(cross-posted at Brains) Well it’s been a while since I have posted anything here…I have been so busy with Consciousness Online and my teaching load that I simply haven’t had time to do anything else! Today I am leaving for the apa in Vancouver where I will present The Reverse-Zombie Argument against Dualism which should be fun. I am also happy to announce that the Journal of Consciousness Studies is allowing me to edit a special issue of the journal devoted to Consciousness Online!!
Anyways, this last past weekend I presented the above titled work in progress at the Long Island Philosophical Society (held at LaGuardia this term). Below is the narrated powerpoint and a draft of the paper. The basic idea is that pain and painfulness are only contingently related (as evidenced by the dental fear phenomenon and Pain Asymbolia) and so we can explain awy the seeming contingency of mind/brain identies in exactly the same way as we do in cases like lighting/electrical discharge and heat/mean molecular kinetic energy. Any comments are welcome!
Richard Brown 
Even given that pain might completely be a brain state, shouldn’t it still be possible to find the brain states associated with dental fear phenomena and with pain asymbolia and thus to regain the reliable correspondence between the brain state and the unpleasant sensation of pain?
For example, a stoic child may silently suffer dental pain that could be relieved with a simple explanation. If the dentist had access to the brain state, then the dentist could know to give the explanation and know if it had been effective. By dismissing the “false” pain as a mistake, it seems you are saying that the explanation does nothing useful. But clearly it changes the brain state to one of less suffering.
In any case, the situation with regard to pain seems different from the situation with regard to lightning. Not all electrical discharges are called lightning, if I generate an indoor simulation of lightning we might disagree about whether it is proper lightning, but we wouldn’t disagree about the situation. The disagreement would just be about whether the label was appropriate.
In the case of pain, the disagreement is not about the label so much as it is about the situation. We cannot find the facts that we would agree (or disagree) constitute pain. Well, I can find these facts, for I know when I am in pain, but then you might tell me that I am mistaken about the pain, by pointing to different facts which relate to your definition of pain.
In the case of lightning we point to a particular phenomenon and argue about the label, but in the case of pain we point to different phenomena and each thinks the other is pointing at the wrong thing. And I can only point in a metaphorical sense, because I am not pointing to a physical phenomenon. There really isn’t anything to point to, I am simply attentive to an awareness.
Also, you seemingly start with the idea that a particular brain state is biconditionally associated with pain, then define that brain state as pain, and then take that definition to the extreme of telling people that they are being untruthful if they claim to feel pain but do not have the brain state, even though in that case the biconditional association no longer holds.
Even to say that there is pain if and only if there is a given brain state does not necessarily mean that the pain and the brain state are identical (although it does mean that research into brain states is likely to be useful in controlling pain).
If I had no choice but to look at my computer monitor, then the sensation of red would be reliably associated with a particular aspect of the state of the software running on my computer, such that in some area of the monitor only the red LEDs glowed. So the premise of the biconditional argument is there, and yet my awareness of red would not be identical with the state of that software, despite the utterly reliable association.
In that case, I might have a hard time convincing an artificial intelligence within my computer that my sensation of red was anything other than the software state that was associated with the glow of only the red LEDs in my monitor, because that artificial intelligence would have access to no evidence of any meaning of the word “red” other than the state of the software. My “real” perception of the red monitor is unavailable to the artificial intelligence.
Nevertheless, if the artificial intelligence can control the monitor, then that can have an impact on my quality of life, so any research done by the artificial intelligence into color, as defined in those software terms, is still potentially very useful.
The fact that a biconditional association exists as a matter of scientific investigation is really pretty strong evidence that the association is a contingent one. If it were a necessary association, then there would be no need to verify it empirically. A scientific experiment has to be able to test between alternatives, and thus can only test contingent propositions. Tests of necessary propositions are more in the domain of mathematics than of science.
So if I put two apples in a box, and add two more, and then count the apples in the box, I am not really testing the necessary proposition that two plus two equals four, but rather the contingent proposition that apples have continuity of existence within boxes. These kinds of experiments, at the quantum level, have surprising results.
Hi Shack Toms,
Even given that pain might completely be a brain state, shouldn’t it still be possible to find the brain states associated with dental fear phenomena and with pain asymbolia and thus to regain the reliable correspondence between the brain state and the unpleasant sensation of pain?
yes it should, but that wasn’t the point I was trying to make. I was arguing that pain and painfulness can come apart, which allows us to run a Kripkean response to Kripke’s argument.
Not all electrical discharges are called lightning
Right, but ‘electrical discharge’ is shorthand for ‘atmospheric electrical discharge’
because I am not pointing to a physical phenomenon
I disagree; you are pointing to a brain state whether you know it or not…
Also, you seemingly start with the idea that a particular brain state is biconditionally associated with pain, then define that brain state as pain
Right, that is because I am trying to defend the identity theory from an objection raised by Kripke. So I start with the identity theory and show how tit can meet the opjection.
and then take that definition to the extreme of telling people that they are being untruthful if they claim to feel pain but do not have the brain state, even though in that case the biconditional association no longer holds.
No, I don’t claim people are being untruthful! I claim that they are making a mistake. The biconditional still holds, it just doesn’t apply in this situation…
If it were a necessary association, then there would be no need to verify it empirically.
Well, this just ignores the whole point of teh argument I am making. Kripke argued, succesfully in my opinion, that some necessities are discovered a posteriori (water=H20 is the classic example). I am arguing that pain=brain state x is like water=H20. It is discovered empirically but is none the less necessary.
I am arguing that pain=brain state x is like water=H20. It is discovered empirically but is none the less necessary.
Even there, it is not necessary. Consider that water, as it turns out, is not H20. Rather it is always a mixture of H20, OH, and H30.
Here again, you might be able to define that only the H20 component was truly water, but we could still agree on the situation. The disagreement would only be about whether the word “water” was appropriate for that situation.
Once we have identified the brain states, and have mapped completely, say, the electrical mechanism between a person’s feeling pain and saying “Ouch!”, then the question will remain “OK, we can understand why the person says ‘Ouch’, but why does it hurt?”
That is, we can have a complete understanding of the electro/mechanical situation, and we will still be in the dark about why there is any non-physical awareness of it.
Even there, it is not necessary. Consider that water, as it turns out, is not H20. Rather it is always a mixture of H20, OH, and H30.
This isn’t really a problem since if we want to include hydronium and hydroxide we can say that water=H20 & H30 & OH; we could even add heavy water if we wanted. The point is that whatever water is identical to it is necessarily identical to it.
That is, we can have a complete understanding of the electro/mechanical situation, and we will still be in the dark about why there is any non-physical awareness of it.
Well, I have spent a lot of time arguing against that claim. If we really have a complete understanding of the electro/mechanical situation then I claim we will have a complete understanding of why it hurts, and further, without having to appeal to any non-physical attributes at all. What is your argument for it?
The point is that whatever water is identical to it is necessarily identical to it.
Yes. The thing that is contingent is the association of the label to the circumstance. Once the label has been defined as a particular circumstance, that circumstance is necessarily identical with the labeled circumstance.
So I agree that if we define pain as a brain state, we can then claim that it is necessarily a brain state. That is OK, except that it has little bearing on whether pain within its ordinary experience is also a brain state.
The question, as with water, is whether the definition matches the term as representing the circumstance of interest. With water, we probably can agree on the circumstance, with pain we seemingly do not.
What is your argument for it?
My awareness is aware of being aware, although my awareness is not an object within my awareness.
My most fundamental arguments for non-physical awareness are private ones. Modus ponens, but the antecedent is private.
That is, even if I accept that every thought that I have corresponds to a brain state, and that these brain states are consistent with the operation of physical laws, there will yet be something that is missing from your analysis—my non-physical awareness. Being in the possession of that awareness, I can conclude that the physical description is incomplete. However I cannot demonstrate this awareness to you (if I could, then it likely wouldn’t be non-physical). All I can do is to see for myself if it exists within the attributes that you enumerate as arising from the brain state.
I don’t see how it is possible to be mistaken about this. I am willing to accept everything that you accept about the brain state, except for your assertion that it is a complete description of awareness. You are demanding physical proof of a non-physical awareness. That is akin to a person in a dream demanding of me that I demonstrate the dreamer within the dream.
Interestingly, I fully expect that there is a brain state that corresponds to the idea that I have non-physical awareness that does not correspond to a brain state. In other words, my disbelief in your argument is probably a result consistent with the operation of physical laws. If we take your way of looking at things, my disagreement with you is a natural phenomenon.
Maybe it goes back to Chalmers. Suppose there really were some people who were microphysically identical to us and yet who lacked qualitative awareness. How could this qualitative awareness possibly be shown to them to exist?
One thing that I can do, however, is to point out that in the case of every virtual world, where we can see the implementing system, the implementing system can only exist within the virtual world as virtual representation of itself, never as itself.
The dreamer only exists within the dream as a dream being, never as himself. Mario cannot perceive the Nintendo system that implements his world, although the programmers might have had fun by including a representation of that system within the game.
So suppose we program a virtual world in which there are artificial intelligences who are debating whether there really is an implementing system. The system itself does not exist within the software model, so they cannot point to it. So suppose one of them comes up with the argument that you have presented. How should the other one justify his belief in the unseen implementing system?
In any case, following my argument even while dreaming, I can correctly deduce that my awareness does not arise from the state of my “dream brain”, the brain that I perceive within the dream. I don’t see anything about your argument that wouldn’t also hold if a dream version of you were making the argument within my dream. Essentially your argument would produce the conclusion that there was no dreamer, and there would be no way to convince you otherwise short of producing that dreamer within the dream.
Since my argument produces the correct conclusion in that case (and in every case where it can be tested), it is possibly valid. Since your argument produces, in that circumstance, an incorrect conclusion, it seems that it cannot possibly be a valid argument.
Maybe it goes back to Chalmers. Suppose there really were some people who were microphysically identical to us and yet who lacked qualitative awareness. How could this qualitative awareness possibly be shown to them to exist?
I don’t know, since I find that I cannot suppose what you are asking me to…
So suppose we program a virtual world in which there are artificial intelligences who are debating whether there really is an implementing system. The system itself does not exist within the software model, so they cannot point to it. So suppose one of them comes up with the argument that you have presented. How should the other one justify his belief in the unseen implementing system?
I don’t see why this matters since everything you describe can be captured in physical terms. The same is true of the dream situation. So your “argument” doesn’t produce the right results since there are no non-physical properties in any of those scenerios. In fact unless the dream brain is radically different than their actual dreaming brain, if the dreamer knew all of the physical facts he would be able to deduce the relevant facts about awareness…
Two more wuick thoughts:
1. there is something inconsistant in your appeal to virtual worlds, since you admit that some mental states will be brain states in the actual world. However in the virtual worlds NO mental states will be states of the (virtual) brain. So these kinds of worlds aren’t our analogous to our world and so you shouldn’t rely on them.
2. The only point that I want to insist on is that no a priori argumets refute physicalism. I am perfectly happy leaving it at the claim that it is an empirical a posteriori matter whether physicalism or dualism is true of our world. So that you (seem) to be able to imagine a situation where some physicalist is wrong is fine with me. But that doesn’t show that physicalism is false. It shows that for all we know right now it might be false. What we need is more empirical investigation. No amount of consideration about zombies or dreams will ever show that physicalism is false.
In fact unless the dream brain is radically different than their actual dreaming brain, if the dreamer knew all of the physical facts he would be able to deduce the relevant facts about awareness…
Exactly! And the only available physical facts within the dream are facts about the dreamed objects, together with the private experience of the awareness of the dream.
Within the dream, I deduce that the awareness arises outside of the dream, whereas you deduce that the awareness is entirely a phenomenon of the dream brain.
You are asserting that my awareness of non-physical awareness is not an available fact. But it is available to me.
No amount of consideration about zombies or dreams will ever show that physicalism is false.
Scientific laws model aspects of reality. It is thus in the nature of scientific laws that the operation of their models can be simulated. Therefore, anything that is true in science must also be true within a computer simulation of the scientific model.
This shows that any statement that is true in physicalism must necessarily be true within a virtual world that models physical laws. It also shows that, if physicalism is true, such a virtualization is possible.
Physicalism, it turns out, cannot be true in any virtual world, and thus it is actually inconsistent with physicalism. Physicalism refutes itself.
1. there is something inconsistant in your appeal to virtual worlds, since you admit that some mental states will be brain states in the actual world.
Not quite, I admitted that my thoughts would correspond to brain states that were consistent with physical laws. I didn’t go so far as to say that my non-physical awareness of those thoughts would be brain states. Awareness is not a thought.
I can think “I am aware!”, but that thought isn’t aware, rather I am aware of the thought.
The physical world is a mathematical abstraction, a set of equations that describe consistent relationships between objects within our awareness. I hold that the “actual world” is not identical with the mathematical abstraction we use to model it.
I think that physicalism turns out to be equivalent to the statement that the actual world is identical with the mathematical abstraction used to model it.
Your statements about the unreality of non-physical awareness amount to an application of that principle to awareness, but I think that it actually holds in general. We end up with the only possibility for physicalism being a tangled hierarchy of mathematical models, a strange loop in which the abstraction can manifest itself.
You are asserting that my awareness of non-physical awareness is not an available fact. But it is available to me.
No. I am asserting that what is available to you is awareness of physical awareness.
Therefore, anything that is true in science must also be true within a computer simulation of the scientific model.
This doesn’t follow. If I model a hurricane on my computer I do not have a hurricane in my computer. If we model the physical world it doesn’t follow that there is consciousness in the model. What physicalism requires if that there is a world which is micro-physically identical to our world then there is consciousness there.
I hold that the “actual world” is not identical with the mathematical abstraction we use to model it.
That sounds reasonable to me. But the physicalist is not committed to denying it. The physicalist is commited to the claim that all of the facts about the actual world can be deduced from the a completed physical theory. There is nothing about the identity of the theory with the world…
This doesn’t follow. If I model a hurricane on my computer I do not have a hurricane in my computer.
You have something that, within the model, behaves exactly as a hurricane behaves.
If we model the physical world it doesn’t follow that there is consciousness in the model.
You have something that, within the model, behaves exactly as consciousness behaves.
So would such a simulated consciousness be real consciousness?
Let’s examine your current assertion that it does not follow that it has real consciousness. So let’s suppose we create a mathematical model of neurons and create a number of simulated brains, made of different kinds of materials.
1. A simulation done with pencil and paper (a time-consuming proposition).
2. A simulation done with a digital computer.
3. A simulation done with an analog computer.
4. A simulation done with manufactured neurons (a collection of analog computers, say, each of which models a neuron).
5. A simulation done with cloned neurons.
6. A simulation done with natural neurons.
In simulating the neurons, we can present them with data that model the physical world, just as our sensory organs do for our physical brains.
So, my question is, where in that progression does consciousness appear? And on what basis do you claim that it is in step n, but not in step n-1?
If consciousness is physical, as you assert, then you should be able to point to it. So what do you point to in step n that is not present in step n-1, and which is likely to have any bearing on whether it is aware?
I think that the idea of consciousness being a brain state (as opposed to being a property of the matter of the neurons) is that consciousness is, in fact, a mathematical relationship.
Is that not correct? Or are you saying that there is some specific part of the brain that is the awareness? I thought it was the relationship.
So, my question is, where in that progression does consciousness appear? And on what basis do you claim that it is in step n, but not in step n-1?
Well, since I think that a conscious pain is identical to a certain brain state, and that this identity is necessary, then I would say that consciousness arises at step 5, though at that point it is no longer a simulation. You could tell the same story about lightning. Simultaed lightning is not lightning. In order ot have lightning you need to have atmospheric electrical discharge, not simulated atmospheric electrical discharge.
Or are you saying that there is some specific part of the brain that is the awareness?
No I am not saying there is a specific brain part which is consciousness, but rather that there is some specific brain state that is. When I have a conscious pain I am in a particular brain state which just is that conscious pain.
I would say that consciousness arises at step 5, though at that point it is no longer a simulation.
OK, so consider that these neurons have parts, and that these parts could be replaced one by one with prosthetics. What are the crucial parts of the crucial neurons that make brain states of neurons “real awareness”, but brain states that involve some semi-artificial neurons a mere simulation.
What is the fundamental difference between a brain state of “assisted neurons” and the brain state of purely natural neurons?
For example, suppose my mental function is supported by a medication that fills a gap, a defect in my neurons. Does this make me less human? Would you argue that because I might take such a medication that my awareness is a partial simulation and that I therefore shouldn’t be considered as feeling real pain?
The physicalist is commited to the claim that all of the facts about the actual world can be deduced from the a completed physical theory. There is nothing about the identity of the theory with the world…
In fact, there is that identity.
With respect to awareness, your assumption of identity between brain states and awareness arises because all of the facts about brain states match all the public facts about awareness. Your challenge to me is to present a fact about awareness that isn’t also a fact about a brain state.
In other words, it seems to me that you are simply denying supervenience and asserting that supervenience and equality turn out to mean the same thing.
If you think that physicalism asserts that reality is different from the set of facts about reality, then I wonder what fact about reality you have in mind which we can use to verify that.
I think that the answer to such a question is that, ultimately, we find something, as a referent of a symbol, that exists as neither a symbol nor as a relationship between symbols.
I think that such a “pure referent” is therefore necessarily private, awareness for example. “Matter” would also work, except that at least with awareness there is private knowledge that it actually exists. I don’t think there is any experience of matter apart from the objects of awareness, which objects are really just experiences of attributes, not of matter in itself.
OK, so consider that these neurons have parts, and that these parts could be replaced one by one with prosthetics. What are the crucial parts of the crucial neurons that make brain states of neurons “real awareness”, but brain states that involve some semi-artificial neurons a mere simulation.
I never said anything that would imply that semi-artificial neurons wouldn’t be good enough…though this is an empirical question…
In fact, there is that identity.
No, in fact, there isn’t.
In other words, it seems to me that you are simply denying supervenience and asserting that supervenience and equality turn out to mean the same thing.
How I am both denying supervenience and asserting that supervenience means the same as equality?
The point that I am trying to make here is simply that there is no good reason to think that menatl properties aren’t physical properties. That’s it. I am not asserting that physicalism is true. I am asserting that there is no good reason to think that physicalism is false. Zombies, Mary, Kripke’s argument, none of them give us any reason to seriously doubt that physicalism is false. The challenge I present to you is to show me some good reason for thinking that physicalism is false; plain and simple.
If you think that physicalism asserts that reality is different from the set of facts about reality,
I don’t think that and I didn’t say anything that implies that. There is the physical world, and then there is our description of that world in mathmatical/theoretical physical terms. The point is that there is no reason to think that a complete theory of our world will need to appeal to mysterious non-physical properties. The further point I was making was that an implementation of that description is not the same thing as the actual world. Simulating a bunch of gold atoms doesn’t give you any gold.
I think that such a “pure referent” is therefore necessarily private, awareness for example. “Matter” would also work, except that at least with awareness there is private knowledge that it actually exists. I don’t think there is any experience of matter apart from the objects of awareness, which objects are really just experiences of attributes, not of matter in itself.
That may be true, or at least plausible if you start from certain assumptions about how sensing works, but that is besides the point. If the “pure referent” as you call it turns out to be necessarily private that still doesn’t give us any reason to think that it isn’t physical. It would then be the case that I could make deductions from a completed physics to my own private conscious experiences. I would then be able to makes deductions from a completed phyics to your conscious experiences, though, of course, I couldn’t HAVE your experiences. But I might still be able to know what conscious states you are in. So you still haven’t gien any reason to think that physicalism is false.
Now, you are right that I am assuming that physicalism is, and should be, the default view and so should be accepted unless shown to be false. But that is another argument.
I never said anything that would imply that semi-artificial neurons wouldn’t be good enough…though this is an empirical question…
Well, you said that completely artificial neurons weren’t good enough, and that completely natural neurons were good enough. I am just trying to figure out what physical characteristics you think are necessary before a person should be considered human, and whether one person should, by virtue of physical characteristics, be considered more human than another.
If it is an empirical question, then there must be some physical test of whether awareness exists. So far, the only physical test you have mentioned is that awareness exists if and only if the brain state is a state between (at least partially) natural neurons. But you seem unwilling to give any reasons for deciding that completely prosthetic neurons are unfit for supporting the relevant kind of brain state.
What I am trying to get at is that there is no physical test for awareness, in the ordinary sense of that word, and you are trying to create such a test by making arbitrary definitions about the identity between qualia and brain states and thus are arbitrarily defining humanity.
How I am both denying supervenience and asserting that supervenience means the same as equality?
Sorry, I was imprecise. You are denying supervenience in that you are taking any evidence of supervenience as evidence of the stronger claim of equality, and then challenging those who believe only in the weaker claim to provide public evidence that your stronger claim is inappropriate.
I don’t think that and I didn’t say anything that implies that. There is the physical world, and then there is our description of that world in mathmatical/theoretical physical terms. The point is that there is no reason to think that a complete theory of our world will need to appeal to mysterious non-physical properties.
But that appeal is exactly what physicalism does. It is an appeal to what, in physics, are called hidden variables. Most physicalists are materialists, they think that the attributes arise because we are measuring local properties of a divisible, impersonal substance called matter.
The smaller problem with this view is that since we can only observe the attributes, and then only through measurement, so there is no direct evidence that matter exists. The larger problem with the view is that there are physical measurements, such as the Aspect experiments showing violations of Bell’s inequality, which are inconsistent with the idea that reality is a manifestation of local properties. “Local realism” simply doesn’t withstand the empirical tests.
The modern trend is to use the word “physicalism” instead of “materialism” because of the empirical problems associated with the idea that reality is made of matter. But the hope is to maintain the fundamental principle of ontological naturalism—that awareness is a mathematical relationship between impersonal attributes.
I grant that awareness is no less mysterious than is matter, however at least there is some evidence for the existence of awareness whereas there is evidence against the existence of matter, at least as it is usually thought of.
Furthermore, there is no operational difference between the physical properties of the world as we measure it and the measured physical properties of a simulated world within a simulation. You assert that lightning within the real world is different from a simulation of lightning within a virtual world, but the fact remains that every measurement within the real world would have an operational analog within the simulation and so there is no operational difference between the real electrical discharge as measured in the real world and the simulated electrical discharge as measured within the simulation.
If you have a theoretical basis that provides facts about the world, and if this theoretical description is complete, then every fact about the physical world should also be a fact within the theoretical framework.
But everything that exists within the theoretical framework is either a symbol or a relationship between symbols, there are no “pure referents” in the framework. Since pure referents do not exist within the theoretical framework, they cannot meaningfully exist within a physicalist interpretation.
I think that is really the crucial point of our disagreement.
I freely admit that it is surprising that awareness exists, but it would be far more surprising to find that it arises as a mathematical relationship between impersonal attributes. If that were the case, then it would seem that a book could be written in such a clever way that it would create the awareness of its reader.
If the “pure referent” as you call it turns out to be necessarily private that still doesn’t give us any reason to think that it isn’t physical.
What do you mean by “physical” in that statement? I prefer to say that there is no reason to think it doesn’t really exist.
The reason that it isn’t physical is that it isn’t a mathematical abstraction and thus cannot be completely expressed within a mathematical framework consisting of nothing but symbols and relationships between symbols. It cannot be modeled.
That is, there is a difference between things that exist and things that do not exist, but that difference is not entirely expressible as a fact within a theoretical model.
I seem to find myself in the position of Samuel Johnson. “I refute physicalism thus!” and I kick a rock. You may try as you will to reduce that entire experience to an abstract mathematical relationship (a physical model), but the thing that makes it real is my experience of it. And that awareness is the “mysterious “pure referent.
That is, we can have a complete physical theory, which includes all of the physical facts, but the thing that distinguishes the reality from the model is awareness.
If I dream of kicking the rock, it may be a dream rock and a dream toe, but the sensations of the rock, the toe, and the pain are all real. Maybe there is matter, and maybe there is not, but there is no need to bring this theoretical, ghostly “matter” into the picture at all. And in every theoretical model we can construct, we find that there really is no such thing as “matter”, there are only models of the ways that attributes interact.
Modern science, for the past 100 years or so, has advanced by discarding the notion that we are measuring an independently-existing reality and by replacing that with the notion that we are describing relationships between the measurements themselves.
It is very hard these days to conceive of a way that reality could possibly exist, independent of our measurement of it, that is consistent with the empirical evidence. As Einstein said, “These days, every Tom, Dick, and Harry thinks he knows what a photon is, but he is wrong.”
Now, you are right that I am assuming that physicalism is, and should be, the default view and so should be accepted unless shown to be false. But that is another argument.
It is strange to treat as the default view one that is demonstrably wrong in every case in which it can be tested, and which conflicts with physical experiments such as the Bell-Aspect experiments.
But my qualms with your theory don’t arise from my rejection of matter, but from your rejection of supervenience. You seemingly claim to have disproved supervenience in favor of identity.
I have given examples of scenarios, virtual worlds, in which the same available facts hold as premises as hold in the real world, and where the answer turns out (perhaps unbeknownst to those within the virtual worlds) to be supervenience but not identity.
Your response seems to have been to accept that your conclusions do not stand, but without showing how this inconsistency in the conclusions should arise from measurements done with available facts within the simulation.
To me, you and Dennett are each disagreeing with Chalmers in related but very dissimilar ways. I had proposed two classes of virtual worlds, some of which (as computer simulations) were intended as examples of Chalmers’s zombie worlds and some of which (as dreams) were intended as examples of zoombie worlds.
That is, I was modeling these worlds with concrete examples of particular ways of going about it, in hopes of finding out how some measurement done within these worlds would support the different conclusions.
Dennett also denies that you can separate the operational characteristics of thought from awareness, but then, unlike you, would have asserted that real awareness (as real as the awareness had by you or me) exists within the simulated zombie worlds.
So, I wonder how you would argue against Dennett, a fellow physicalist. What would you propose as an experiment to decide, between yourself and Dennett, whether the simulated intelligences had real qualia?
What is the appropriate physical test for awareness?
I suspect that Dennett would be as unhappy as I am with your defining qualia as necessarily involving “wet matter”.
Well, you said that completely artificial neurons weren’t good enough, and that completely natural neurons were good enough.
No, I didn’t. I said that a simulation of neurons on a computer isn’t good enough. And for the same reason that a simulation of an H20 molucule (using computers that model the elements) isn’t a water molucule.
What I am trying to get at is that there is no physical test for awareness, in the ordinary sense of that word, and you are trying to create such a test by making arbitrary definitions about the identity between qualia and brain states and thus are arbitrarily defining humanity.
I am not making an arbritray definition. I am defending a proposed scientifically discovered identity. You seem to keep forgetting that in this post all I was doing was responding to a famous argument that mental and physical states cannot be identical. I was trying to show how the proposed problem can be solved. In fact, all such problems are eaisily solved. So there is no reason not to think that mental properties/states are identical with physical properties/states. That is the better theory, and since there is no reason to think that it is false I propose that we think that it is true.
Sorry, I was imprecise. You are denying supervenience in that you are taking any evidence of supervenience as evidence of the stronger claim of equality, and then challenging those who believe only in the weaker claim to provide public evidence that your stronger claim is inappropriate.
Well, since physicalism is the claim that any world physically identical to ours is a world where there is consciousness that means that physicalism is the claim that the mental logically supervenes on the physical, where as dualism is the view that it only nomologically supervenes on the physical. Does that help?
Furthermore, there is no operational difference between the physical properties of the world as we measure it and the measured physical properties of a simulated world within a simulation. You assert that lightning within the real world is different from a simulation of lightning within a virtual world, but the fact remains that every measurement within the real world would have an operational analog within the simulation and so there is no operational difference between the real electrical discharge as measured in the real world and the simulated electrical discharge as measured within the simulation.
This whole preceeding bit assumes a view of science and a philosophical theory of perception that I do not accept. Of course there is no operational difference, but what has that got to do with the real world? But, seriosuly, I don’t want to get caught up in this debate
If you have a theoretical basis that provides facts about the world, and if this theoretical description is complete, then every fact about the physical world should also be a fact within the theoretical framework.
See, this is where I disagree. It is not the case that every fact about the physical world should also be a fact within the theoretical framework! Rather, as I have said, it is that all of the facts about the physical world can be deduced from the theoretical framework. Doing it your way makes it the case that being a chair is a non-physical fact (it is not a fact in the theoretical framework), but that’s just silly!
It is strange to treat as the default view one that is demonstrably wrong in every case in which it can be tested, and which conflicts with physical experiments such as the Bell-Aspect experiments.
I would LOVE to hear how physicalism conflicts with the Bell-Aspect experiments…and as I have already stated, I just don’t agree that a virtual model is a good test of physicalism.
But my qualms with your theory don’t arise from my rejection of matter, but from your rejection of supervenience. You seemingly claim to have disproved supervenience in favor of identity.
No, as I have said, what I claim is nothing you have said disproves identity…
I have given examples of scenarios, virtual worlds, in which the same available facts hold as premises as hold in the real world, and where the answer turns out (perhaps unbeknownst to those within the virtual worlds) to be supervenience but not identity.
No you haven’t. There is no consciousness in a purely virtual world and in a world like the Matrix world the conscious that is there is completely physical…perhaps the people ‘in’ the simulation can’t tell that, but that has always been one of the strategies of physicalism.
To me, you and Dennett are each disagreeing with Chalmers in related but very dissimilar ways. I had proposed two classes of virtual worlds, some of which (as computer simulations) were intended as examples of Chalmers’s zombie worlds and some of which (as dreams) were intended as examples of zoombie worlds.
A virtual model of the physical world is NOT a zombie world. A zombie world is one where there are actually electrons, protons, neutrons, etc (or whatever a completed microphysics apeals to) and yet lacks consciousness. A virtual world does not satisfy the first part of the criterion.
Now, if it turns out that we discover that this world is just a simulation (and that there is nothing outside of it which influences it, so not the Matrix) then we will have discovered that consciousness does arise in a simulation. That is because we have good reason to think that whatever OUR brains are, they give rise to consciousness.
As to the second point, I certainly like you trying to bring in zoombies, but a dream world is NOT a zoombie world. When I am dreaming, everything that is going on is going on in my vrain, which is purely physical. There are no non physical properties in a dream, unless one is already assuming that physicalism is false (and I am not!).
I suspect that Dennett would be as unhappy as I am with your defining qualia as necessarily involving “wet matter”.
That’s because Dennett is a functionalist and I am an identity theorist…
Richard,
I was a little surprised you picked step 5. To me it is probably step 2.
While there is no way to accurately create a model of something without creating a real duplicate of it you could create something that can reasonably be called conscious by most people. Of course the conciousness is based on real matter as opposed to being just virtual.
Tom,
I’d say the reason why we don’t have an ideal test for consciousness is that we (most people) aren’t comfortable giving consciousness a tight physical definition. That is totally different from it being impossible to give it a tight physical definition.
For example, I could describe some tests but my definition would probably be very broad… For example one might use an argument in regard to animals or the development of babies asking where does consciousness begin – when does the dualist ‘X factor’ kick in? and why?
It seems to me the only good answer is that there is no non-arbitrary point (which leave some uncomfortably tough bullets to bite).
In a similar vein I propose a thought experiment here would be to define an arbitrary boundary that defines ‘you’ (maybe just your brain?) and then slowly add or subtract a miniscule things and ask what each of those “beings” is experiencing first person. I suggest there is an answer to those questions and no principled reason to draw a clear line where you can stop. The point being that even your first person experience merges seemlessly into the rest of the universe.
Regarding quantum mechanics – the results are largely explained by multiple worlds model without having to get all mystical. The fact that your “awareness” follows a path through those worlds gives you an illusion that it has a special status fully explained / required by the theory. I.e. that most physicists would actually say there is no conflict with physicalism in their work, very far from refuting it.*
I’d expect they would also be unconfortable with the idea of a “pure referent” and more comfortable with the idea of seeing the same thing from different angles even if your angle of your experience is by definition a unique one.
* I do see matter and space as best described as a collection of numbers. That is just that our models of reality are finite and not by definition based on truth and thus more flexible models like maths tend to be better. So someone who suggested, for example, atoms as being fundamental some centuries ago was inaccurate but on the right track in the same sense as later people who suggested particles or waves as being fundamental.
Hey GNZ,
I pick step 5 because I think that mental states are identical to brain states and computer simulations of neurons don’t give you brain states, just as a computer simulation of a molucule with atomic number 86 doesn’t give you gold (but also as I said, this is an empirical question)…but if you want to pick step 2, the question is then what’s the relevant difference between that and step 1?
As for your second claim, if you mean to be suggesting that we could build an artificial brain that resulted in a robot, or computer, or whatever, being conscious then I don’t necessarily disagree. My disagreement is with the idea that a simulation is as good as the real thing. That seems obviously false to me.
With the pencil and paper model it seem like too much work is being done by the person drawing up the next stage of the model. In the computer model there is an analogous thing doing that work, but it is part of the model.
No, I didn’t. I said that a simulation of neurons on a computer isn’t good enough.
The situation in step 4 can be taken as a simulation of a brain with artificial neurons, each of which models a brain neuron, interconnected in the same pattern as brain neurons are interconnected. Essentially, these are prosthetic neurons.
They are not neurons, but there is a mapping between the relationships between these artificial neurons and the relationships between real neurons in a brain. That’s why I went back to whether you thought that awareness was an object or a relationship between objects.
Also, you seem to be arguing here that the relationship is real with respect to awareness and brain states (so that awareness actually exists, despite its arising only in the relationship between neurons), but in another post about Hume’s ideas of causation you hold that causation relationships between objects are not real in the same way.
See, this is where I disagree. It is not the case that every fact about the physical world should also be a fact within the theoretical framework! Rather, as I have said, it is that all of the facts about the physical world can be deduced from the theoretical framework.
Well, I don’t think my argument requires you to go any farther than that.
Let P be the set of facts about reality that can be deduced from a completed theoretical framework. What we mean by that, I think, is that there is a mapping from the facts in P to reality such that the relationships between the symbols of P correspond to the relationships between the referents of those symbols in reality (i.e. symbols map to referents and relationships between symbols map to relationships between referents, such that the invariants in the relationships between symbols correspond to the invariants in the relationships between referents).
A simulation can (in principle) be created in which all of the facts in P can also be mapped to objects within the simulation, again with corresponding relationships. Thus all of the facts in P apply to the simulation as well as to reality.
It seems to me that you are objecting that there are additional facts, call the set of them S, about the simulation (but not available within the simulation). I certainly agree that this is the case. For example, we know there is a real dreamer with a real brain, but this is not an available fact within the dream (we cannot do an experiment within the dream that locates the brain of the dreamer).
So the brain state of the the dreamer is a fact within S, and within P as it is mapped to reality, but not within P as it is mapped to the dream. (The facts about the dreamer within P as mapped to the dream refers to a dreamed dreamer who is having a dream within the dream, not to the physical dreamer.)
You seemingly hold that awareness arises as a fact within P when it is mapped to reality, but that does not arise as a fact within P when it is mapped to the dream, arguing that where we have (P union S) your argument does not apply to the facts of P. That is what I don’t understand.
That is, if I can deduce “no awareness except as it corresponds to a fact of P” from the facts of P in the absence of S, then I should also be able to deduce that from the facts of P in the presence of S. Adding additional facts does not threaten the verity of the conclusion of a valid argument.
Thus it seems to me that, even within physicalism, the existence of the awareness within the dream as a fact of S, and not of P (as it corresponds to the dream), shows that your argument is invalid.
em>Doing it your way makes it the case that being a chair is a non-physical fact (it is not a fact in the theoretical framework), but that’s just silly!
What we mean by the reality of a chair as an object is that there is a particular way in which we can be aware of it. Not that it is consistent with some mathematical abstraction (physical laws) but that it exists (or can exist) within our awareness.
So what do you mean by “physical”?
I think that if you don’t appeal to awareness, then the meaning of “physical” refers only to the idea that the abstract model can be mapped to the reality with corresponding relationships between the parts. But the simulation is also consistent with the mathematical abstraction in that way. And, for that matter, the mathematical abstraction corresponds to itself in that way (via an identity mapping).
Frankly, I would like to understand your argument better. I have a sneaking suspicion that identity theory could reconcile the idealist and physicalist positions as two ways of looking at the same identity.
That is, you see the identity as a reduction to the physical and I might see it as a reduction of the physical to awareness, but an identity ought to work in either direction. Despite the reduction, I think you believe that awareness is real, just as I believe that chairs are real, we just have a different idea about which kind of reality is prior.
GNZ: why does that matter?
Shack:
Also, you seem to be arguing here that the relationship is real with respect to awareness and brain states (so that awareness actually exists, despite its arising only in the relationship between neurons), but in another post about Hume’s ideas of causation you hold that causation relationships between objects are not real in the same way.
What post about Hume and causation are you refferring to? The only one I remember doing I argued that the causal relationship was real and was observable…but you are right that I think that cobsciousness is real. I am not an elimativist about consciousness…
I am afraid that I don’t really understand what you are getting at in the rest of your comment. You seem to wan to say that there is consciousness “in the dream” but that makes no sense! There is consciousness in the world that the actual dreamer is located in but not in the dream world. But if we do assume that what we perceive as the physical world is a vast representation, like what people in the matrix are experiencing (and this seems to be what you think) then in such a situation we would be able to deduce that certain brain states (in the matrix) result in conscious experience (in the actual brain though we mistakenly attribute it to the represented brain) if the representation-reality relation is faithful then we will still be essentially right. So even though we cannot locate the actual physical brain we will be right that states of such a thing result in conscious experience by proxy. If the representation-reality relation isn’t faithful then we don’t really have access to the completed microphysics and yet it would still be right tot say that access to such a theory would allow us to make the required deduction. So talking about the dream world as though it were a seperate world is a mistake.
Frankly, I would like to understand your argument better. I have a sneaking suspicion that identity theory could reconcile the idealist and physicalist positions as two ways of looking at the same identity.
In principle I agree. When I say that conscious is physical I mean that it is the same, ultimately, as table and chairs. If idealism is right and tables and chairs are simply ideas then physicalism as I understand it could be re-interpreted in the right way and still come out to be true. The claim is that when we have a completed theory of the ways that tables and chairs (and all of the other stuff we iinteract with) behave the fact that there is consciousness will follow from that theory. The argument I am presenting is an argument against dualism not against idealism…as you may know from my other posts I am neutral about whether idealism has been refuted (I suspect that we have simply defined it out of existence by adopting certain theories of perception)
[…] considered as actual even though the painful stuff picks out C-fiber firing here? The answer is arguably yes! This is something that may be surprising but it is arguably shown to be true by empirical results. […]
I am afraid that I don’t really understand what you are getting at in the rest of your comment. You seem to wan to say that there is consciousness “in the dream” but that makes no sense! There is consciousness in the world that the actual dreamer is located in but not in the dream world.
Sorry for the delay.
The point is that physicalism asserts that every fact in reality corresponds to a fact in the theoretical framework.
Physicalism applies to the virtual world, because everything in that world corresponds to a fact in the physical framework. In a complete simulation of the theoretical framework, every fact in the framework will correspond to a fact in the virtual world. Because the theoretical framework is simply a mathematical model, such a simulation is possible.
You assert that there is some “physical dreamer”, but that assertion only seems to make sense within physicalism as saying that there is some aspect of the theoretical framework that corresponds to the physical dreamer. There is, within physicalism, no real dreamer in any sense other than that.
But that aspect of the theoretical framework also corresponds to an object within the dream (i.e. I can dream about myself).
That is, there is no actual substance in physicalism, you never transcend the abstract theoretical model to talk about anything that actually exists.
The meaning of the word “exist” does not correspond to facts in the theoretical framework. That is, a completed theoretical framework will still not be able to show that reality actually exists.
Idealism escapes that by asserting, contrary to physicalism, that there is something (namely awareness) that exists in reality but does not correspond to anything in the theoretical framework. It is only because idealism is *not* a physicalism that awareness can thus enter the dream from the outside.
In idealism, the objects of the real world actually exist, because they are objects of awareness, whereas in physicalism they never transcend the abstract, because there is no fact in physicalism that is not a fact within the abstract theoretical framework. Physicalism turns reality into a mere abstraction. In idealism, real objects actually exist in a meaningful way (e.g. when you kick a rock you are aware of it in a way that isn’t just theoretical and abstract).
If idealism is right and tables and chairs are simply ideas then physicalism as I understand it could be re-interpreted in the right way and still come out to be true.
You see, this is the kind of statement that I think gets it exactly backwards. It is in idealism that tables and chairs actually exist (i.e. we can be aware of them). In physicalism there is no fact about a table or chair that does not correspond to a fact in the abstract theoretical framework, thus it is in physicalism that the tables and chairs are merely ideas.
[…] instead thinks that they are in pain but that pain and painfulness are (contrary to appearances) only contingently related. What are we to say? In the paper Dave says, A fifth issue is the worry that subjects might […]