Is the existence of a p-zombie a self-consistent idea?
I suppose most of us are familiar with the concept of philosophical zombies, or p-zombies for short: beings that appear and act like humans and are completely indistinguishable from humans but do not have consciousness.
However, a p-zombie, despite having no consciousness, reacts to stimuli in exactly the same way a human being would. A human, however, is aware of its own consciousness and sentience, and this awareness in itself is a perception that human reacts to. Therefore, wouldn't a p-zombie notice its lack of consciousness and experiences and comment on these, thus not being completely similar in its actions to a human being?
If this kind of being - one that reacted to all its perceptions in a way a human would, but did not have the perception of the conscious experiences or thoughts - what would its reaction to this then be like? Would the p-zombie appear to be panicking or show some other emotion, or, upon realizing that it had no emotions, would it also stop expressing them?
I think this rises yet another quite interesting question: what do we even react to? Do we react to external stimuli, or our experiences of them? I think this is what defines the answer to the previous question as well; if the answer is the former one, the p-zombie would indeed mostly act like a human, but were it the latter one, the p-zombie would have no motivation to perform any action at all or react to any stimuli except by its reflexes, and die of starvation.
However, a p-zombie, despite having no consciousness, reacts to stimuli in exactly the same way a human being would. A human, however, is aware of its own consciousness and sentience, and this awareness in itself is a perception that human reacts to. Therefore, wouldn't a p-zombie notice its lack of consciousness and experiences and comment on these, thus not being completely similar in its actions to a human being?
If this kind of being - one that reacted to all its perceptions in a way a human would, but did not have the perception of the conscious experiences or thoughts - what would its reaction to this then be like? Would the p-zombie appear to be panicking or show some other emotion, or, upon realizing that it had no emotions, would it also stop expressing them?
I think this rises yet another quite interesting question: what do we even react to? Do we react to external stimuli, or our experiences of them? I think this is what defines the answer to the previous question as well; if the answer is the former one, the p-zombie would indeed mostly act like a human, but were it the latter one, the p-zombie would have no motivation to perform any action at all or react to any stimuli except by its reflexes, and die of starvation.
Comments (37)
yeah you're right. The closest thing we have are people with Aphantasia who lack a "mind's eye" https://en.wikipedia.org/wiki/Aphantasia however during interaction with other people they are able to comprehend they are lacking something the others have.
The p-zombie argument is built on the idea that consciousness is epiphenomenal. So you can have a zombie that relates every detail but has no phenomenal version of events. In that sense it's stupid but works well as an argument against mechanism. If anyone thinks it can be true besides being a philosophical thought experiment (sort of like the Socractic dialogues) then they are nuts.
He'd start rambling on about consciousness like David Dennet :smile:
Take the Conscious Visual experience of the scene the p-Zombie is looking at. Without the Conscious Visual experience the p-Zombie would be Blind and would not be able to move around in the World without bumping into things. The Conscious Visual experience is the final stage of the Visual process. People who think the Conscious experience is not necessary for the p-Zombie to move around in the World are exhibiting Insane Denial of the purpose of the Conscious Visual experience.
Sure, a p-zombie would notice something missing, just like a blind person notices something missing when they hear others talking about their visual experiences.
P-zombies wouldn't have dreams or nightmares.
P-zombies wouldn't have that internal voice telling them what would be right or wrong, or self-reflection. Being a p-zombie would be like having blind-sight. People with this condition are aware that there is something there, but are not clear on the details of what that thing is. They do not behave like normal humans as a result of this lack of visual consciousness.
The behaviour of computers are driven by bottom-up processes (mechanical/hardware) and top-down processes (non-mechanical/software). P-zombies would be like "computers" without software. They would be more like a typewriter that simply reacts to external input and can't perform actions based on its own internal programming and stored information. That would be the kind of difference we would see between a human being and a p-zombie.
How would a non-conscious being reflect on its own condition? Doesn't the term ''zombie'' specifically deny self-awareness of any kind?
Good question, as a thought experiment I think one could think of a conscious being losing their consciousness, thus having knowledge of the nature of consciousness without having it.
Quoting TheMadFool
Depends on how self-awareness is defined, I guess. By the definition of p-zombie, it should act like it had self-awareness.
This is a puzzle to me. A p-zombie, the way you describe it, should be indistinguishable from a person who actually is conscious.
Isn't this a ''dead'' end?
If you define y as x and then it becomes impossible to inquire into how x and y may be compared.
Anyway I started a thread on self-awareness and whether that is a good thing or not.
I love the way some philosophers (e.g. Dennett) handle it though: they cut the Gordian knot by saying simply (in complicated ways) that we are p-zombies! :)
More of "y should (appear to) be x". Sure the zombie would not have self-awareness in the sense that it wouldn't exactly have any knowledge at all, although its communication would reflect the information stored in its brain,
The notion of a philosophical zombie is a manifestation of the Spiritualist confusion of academic philosophers.
Any device that can do what a person or other animal can do has "Consciousness". That's how it does those things, you know.
We're purposefully-responsive devices, designed by natural-selection, to achieve certain material goals and purposes. Consciousness is the property of being a purposefully-responsive device.
But the word "Consciousness" is, of course, used very chauvinistically. We only apply that word to those purposefully-responsive devices that are sufficiently similar to us. Other humans, or maybe other mammals. ...maybe the vertebrates, or maybe include the insects...and so on.
Yeah, but when it comes down to it, you're not different in kind from a mousetrap or a refrigerator lightswitch.
Michael Ossipoff
No, it does those things through causality causing it to do those things. Doing those things does not imply consciousness being involved.
Quoting Michael Ossipoff
Where is the consciousness in a mousetrap? How does it rise from its physical nature?
I’d said:
You replied:
You could say that about any event, explain it by saying that it’s caused by causality. For example, that could be said about the operation of a Roomba. What the Roomba does is caused by its built-in purposeful-responsiveness. …as can also be said of any animal, or humans too. The behavior that makes you say that a person has Consciousness is the result of designed-in purposeful responsiveness. If there’s any Consciousness there, it’s the purposeful-responsiveness itself.
The (hypothetical) philosophical-zombie has that purposeful-responsiveness too. How else do you think it does what a human does?
Correct. It doesn’t imply the kind of Consciousness that you’re talking about. …some Spiritualist notion of a separate entity, separate and different from the body.
But animals have what you could call “Consciousness” because animals are designed to do various things, to accomplish particular goals and purposes. What you regard as a separate entity called “Consciousness” consists of the design-purpose built into animals (including humans). …their designed-in purposeful-responsiveness, to accomplish their design goals.
I’d said:
You replied:
For the purpose of our animal-chauvinist usage in speech, I define “Consciousness” as purposeful-responsiveness of a device with which the speaker feels kinship.
You don’t feel kinship with a mousetrap.
Yes, it’s a vague definition, because our use of the word “Consciousness” is imprecise. Really, “purposeful-responsiveness” is a better, more uniformly-used term.
It isn’t some separate thing that “arises”, “supervenes” or “emerges”, or whatever. It’s just the property of purposeful-responsiveness. …but you don’t call it “Consciousness” when possessed by something you don’t feel kinship with.
But would you say that insects aren’t conscious? They go about their business, as do mammals, and they experience fear. In those regards, they’re recognizably similar to us.
Michael Ossipoff
I think the problem is that questioning whether other people have consciousness isn't really a well-formed question (or, it's not a question that actually makes any sense), it just seems to be.
Actually digging into the philosophy a bit: there are broadly two senses in which we use the concept of consciousness:-
A. A publicly verifiable thing. We observe the world, we see things avoiding some things, cleaving to other things, and that kind of behaviour is what we call "conscious." Animals have it, although there's a fuzzy line where animals don't seem to have it any more and their behaviour can be explained as purely mechanical and "robotic" (e.g. bacteria, ants, things like that). And this sense of consciousness has reasonable, publicly verifiable tests, like "How many fingers am I holding up?"
B. Something each of us has or is that is completely private (no one else can see your consciousness, or can tell whether you "have" it or not).
I think the problem is that, generally speaking, the latter sense, which probably derives from religion (ideas about soul, spirit, etc.), is often confused with the former sense, in that having B is supposed to be what makes A possible. B is supposed to be an un-publicly-verifiable inner thing that makes the publicly-verifiable A behaviour happen.
Most of the philosophical battles in philosophy of mind seem to be around some people thinking that B is a real thing, and that if you don't have it, you can't be conscious, even under conditions when most would say you're conscious in sense A (the p-zombie).
Detractors of B point out that we seem to get along perfectly well in the A sense regardless of whether we have B or not.
I don't know what the answer is, it depends on how you squint at it. Sometimes one thinks having this inner "registering" of outside events and of inner thoughts is the most weird and important thing in the world, and that denying it would lead to hideous atrocity; sometimes the detractors seem right and it's all a fuss about nothing.
One point: it looks like we're getting pretty close to being able to scan peoples' brains and know what they're thinking. If that's the case, then in the end, the complaints just come down to an ontological truism elevated to something grand and mysterious: I am not you, and you are not me.
It's also incorrect as it does not describe what the term "consciousness" refers to. As I stated, purposeful-responsiveness doesn't imply consciousness, but strictly speaking consciousness doesn't imply purposeful-responsiveness either. In contrast,
Quoting Michael Ossipoff
is what the word means and refers to.
All conscious beings that exist are purposefully-responsive because they are a part of our universe, a property of which causality is, making actually everything purposefully-responsive from some points of view. Being a part of the universe isn't a good definition for conscious beings either.
What it amounts to is that you're using "Consciousness" to refer to something that I don't believe in.
You're in good company. A lot of academic philosophers do too.
But you're implying that the meaning that you give to that word is its only correct meaning.
I don't think that there's a consensus about what the word "Consciousness" means.
If someone built a perfectly human-impersonating robot, and said that it's a philosophical zombie, and if its behavior and reactions were indistinguishable from those of a human, then how would you prove that there's some un-bodily, Dualistic, "Consciousness" that it lacks? Don't you see the weakness of that position?
There'd be no reason to believe that that robot lacks anything of a human.
In actuality, robots will probably be made without the natural-selecton-built-in self-interest that humans and other animals have. No need to worry about a selfish robot take-over. They'd have more resemblence to ants than to self-interested mammals. Of course ants and bees will protect themselves, but they value, first, their service to the colony.
Michael Ossipoff
I suppose the Consciousness that you're referring to needn't be unbodily. Some people are claiming that there's some mysterious way that the brain causes it. Yeah: Purposefulness. That's it. Nothing mysterious there.
But Humans don't work like Robots. Humans and probably all Conscious beings have a further processing stage that presents the Visual experience to them. The Visual experience is what we use to move around in the World. When I reach out to pick up my coffee cup I see my Hand in the Conscious Visual experience. If my hand is off track I adjust my hand movement until I can touch the handle and pick up the coffee cup. It would be much more difficult to do this without the Conscious Visual experience. The Conscious Visual experience contains an enormous amount of information that is all packed up into a single thing. The Neural Activity is not enough. We would need far more Neural Activity to equal the efficiency that the Conscious Visual experience provides us.
If a Computer could experience for example the Color Red then I would agree. But a Computer does not Experience anything. A Computer can be programmed to scan pixels in an image to find the Red parts. A Computer will look for pixels with values that are within a certain range of numbers. A Computer never has a Red experience but it can find the Red parts of an image. So just because it can find the Red parts of an image, like a Human can, it does not mean it has a Conscious Red experience while doing this. A Computer works in a different way than a Conscious being does. Science doesn't understand enough about Consciousness yet to design Machines that have Consciousness.
...and you know that....how? If you aren't a computer, then how can you speak for what a computer does or doesn't experience?
What does experience mean? I define "experience" as a purposefully-responsive device's interpretation of its surroundings and events in the context of that device's designed purposes.
By that definition, yes a computer has experience.
As I said, we tend to use "Consciousness" and "Experience" chauvinistically, applying those words only to humans or other animals. That's why I try to cater to that chauvinism by sometimes defining those words in terms of the speaker's perception of kinship with some particular other purposefully-responsive device..
When you find the red part of an image, why should I believe that you have a red experience in a meaningful sense in which a computer doesn't?
The computer finds the red part of the image. You find the red part of the image. Period (full-stop).
You wouldn't report the red part of the image if you hadn't experienced it. The same can rightly be said of the computer.
You call it a Conscious Experience when it's yours, or of another person, or maybe another animal. ...you or a purposeful-device sufficiently similar to you, with which you perceive some kinship.
...because you define a Conscious Being as something very similar to yourself.
...if you're defining "Consciousness" as "ability to pass as human".
Current technology can't yet produce a robot that acts indistinguishably similarly to a human and does any job that a human can do.
Imitating or replacing humans is proving more difficult than expected. Life has evolved over billions of years of natural-selection. It wasn't reasonable to just expect to throw-together something to imitate or replace us in a few decades.
If such a machine is ever built, some would say that it has Consciousness and Experience (as do we), and some would say that it doesn't (and that it's a philosophical zombie merely claiming to have feelings and experiences).
Of course the former would be right.
Michael Ossipoff
Seriously ... you think a Computer experiences the color Red like we do? You know that the only thing happening in a computer at any instant of time is : Add, Subtract, Multiply, And, Or, Xor, Divide, Shift Left, Shift Right, compare two numbers, move numbers around in memory, plus a few more. If you have 4 cores then this is happening in 4 different places in the computer chip. Which one of these operations experiences the color Red?
Quoting Michael Ossipoff
Question is: Do you have a Red experience in any meaningful sense. Think about your Red experience. Think about the Redness of the Red. That Redness is a Property of a Conscious phenomenon. Think about how a Computer works. Add, Subtract, Multiply, etc. There are categorical differences with how the Human Brain functions and how a Computer functions. A Human Brain has trillions of Neurons firing simultaneously at any instant of time. A 4 core processor chip only has 4 places where things can happen at any given instant of time. Effectively a 4 core computer chip has only 4 Neurons.
Quoting Michael Ossipoff
I showed you how a Machine detects Color. It compares numbers in memory locations. It makes no sense to think that it also has a Red experience. It doesn't need a Red experience to detect colors. Machines and Brains do things using different methods.
Quoting Michael Ossipoff
I'm not defining Consciousness as the ability to pass as Human. Most Birds can probably have a Conscious Red experience.
I think that since our Brains are made out of Physical Matter that other kinds and Configurations of Matter could produce Consciousness. Science just needs to understand Consciousness more. I think that someday there will be full Conscious Androids and not mere Robots.
So, how to test for a P-Zombie: ask some questions, like, 'how do you feel right now?' 'what is the most embarrassing thing that ever happened to you?', or 'what's you're favourite movie, and what did you like about it?' Engage it in conversation. I can't see how it could maintain the pretence of being, well, 'a being', for very long, as all it can do is regurgitate, or combine, various responses and information that has been uploaded into it (how, by the way? Is it a computer? If so, could it pass the Turing Test?)
These challenges from 17th century philosophers still seem germane:
René Descartes, Discourse on Method (1637)
Gottfried Leibniz, Monadology
Quoting Wayfarer
Isn't that what humans do as well? We are fed information through our senses in infancy and childhood that we over the course of years learn how to react to.
Is the converse true? I think a robot works, although in a simplified way, like a human, making it possible for it to replicate the actions of conscious beings.
Quoting SteveKlinko
I think the opposite is the case. A conscious experience, whatever its benefits are, cannot be efficient. While containing all of the visual data provided by eyes, it also contains the experience of that data, which is such a rich experience we ourselves can't even begin to comprehend how it is created. The brain also unconsciously organizes and edits that data to a huge extent, filling gaps, causing us to perceive illusions, basically expanding our visual experience beyond what information is provided by the senses. For example,
Quoting SteveKlinko
a robot would only need to find a specific kind of group of pixels with a color matching the color of the cup. Conscious mind, for some reason, in a way wastes energy forming an idea of "cupness", equating that cup with other cup and connecting it to its intended usage as well as all the memories (unconscious or conscious) an individual has relating to cups. All that information could be broken down to individual points and be had access to by a robot, but instead human mind makes something so complex and incomprehensible.
The existence of that idea also allows me to, while seeing a simple cup, appreciate my conscious perception of that cup. I still can't see the evolutionary value of that appreciation, though.
I think the Descartes of the quote I provided could have anticipated something like Siri.
Quoting BlueBanana
You're taking a lot for granted, and in such matters, that is not wise.
You're missing the reality that the Robot would most definitely need the concept of cupness to operate in the general world of things. Knowing the color of the handle of one particular cup might help with that cup. In the real world the Robot would need to understand cupness in order to find a cup in the first place. Then when it finds a cup it can determine what color it is.
Like I said, p-zombies cannot be programmed. They are dead inside. Humans are more like robots, where p-zombies are more like a mechanical contraption without any capacity for programming. Humans are programmable. P-zombies are not.
Quoting Wayfarer
Every time you are asked what it is that is missing when we compare humans to computers, you weasel out of answering the question.
Quoting SteveKlinko
So, we design a robot with templates - a template for cups, for humans, for dogs, for cars, etc. - just like humans have. We humans have templates stored in our memory for recognizing objects. We end up getting confused, just like a robot would, when an object shares numerous qualities with different templates. The solution is to make a new template, like "spork". What would "sporkness" be? Using the word, "cupness" just goes to show what is wrong with philosophical discussions of the mind.
It would need the information the concept of cupness holds - or, some of that information. It doesn't need to compose all this knowledge into this abstract construction that also holds much unnecessary data. Cupness is much more than what a cup is and what counts as a cup, or even all the knowledge concerning cups there is. Cupness is like a generalized form of a thought (or dare I even say feeling) of a cup, something only a conscious and sentient being can grasp. That simultaneously compressed but also overly complex thought is something I don't believe would be necessary.
If they have memory, they can learn, and therefore they can be in a sense programmed through conditioning. So, do they have memory? Is memory a property of only mind, only brain or both? I'm not an expert but afaik the existence of memories in brain has been scientifically confirmed.
Either way, the reflexes of p-zombies should work normally so at least classical conditioning should work on them.
Let's interpret a footstep as a memory of something stepping there. What entity has this memory? Ground? The Earth? Some even larger system? Can any of these be said to have a mind?
Quoting Harry Hindu
Sentience to begin with. Conscious experiences.
Quoting Harry Hindu
They don't express any attributes that would imply them having a mind, and neither does their physical structure.
I agree about templates but don't understand your objection to saying cupness or even sporkness.
The Computer Mind would be equivalent to the Physical Human Mind (the Brain). But Humans have a further processing stage which is the Conscious Mind. When Humans see the Color Red there are Neurons firing for Red. But with Humans there is also that undeniable Conscious Red experience that happens. You can't really believe that a Computer has an actual Red experience. That would imply some Computer Self having that experience. Science knows very little about Consciousness so who knows maybe even a grain of sand has Consciousness. But you have to draw a line somewhere in order to study the problem.
I don't know what you are asking here. IF we were to interpret a footprint as a memory of something stepping there, then yes, it would be part of the the mind of the Earth, I guess. But we don't call footprints "memories". We call them "evidence". We can also call your recall of a crime as evidence as well, so in a way, yes, memories are effects of certain causes, just like footprints. What makes footprints and memories distinct is that footprints are not used to obtain a certain goal by the same system that it is part of. Footprints are on the ground and part of the surface of the Earth. The Earth has no goals for which to use the memory/knowledge of that footprint, or any footprint for that matter. Organisms and computers (currently only working for human goals) are the only things that use information to obtain goals.
Quoting BlueBanana
...and what are conscious experiences?
Quoting BlueBanana
What attributes would imply that one has a mind? Wouldn't one attribute be that it uses stored information to act for it's own interests?
How do we know that "physical" structure has anything to do with it? What is "physical" anyway?
Then are you sure that you agree with me about templates? My point was that we have better, more accurate terms to use ("cup template") instead of these philosophically loaded terms, like "cupness".
Quoting SteveKlinko
No. You have to study it first to see where you should draw the line, or else that line would be subjective - arbitrary.
There is no red out there. Red only exists as a representation of a certain wavelength of EM energy. Any system could represent that wavelength as something else - the written word, "red, a sound of the word, "red" being spoken, another color, or even something else entirely. No matter what symbol some system uses to represent that wavelength of EM energy, others that also have a different representation could eventually translate the other's symbol for that thing. That is what we do with translating languages.
What we see is not what the world is. Our minds model the world as physical objects with boundaries and borders, but the world isn't like that. It is all process, which can include "mental" processes, and non-mental (what many might call "physical" processes). When you look at someone you see them as a physical being, but they are just an amalgam of certain processes, some of them being "mental" in nature. What I mean by "mental" is goal-oriented sensory information processing. Brains are what we see, but they are just models of other's mental processes.
YOU are a process. What I mean is, your mind is a process - a mental process. It is what it is like to be your mental process of simulating the world in fine detail so that you can fine-tune your behavior for the extremely wide range of situations you will find yourself in during your lifetime. Your conscious experience is just a predictive model of the world and is not as the world is in real-time. It is continually updated with new sensory information milliseconds after the events in the world.
So to say that computers cannot have minds seems to be out of the question, if we designed them to learn using the information they receive about the world and their own bodies through sensory devices and to represent the world (using the information from it's senses) in a way that enables it to fine-tune it's behavior to achieve it's own personal goals. In other words, the computers you have on your desktop probably do not have minds in the same sense that we do. There may be something it is like with it being a process like everything else, but it is without any self-awareness or independent thought.
So you are just arguing about symantics? For me Cup Template and Cupness have the same meaning.
Quoting Harry Hindu
I think Science will get nowhere if it insists that a grain of sand has Consciousness.
Quoting Harry Hindu
I agree except that the Conscious experience of something like the color Red is more than "Just a Predictive Model of the World". It is a Conscious experience. Science does not know how to explain any Conscious experience yet.
Quoting Harry Hindu
I say Computers that we have today don't have Minds but I didn't say that they can never have Minds. We first have to understand how the Human Mind works and only then will we be able to design Conscious Machines. But with Consciousness everything is still on the table. Maybe we should study the Consciousness of a gran of sand first, but I doubt the Wisdom or Logic of doing that..