I can barely imagine how a p-zombie even exists in any form, and the concept seems almost absurd to me. How can a being act exactly as if it had consciousness without having consciousness, unless the concept of consiousness itself has no meaning?
Daniel Dennet imagines the possibility of ‘zimboes’, an inversion of zombies; each zimboe thinks it is conscious. thinks that it feels pain and concern, thinks that it is in love, has self-awareness so on; a zimboe can even think that it is philosophical about the experience. It is just wrong about these beliefs in a way that cannot ever be detected.
If a zimboe is capable of philosophy and love, yet is not conscious because of an indetectable quality that is absent, then consciousness is a meaningless concept.
And yet, note that I said 'almost ’ absurd. I think that I can conceive of a situation where consciousness is absent, yet human-like zombies act as if they were conscious. If all human minds were replaced by computer-controlled automata, which in turn were controlled by an arbitrarily large computer program that could simulate every thought and feeling in the minds of the human population, the p-zombie state is just about conceivable.
Moderately convincing simulations of humans are already included with many electronic games and other media; given an arbitrarily sophisticated computer program the entire world could be replaced by such simulations, all controlled by a system that need not itself be self-aware or conscious in the same way that we consider ourselves to be.
I suppose the great test of such a system would be innovation; if a computer-simulated population of humans remained at a static level of development, never progressing in technological achievement or in social sophistication, then we could be fairly sure they were mere automata; but if this hypothetical virtual society progressed and evolved over time, then we could be fairly sure that the members of this society were functionally equivalent to humans, whether they possess ‘true’ consciousness or not.
Well, you beat me to it, but sure I can conceive of an entity acting conscious but not being conscious.
I might have a circuit in my mind that makes me grab my left foot and say “ow, ow, ow” after I dropped a bowling ball on it, despite not actually feeling any unpleasant sensation.
Of course there’s no reason for such a circuit to exist. Which is part of why I assume when you say “ow!” you are experiencing similar unpleasant sensations to me when I say “ow!”. But that’s besides the point.
The “zimbo” hypothetical I think ultimately just restates one of the unique properties of consciousness, that it cannot be an illusion. That if you feel as if you are in pain, then you’re in pain. So a zimbo is indeed the same as a conscious agent, but not the same as a p-zombie.
This is one of the arguments we Strong AI people use. By the time an advanced computer (or other synthetic mind) gets to the point where it can engage in philosophical thought, then consciousness has emerged.
The “qualia” might be totally different. It (they?) (is qualia singular or plural?) might not even exist at all. Just as you imagined an entity acting as if conscious, but not being conscious, so an entity might act as if it perceived qualia…but didn’t really. It doesn’t matter, because such an entity is reacting on the basis of the perception of something that models the external world. It might not be a personal qualia; it might only be a map, or a flowchart, but it is something.
Operating on a model of your own mind is a really big chunk of the definition of consciousness.
Would a person be considered to be conscious if they were, say, hypnotized? They might look conscious, but they are unaware if what’s going on. And yet I have heard accounts of where the hypnotized person remembers doing things but was incapable of stopping themselves from doing them. There was no conscious volition. One could perhaps say the same about brainwashing and cult techniques.
What is consciousness is the fundamental question of neuroscience, philosophy and theology as well as merry pranksters.
Big Al had this to say on the subject on the death of a friend: “Now he has departed from this strange world a little ahead of me. That means nothing. People like us, who believe in physics, know that the distinction between past, present, and future is only a stubbornly persistent illusion.”
Robin Williams posited: “Reality. What a concept.”
The difference between thee and I is only from where we experience our point of view. And you and I are temporary in every sense of the word. – TSS
How about deeply absorbed in reading a book or watching a movie?
Probably all of us have had that experience: we’re watching Casablanca (or Who Censored Roger Rabbit) and lose all sense of self. Our entire awareness is absorbed into the screen and the story.
Is it valid to say that consciousness comes in degrees, and can be partially impaired without being fully absent? Say, you’re really sleepy… Or you’ve had four beers… You’re still conscious…but you’re less conscious than you might otherwise be.
It doesn’t have to be Boolean. During the lassitude that follows having a big hearty meal, one might be “75% conscious.” When you’re on the freeway and a big semi blares his air-horn right behind you, you jump to 99.999% conscious, with an adrenaline chaser.
I would think at that moment the semi is probably the ONLY thing you become conscious of. I imagine full consciousness is when you become aware the most of everything around you. Or maybe not…
Maybe the really heavy and fast pounding of your own heart…
Interesting question… Is consciousness over a wide horizon greater or lesser than tightly focused consciousness? If I’m on a mountaintop, taking in the whole landscape, is that greater or lesser than when I’m squinting through a loupe at a fine detail on a D&D miniature I’m painting?
Clearly, the two kinds of consciousness are different, but can they be put into an ordering of “higher and lower” or “greater and lesser?” Honest truth, I don’t know!
It doesn’t have to be tricky. Just ask the AI if he’s conscious, and be sure he’s telling the truth.
AHUNTER3 knows that consciousness exists because he experiences it. However, he can’t assess whether you (or anyone/thing that is not him) are conscious. Why not? I assume it’s either because he cannot be sure you fully understand the question (i.e. “what is consciousness and do you have it?”), and/or because he cannot be certain you are telling the truth. If we’re certain that another intelligence understands our agreed-upon meaning of consciousness (including self-awareness, qualia, etc.) and we know that it’s telling the truth, then if it says, “yes, I am conscious”, it is conscious (…unless it’s self-deluded, but let’s not muddy the water with that possibility—I think self-delusion can be eliminated as a risk factor with AI).
We need not be hampered by the constraint of the Turing Test. We’re not trying to ascertain whether or not we are conversing with a human or AI, we’re assessing whether or not an AI has evolved consciousness. Turing subjects are programmed to lie and deceive; ours will be programmed to tell the truth.
Or, perhaps, it’s better to program our AI with the ability to lie (who knows, deception may be an integral part in the emergence of human-like consciousness), but assure that we can detect the lies 100% of the time. Heck, even a polygraph machine can detect human deception most of the time (even if you’re a sociopath or taking beta-blockers), and that’s using subjects (i.e. humans) that we (humans) did not program ourselves, using measuring devices that are rather primitive. I don’t believe a human being can knowingly tell a lie without there being reliably detectable physiological changes. We just need a more precise instrument to make lie detection 100% accurate. With AI, lie detection is even easier to master: just program his robo-nose to grow when he tells a lie.
You: “Robato, are you conscious?”
Robato: “Why yes, yes I am. Quite proud of it, too!”
You: “Robbie…your nose is growing…you’re fibbing to me aren’t you?”
Robato: *“Ya got me…I’m not conscious.” *
It is my understanding that squirrels don’t lie, but if you ask one if he’s conscious, he probably lacks the intellect to grasp the concept of consciousness and therefore cannot give you a meaningful answer (or, maybe he does give you a meaningful answer, but you lack the intelligence to understand his chirps). While deception may be our main stumbling block in assessing whether or not high-intelligent beings are conscious, lower intelligence subjects and/or language barriers are the primary stumbling blocks in assessing whether or not non-human biological beings here on Earth are conscious. I believe AI can come to the rescue in both cases. Lie detectable AI takes care of the first case.
Can AI someday be developed that will be able to answer the question, “what is it like to be a bat?”? I believe that it will be developed and be accurate…to a certain degree. Simply program the mind of a bat into a machine and run a bat-life simulation through it (complete with echo-location perception, etc). Then, ask him what it’s like to be him. And he will answer…nothing. He still has the mind of a stupid bat, remember? Plus, there’s that whole language barrier thing. So, we do have to change a couple of parameters in our experiment: we need to bump up the sim-bat’s intelligence at least to the point of understanding the meaning of consciousness, and we need to include a means of communication…and, of course, we need to make sure he’s not a liar, or bat-shit deluded. Then, we can ask:
You: “Are you conscious?”
Bat: “Yes, I am.”
You: “What’s it like to be a bat?”
Bat: “It’s like being a squirrel, only more psychedelic, man.”
I believe AI will soon give us a much deeper understanding of consciousness, both biological and artificial. Paradoxically, I think testing with AI subjects will give us a better understanding of biological consciousness than testing with biological subjects. We know exactly what is programed into AI; not so much with messy organic brains with their divergent evolutionary pathways and all. At the point AI tells us they now experience self-awareness and qualia, and we know we did not program those perceptions into it, we have proof those are emergent properties and the precise point when consciousness flames on.
Pain. Must one be conscious in order to feel pain? When one is in a semi-conscious state on an an aesthetic, would that be an indication that for one to be fully conscious, one must have the capacity to feel pain?
Another interesting thought exercise that (I think) Susan Blackmore proposed… using a dozen or so sticky notes, write upon each the phrase “Are you conscious now?” and place them at inconspicuous locations around your home.
What you (or at least some people) experience is that most of the time we’re not truly conscious until something prods us to meta-awareness.
This seems like a good thing to me. Veering off into my own opinion, consciousness is just an illusion, a momentary way to project our cognition in relation of external and future events. As such, it’s flawed by definition, and it is probably unhealthy to spend all one’s time in a fully conscious state (as I’ve characterized it).
In other words, if consciousness is just an illusion, then who’s being fooled? (Paraphrased from Conscious Entities, a blog that’s by now evolved to a valuable repository regarding the issues surrounding consciousness.)
While meditating, when one forgets about his body/mind , surrenders and lets go, the consciousness expands and one experiences a state of super-consciousness .