Friday, August 17, 2007

Blake Ostler

I have recently read bits and pieces of Blake Ostler's Mormon Theology and have found myself agreeing with him. In a review by FARMS, his overall position is laid out, and it very closely matches the ideas that I had come up with, namely:

1. Libertarian, incompatiblist freewill

This means that our choices can change the future. There is no fixed future that can be known beforehand, only a set of possibilities.

2. Conditional prophecies

Because of 1, prophecies in general are conditional on our free choices.

3. God exists within time

God is still a person: he can think thoughts, take actions, feel emotions. All of these are impossible if God doesn't experience time.

4. Choice is an act of creative perception.

I'm inferring a bit here, because I can't find a direct quote. But my idea, which I think he would agree with, is: Creativity, perception, and will are all essentially the same thing. Whenever our nerves bring information to our consciousness, we have a choice about how to interpret it as a perception. This includes our own state of thought and emotion. This interpretation isn't caused by the signal; instead, it's literally new, created information. It is that perception that causes our future actions. For example: when you have two conflicting desires, you can percieve one as unimportant and the other as important. The one you percieve as important is the one that will influence your actions. Creative artwork is an act of choosing to see things in a new way.
An example is the "Necker cube" illusion (the illustration to the right.) You can choose to see it as open to the upper left or open to the lower right. There is a measurable change in the state of the neurons in your brain when you make this choice. All choices, I propose, are this sort of choices.


mike said...

Last week I read Martin Gardner's review of Douglas Hofstadter's new book, I Am a Strange Loop. Hofstadter, of course, argues that qualia and agency are an epiphenomenon, the result of a bunch of self-referential systems. Gardner says that Hofstadter describes self-reference very well, but fails to even make a beginning into understanding how consciousness arises:

"We mysterians are persuaded that no computer of the sort we know how to build--that is, one made with wires and switches--will ever cross a threshold to become aware of what it is doing. No chess program, however advanced, will know it is playing chess anymore than a washing machine knows it is washing clothes. Today's most powerful computers differ from an abacus only in their power to obey more complicated algorithms, to twiddle ones and zeroes at incredible speeds... There may be advanced life forms in Andromeda who know the answers. I sure don't. Nor do Hofstadter and Dennet. And neither do you."

Mysterians are those who hold that there is something more than computation involved in qualia and agency.

I also read Greg Egan's short story, "Oracle." It's an alternate universe story; the main characters are clearly Alan Turing and C. S. Lewis, though they have different names: Stoney and Hamilton, respectively. Near the end, Stoney and Hamilton have a debate about whether machines could possibly think. In the story, Hamilton offers Penrose's argument that because a human can understand Goedel's incompleteness theorem, that their thoughts must be uncomputable. Stoney makes the plausible argument that humans' thought processes are far more complicated than a simple finite axiomatic system, so a computer that follows the same process would have similar ability to reason about Goedel's statement. What does not follow, for mysterians (which presumably would include Hamilton), is that the computer would experience anything while reasoning in this way.

If, as you propose, that qualia and agency are the same mystery, then Conway's Free Will Theorem becomes relevant. It says that if an experimenter has the property that his behavior (specifically, his choice about which way to measure a particle) is not entirely determined by his past light cone, then every spin-1 particle in the universe has the same property. Conway believes in free will, and thinks that it arises somehow from some interaction of the individual freedoms of the particles. (Actually, he thinks a system is freer if it's not impeded by judgment, and suggested the term "free whim.")

In a similar way, perception of qualia might be a fundamental property of matter, and our perceptions might arise from some interaction of the individual perceptions of the particles.

Perhaps it's my modular programming background, but I've got to look at experiments like replacing the hippocampus with a chip and assume that consciousness is independent of the implementation. For the scenario under consideration, that means that it's not the electron that perceives, but perhaps the quantum state that perceives. This implies that there's some kind of algebraic structure on Hilbert spaces that tells how two interacting quantum systems' perceptions and freedoms combine.

On closer inspection, it can't be the quantum state that's perceiving, since we're free to choose the basis, and every state is in the same equivalence class.

The next possibility is to consider perception as a process in time. This would make the unitary operations into the things that perceive. We have the freedom to change the basis, but two transformations T_1, T_2 that take the initial state to orthogonal states in one basis will do the same thing in any other. So we can separate linear transformations into equivalence classes, and assume a map from these into the set of distinct perceptions.

D said...

That review is wonderful. I hadn't realized that Gardner would take such a position.
Clearly a computer can have a separate representation of its internal state. What it can't have is a phenomenal experience of that representation, or a free will response to the phenomenal representation.
If qualia arise from the perceptions of individual atoms (which is the logical hypothesis for a scientist to make, by the way, absent any evidence of life after death or before birth) you have to ask, "what is it like to be a corpse?" because in that case, there should still remain awareness of some kind in your atoms after death. Without memory, senses, or cohesion, it would be a very limited sort of existence. Unless the atoms become part of another consciousness.
I personally think there are other ways out of the "dancing qualia" argument (replacing qualia experiencing modules with a chip) than accepting that consciousness is present in anything that takes the same inputs and simulates its effects.
It's nice to think that consciousness is somehow associated with the quantum (yet more conservation of mystery) but I'm not really qualified to have an opinion.
Would you like to read Strange Loop? You're welcome to borrow my copy.

D said...

Other ways out:
Suppose you are hypnotized. The hypnotist says, "your hand feels hot." Suppose that your hand does not actually feel hot, but you act as if it does because you have been hypnotized. (This is one thory of how hypnosis works: convincing people on a deep level to want to act like they are hypnotized.)
Couldn't the module act in the same way, by removing both qualia and free will at the same time and replacing your self with something robotic? In my system, there's no way to remove the qualia without at the same time removing the ability to respond to the lack of qualia freely.
I think this is the right way out of the problem. It also avoids the "dancing pixies" argument.

mike said...

You've proposed that qualia and free will happen in the same place, but there's some evidence that they don't. In Sacks' "Island of the Colorblind," he writes

“As a child I had visual migraines, where I would have not only the classical scintillations and alterations of the visual field, but alterations in the sense of color too, which might weaken or entirely disappear for a few minutes. This experience frightened me, but tantalized me too, and made me wonder what it would be like to live in a completely colorless world, not just for a few minutes, but permanently. It was not until many years later when I got an answer, at least a partial answer, in the form of a patient, Jonathan I., a painter who had suddenly become totally colorblind following a car accident (and perhaps a stroke). He had lost color vision not through any damage to his eyes, it seemed, but through damage to the parts of the brain which ‘construct’ the sensation of color. Indeed, he seemed to have lost the ability not only to see color, but to imagine it or remember it, even to dream of it. Nevertheless, like an amnesiac, he in some way remained conscious of having lost color, after a lifetime of chromatic vision, and complained of his world feeling impoverished… — his art, his food, even his wife looked ‘leaden’ to him.”

I don't think he's actually perceiving color but compelled to behave as though he isn't!

mike said...

> What is it like to be a corpse?

It's rather like a trillion things being molecules all at once. ("What is it like to be two people that don't know about each other?") The system that was the physical intantiation of "you" has broken down into several noninteracting systems.

D said...

On "what is it like to be a corpse"? I agree that that is what it would be like. My point is just that it's a terrible fate to contemplate.
On the example from Sacks: I would say that the part of him that experiences color has not been removed or damaged. It is experiencing the lack of color, and he despairs about that fact. The signals that generate color are being blocked somewhere upstream of that. But it's just a theory.