= 27d567bc2d48d9f40e9ccc20ea370b15)
eternal_neophyte comments on My recent thoughts on consciousness - Less Wrong Discussion
You are viewing a comment permalink. View the original post to see all comments and the full post content.
= 783df68a0f980790206b9ea87794c5b6)
Loading…= b715d02e85f7b3b4ae65ca3d3e450868)
Subscribe to RSS Feed
Powered by Reddit
= f037147d6e6c911a85753b9abdedda8d)
You are viewing a comment permalink. View the original post to see all comments and the full post content.
Comments (64)
My problem with materialist reductionism is that it entails that explanations should suffice to provide descriptions. The taste of honey refers to something entirely descriptive, both without the power of furnishing an explanation of anything about the honey, and incapable of being grasped by means of anything that does explain something about the honey.
You could model the world that provokes you into experiencing sensations without any access to the sensations themselves (you wouldn't know what you were modelling however) using nothing but a collection of flavourless tokens related by explanatory mechanisms.
Neither you nor I will ever know what a glucose molecule is in the way that we know what an orange tastes like. If you found out tomorrow that molecular theory has been a grand, ingenious, astounding and improbable swindle, nothing about your beliefs concerning the taste of oranges will change in the least. I really cannot see how explanatory statements can bear descriptive burdens, which is ultimately what the problem of "qualia" is driving at.
There are even further problems I see, for example if your language is restricted to explanatory statements then your ability to communicate is effectively restricted to statements about the states of some machine (for the sake of completeness of argument, take this to be a turing-complete computer) and changes in this state. This leaves no room for the possibility of a description, any statement concerning a "qualia" could not augment the information we have about the state of the machine or the rules by which it changes. It follows from this that all "senses" available to the machine could at best only be different schemes of drawing up a more compact declaration about machine states and programs, there could be no sense of "sight" as distinct from the sense of "sound" or "emotion". For a machine, the sensation of "sound" might as well be identical to the sensation of sight, but always accompanied by a unique, peculiar shade of blue. If you deny a human being the ability to separate out information according to the sense by which it arrives, he's hardly capable of communicating anything whatsoever.
Explanations are useful because they organize descriptive information, rather than vice versa.
Pan-psychism resolves all philosophical hitches I've been able to come up with, and I would argue that it's not a "mysterious answer" but simply an assumption that our inner experience of consciousness is in fact a feature of the world as mass-energy is assumed to be, without the need for further explanation and not admitting of any such a possibility. Now I don't actually enjoy insisting that no explanation is possible of what is a very confusing topic, but in this case it's not the topic that needs to be explained but the process by which we confused ourselves over it.
Does it answer such questions as "how does consciousness work?" and "how can we make one (by other than the traditional method)?"?
I don't believe it does, it suggests no particular theory of how to build one machine that is more conscious than the other. But that is, I believe, precisely its strength. If consciousness is a basic feature of the universe, your theory of consciousness ought not to provide you with any such information, otherwise your theory would have some internal components and consciousness would not be the terminal level of analysis. Would you expect string theory to tell you how to make strings?
I would expect it at least to make testable predictions. The difficulty of doing so is an argument made by some physicists against its value, notably Lee Smolin. But at least there is something there, ideas with mathematical structure and parts, that one can study to attempt to get observable consequences from. I don't see even that much with panpsychism. We don't know what consciousness is, only that we experience it ourselves and recognise it from outward signs in people and to various extents in other animals. What can I do with the claim that rocks are conscious? Or trees, or bacteria?
Strictly speaking, what can you do with the claim that "apples are red"? Can you test apples for their inherent redness in a way that doesn't rely on your own petulent insistence to magically intuit the redness of apples? Ofcourse not, whatever chemicals you show to exist in the skin of an apple to prove its redness will rely on an association between that chemical and that colour which you will itself defend on the grounds of being able to see that the chemical produces redness in certain circumstances. Your ability to use the concept of redness to distinguish red apples from yellow ones similarly relies on your having direct, unmediated knowledge of redness. Conceptual analysis has to terminate somewhere, and it might as well (and arguably, ought to) terminate with whatever ideas we find necessary but impossible to investigate.
What can you do with the claim that people are conscious?
I can tell this to someone who is unfamiliar with them, and they will be able to predict what they will look like. (Of course, we are both glossing over the irrelevant detail that apples come in a variety of colours.) They will also be able to predict something of their objectively measurable reflectance properties.
But this is well into the land of Proves-Too-Much. What can I do with the claim that water is made of hydrogen and oxygen, that doesn't rely on "your own petulent insistence to magically intuit" (we're into the land of Straw Men also) its constitution? No, whatever (etc.etc., paralleling your own paragraph).
I can describe my sensation of my own presence, and say that this is what I am talking about. If the other person experiences something that my words seem to describe, then they will recognise what I am talking about.
Can you do anything similar with the claim that rocks are conscious? You can say, whatever conscious experience is, that you and I recognise, rocks have it as well. But that doesn't help me recognise it in a rock.
If rocks are conscious, so presumably are corpses. How does the consciousness of a corpse relate to the consciousness that animated it in life?
There's an ancient philosophical chestnut: "is my red your red"? So this is in fact not clear at all.
Same argument as for the chemicals applies. You won't be able to make any useful prediction that doesn't ultimately rely on your ability to simply perceive red.
Derive its chemical properties. There is some intuition involved in your knowledge of mathematics, but that's not the same as relying on an innate intuition as to its constitution. There was some point in time when the constitution of water was unknown, and anyone with enough knowledge of chemistry would have been able to make valuable predictions about the behaviour of water under various experiments once they learned how it was constructed, which did not rely on his ability to intuit the H20ness of water.
It doesn't help you recognise it in somebody with total bodily and facial paralysis either. Does it mean that it's nonsensical to ascribe consciousness to such persons?
By degree of complexity and organization, if nothing else.
I don't know what point you're making now. Of course I see my red and from my description of the apple he will know to expect his red. It makes no difference to the present topic whether his red is the same as mine or not. It will make a difference if one of us is colourblind, but colourblindness is objectively measurable.
How do we measure the complexity and organization of the consciousness of a corpse at above zero?
That there are meaningful statements that cannot be empirically grounded - and the fact that you cannot communicate your own specific experience of redness to someone else shows that it's not empirically grounded: nobody would (or at least, I've never found anybody who seemed to) argue that the concepts of molecular theory or other statements about the material world are similarly ineffible. Insisting that any characterization of the nature of conscious experience in general is superfluous if it yields no predictive power (even if it resolves conceptual issues) is to insist that - categorically - statements that don't make any predictions about the material world are vacuous. The experience of colour as such serves as one particular counterexample.
My entire point is that the idea that you could measure consciousness under any circumstances whatsoever, of a rock, a tree, a person or a corpse, follows from an incorrect application of empirical epistemic standards to conceptual problems.
A blind man once said that although he had never experienced red, he imagined that it was something like the sound of a trumpet, which I think is pretty good. And fictionally:
— David Gemmell "Legend"
Medics routinely assess the state of consciousness of patients. People routinely, automatically assess the states of the people around them: whether they are asleep or awake, whether they are paying attention or daydreaming.
To me, our experience that we have experience, and our simultaneous inability to explain it, amount to our ignorance about the matter, not a proof that there is any conceptual error in seeking an explanation.
ETA: BTW, I'm not the one who's giving you a -1 on every post in this thread, and I wouldn't even if I was not one of the participants.
We need to look at the brain activity, whether seeing "red" activates the same parts of the brain for different people.
Take one person, show them a red screen, a green screen, a blue screen. Record the brain activity. Do the same thing with another person. Based on the first person's data, looking at the brain activity of the second person, could you tell what color do they see?
Thoughts and feelings are not immaterial, they can be detected, even if we still have a problem decoding them. Even if we don't know how exactly a given pattern of brain data creates the feeling of "red", these things could be simple enough so that we could compare patterns from different people, and see whether they are similar.
Such an experimental procedure depends on materialism; and materialism itself is the topic under scrutiny. Which is to say its results would under-determine the materialist/psychist dichotomy.
It can in principle, in the same way that atomic theory eventually told us how to transmute lead into gold. It's the right approach -- decompose into simple parts and understand their laws.
It's a long stretch from Epicurean atoms to nuclear physics, too long for me to regard the former as an explanation of the latter. Atomic theory wasn't of any use until Bernoulli used the idea to derive properties of gases, and Dalton to explain stoichiometric ratios. Pan-psychism consists of nothing more than hitching the word "consciousness" to the word "matter", and offers no direction for further investigation. Principles that suggest no practice are vanity.
Ok, but if you have a choice of theory while being an ancient Greek, the rightest you could have been was sticking with the atomic theory they had. Maybe you are an ancient Greek now.
Panpsychism offers a way forward in principle, by reverse-engineering self-report. Folks like Dennett aren't even addressing the problem.
What could they do, what did they do, with their atomic theory? Conceive of the world running without gods, and that's about it, which may be significant in the history of religion, but is no more than a footnote to the history of atomic theory.
What can we do with panpsychism?
In principle, try to construct a mapping between experience self-report and arrangements of "atoms of experience" corresponding to it.
Even if they ended up doing nothing, they were still better off sticking with the atomic theory, than with an alternative theory.
Rocks can't talk. Experience self-report only helps for those systems that are capable of reporting their experience.
Panpsychism might be an interesting idea to think about, but it is a question, not an answer. Does everything have a soul? (I use the shorter word for convenience.) If I split a rock in two, do I split a soul in two? If not, what happens when I separate the pieces? Or grind them into dust? Are the sounds of a blacksmith's work the screams of tortured metal in agony? Do the trees hear us when we talk to them? Do we murder souls when we cut them down? Does the Earth have a single soul, or are we talking about some sort of continuum of soul-stuff, parallel to the continuum of rock, that is particularly concentrated in brains? Is this soul-stuff a substance separate from matter, or a property of the arrangement of matter? An arrangement that doesn't have to be the sort we see (brains) in the definitive examples (us), but almost any arrangement at all will have a non-zero amount of soul-nature?
Plenty of fantasy story-seeds there, but I see nothing more.
Yup. Still useful (just very very hard).
Not super interested in arguments from incredulity.
Note that I am not aware of any competitor in the market place of ideas that offers any way forward at all.
That was an argument from the current absence of any way of answering these questions. It is not that the hypothesis is absurd, but that it is useless. As I said before, panpsychism merely utters the word "conscious" when pointing to everything.
You can do experiments on people to investigate how consciousness is affected by various interventions. Drugs, TMS, brain imaging, etc. There's lots of this.
Here's a rock. It's on my bookshelves. How does panpsychism suggest I investigate the soul that it claims it to have?
This is new to me, but googling "panpsychism reverse engineering", "panpsychism reverse-engineering self-report", "panpsychism self-report" doesn't bring anything that seems relevant. Has this been discussed anywhere?