Wednesday, October 22, 2008

Thought Provoking, the 'Harder Problem'

Ideally, you would read this article (also HTML) and simply relay to me* all the questions it raises for you, and I would answer most of them, and then you would argue with me about how my responses don't make any sense. This method would, I think, be much more useful to you, and also allow the debate to stay on track, since there would be this shining goal of a question you wanted answered.

*(Even more ideally, I would be but one choice of many philosophers you might choose.)

I didn't feel like commenting on the earlier parts of the paper, so I didn't. However, if you are interested, I can do.

Instead, the below.

"If we reject property dualism, then the issue of whether Commander Data is conscious depends on extrapolating a concept of consciousness grounded in our physical constitution"
Don't reject property dualism. In either case, begin with a physical account of consciousness. However, if you do not reject dualism, you may find that you have accounted for all observed properties of consciousness without actually reaching the second property, in which case you've finished and can abandon dualism. Conversely, in a dualist world, the physicalist repeatedly rejects evidence for dualism, seeking alternate solutions. As such, dualism is the right mistake to make, not simply the intuitive position.

First let me summarize the below. Subjective intensities relate to objective intensities by simple power laws, a fact which applies to all tested subjective experiences.

"An example will be useful to clarify this point. All human sensory systems obey a power function, an exponential function relating stimulus intensity to subjective intensity as judged by subjects’ reports. That is, subjective intensity = stimulus intensity raised to a certain exponent, a different exponent for different modalities. For example, perceived brightness is proportional to energy output in the visible spectrum raised to a certain exponent. This applies even to outrĂ© parameters of subjective judgments such as how full the mouth feels as a function of volume of wadges of paper stuck in the mouth or labor pains as a function of size of contractions. Should we see the question of whether Commander Data’s sensations follow the power law as a litmus test for whether Commander Data has conscious experiences? No doubt the power law taps some neural feature. Is that neural feature essential or accidental to the nature of consciousness?"

Accidental. Take the mind node not as true but as a metaphor. Given a mind node, the input can be varied along the power axis arbitrarily. In a real brain, different stimuli have different powers,* and also (probably) that different individual have different power distributions. Thus, the power is arbitrary, and thus accidental.

*(Otherwise it would not be a power law, but a simple power, and Block would have listed the approximate power.)

"Chalmers (op. cit., p. xvii) recommends this orientation, saying “I find that discussions framed in terms of identity generally throw more confusion than light onto the key issues, and often allow the central difficulties to be evaded. By contrast, supervenience seems to provide an ideal framework within which key issues can be addressed.”"

The properties of bulk water supervene on microphysical properties of water molecules, which supervene on the properties of the arrangement of subatomic particles in water, and so on... But, as stated earlier in his piece, water=H2O, as in, not water supervenes on H2O. All superveniences can be re-stated as identities. The only convenience in supervenience is that unlike identities, the properties do not quite have to match up 1:1, or put another way, the definitions can be a little sloppy around the edges, as long as any change in the properties of one definition entails a change in the other. For instance, once the power law is established, the subjective sensation supervenes on the objective signals being sent. This is slightly easier than having to say that a power of the objective signal is identical to the subjective signal. But, in my opinion, you should just state the identity or at least the causality, and be not-sloppy.

In short I'm saying I don't understand how this concept is supposed to be useful. (My understanding of supervenience supervenes on my conception of how useful it is.)

"But the Harder Problem depends on the puzzling nature of multiple physical constitution of consciousness"
When I got to this part, I finally figured out what "The Harder Problem" is supposed to be. Incidentally I've already solved it. The fact is that consciousness arises due to an abstract mathematical property, and thus multiple physical instantiations can easily give rise to the same or very similar properties. Further, since consciousness primarily operates non-physically, the physical differences are of only ancillary relevance.

But yes, in a physicalist or materialist worldview, it would be very puzzling.

On second though, no it wouldn't. It was puzzling before I thought about it. Now I know that the only respectable way to formalize consciousness monally is through panpsychicism, and so we could naturally assume that not only Data is conscious, but that my soup bowl is conscious.

"Even the epistemic difficulty may be temporary, unlike the epistemic difficulty of the concept of the gold mountain that no one will ever have evidence of."
I actually don't know what Block's trying to say here, so I'm going to reiterate my ideas on existence with reference to the gold mountain.

Think about what 'to have evidence of' means. It means that this entity, this gold mountain, acted upon you. Further, because of Newton's Third, you acted upon it. And thus, this assumption that nobody will ever have evidence of it means that nobody will ever interact with it, which is precisely equivalent to saying it doesn't exist. The primary property of things which exist is that I can potentially interact with them. (Further, their specific properties define how I can interact with it - a property which I can't interact with is a property that doesn't exist.) To admit that this mountain exists in some Platonic sense is to admit to existence everything I can't interact with.

Intriguingly, it doesn't matter whether the gold mountain 'actually' doesn't exist, or if it 'exists' on some planet that nobody will ever see or visit. (Slightly broken because, for instance gravity, has potentially infinite range.)

"Perhaps we will come to understand the nature of human consciousness, and in so doing, develop an objective theory of consciousness that applies to all creatures, independently of physical constitution."
Oh really? Yeah, just perhaps.

"The Naturalist will want to reject Dualism, but it is cold comfort to be told that the only alternatives are doctrines that are epistemically inaccessible."
I only found alternatives of colder comfort than his alternatives. ("a choice of Superficialism, Disjunctivism and Dualism." The first is that we can only correlate consciousness on purely superficial processes, the second that the definition of consciousness is discontinuous; it has the form 'this is consciousness, and also this apparently unrelated thing.' I find neither coherent.)

"One way to restrict phenomenal realism is to adopt what Shoemaker (op.cit) calls the “Frege-Schlick” view, that comparisons of phenomenal character are only meaningful within the stages of a single person and not between individuals."
This would literally mean that we're all our own little incommensurate worlds.

"That is, Commander Data either has no consciousness or there is no matter of fact about his consciousness."
If there's no matter of fact about Data, there would be no matter of fact about my consciousness either. (Relativity; no special reference frames.) Since there is such a matter about me, there exists such about Data. (No-ish: he doens't actually have properties, since he doesn't actually exist.)

If you decide to read the rest, you can ignore all the technical isms; ("I apologize for all the “isms” [deflationism, phenomenal realism and one more to come]") Block does eventually stop fishing for marks and gets his point across without using them. Still, there's an incredible number of wasted words; it's almost like he was educated at some kind of government institution. Block clearly demonstrates either deliberate padding or else padding by habit.

No comments: