. not ants etc. where genetic interests of individuals are much more aligned) tend to have a more developed brain and more observable traits of consciousness. It also squares with the notion of all of us being "honest liars" because we all (apart from the clinically depressed) have a more flattering view of ourselves than is merited. This helps a lot to induce a more favourable view of ourselves in others.

Good theories, don't you think? I would query one more thing though:

"Consciousness developed because it is advantageous": Is it really so, when we come to think about it ? Technically speaking, no. In evolutionary terms 'success' is quantified in terms of longevity, reproduction and survival. The most successful organisms on the planet are single-cell organisms, amoeba, bacteria, etc., etc. Do these have consciousness? We don't know, but we doubt it. Trees, technically speaking, are more 'successful' than living organisms with consciousness. Are they conscious? We don't know. Consciousness is not necessarily advantageous to survival. Besides which, we still don't know how unconscious matter became conscious in the first place, so long before we theorise on the evolutionary benefits of consciousness, we have the problem of answering how consciousness arose in the first place.

Well, this constant reality and free-will simulation also incurs massive costs - not even the great apes are quite up to it because they can’t ingest enough calories. So it’s a cost/benefit thing. The cost should be low enough, i.e. most mental processes pressed into the service of self-awareness should already be in place. And the benefits should be great enough which is mostly the case were constant management of our relations with individuals from our own species (and again, sufficiently genetically different from us) are essential for surviving and producing offspring.

None of this applies to microbes, trees, or iPhones.  Which also means that not only is this “consciousness everywhere” is complete bollocks, but also that consciousness doesn’t quite “arise automatically”. Only if survival and reproduction of AI depends on tricking the minds of other conscious beings than the mere preconditions for it to arise are in place.

Tricky stuff!