Please register to participate in our discussions with 2 million other members - it's free and quick! Some forums can only be seen by registered members. After you create your account, you'll be able to customize options and access all our 15,000 new posts/day with fewer ads.
Nothing is "missing" from a rock. It is matter. You are matter. The only difference is how that matter is put together - and in what form and configuration. There is nothing different about "living" matter than rock matter - except in how that matter is structured.
I'm going to jump in on this and say I agree - although, of course, I've also been saying that our current concept of matter is inadequate insofar as it provides no logical basis for explaining how qualitative experience emerges from the structure/activities of matter. Current models of physics can account for recursive/self-referencing material activity - which can seemingly account for the sorts of physical behavior that we attribute to conscious creatures - but there is still nothing in our current concept of matter, or the laws of physics, that gives any clue as to why material activity of this sort (or any sort) should qualitatively experience anything like pain or the redness of a sunset. We can account for wincing and rubbing a wound and, generally, engaging in "pain behavior," but no one has yet shown any logical basis in current physics for explaining why material systems engaged in this type of behavior should subjectively feel pain. The reality of self-reference gives us a good basis for explaining how there can be unique perspectives (i.e., subjectivity), but we still have no basis for explaining the qualitative feel of being a system adopting a unique perspective.
The best anyone has offered is a brute-fact correlation: "Material activity of type P just is pain, and that's the end of the story." That might be all that science can ever do, but I think that, with the right kind of conceptual breakthrough, science might be able to do better. (I acknowledge that, no matter how good your explanation is, someone can always ask another "why" question, and, realistically, we need to accept "brute facts" at some point, but I propose that physics has not yet hit the explanatory rock bottom with regards to qualia. With a better conception of matter, we can go at least one step further.)
Now, concerning the recent exchange between Mystic and monumentus, I'm wondering if there is any chance that we could all agree on specifically what we mean by "robot" and "machine." I'd like to propose the following:
Robot = An algorithmically-driven computer housed in a mechanical body that is capable of interacting with the world (e.g., wash dishes, play the role of a butler, engage in verbal conversation, etc.)
Machine = A physical system engineered and built by intelligent creatures (i.e., people).
By these definitions, all robots are machines but, of course, not all machines are robots. I've left the term "machine" extremely wide-open. A machine could be a self-replicating, self-maintaining, self-referencing system that can grow/evolve beyond whatever initial "programming" was build into it. A machine, for example, designed in accordance with the principles of a "neural network" could learn behaviors and display types of goal-seeking behaviors that the engineers/builders could never have predicted.
With these definitions, I'd say that a robot could not be conscious, but some types of machines, perhaps, could be. (Commander Data on Star Trek would be a sci-fi example of this type of machine.)
A key sticky point will be the concept of "algorithmic". An algorithm is a step-by-step procedure for calculating some more or less determinate output from a set of initial conditions/inputs. I say "more or less" determinate because an algorithm can incorporate a random number generator (RNG). Your laptop has a pseudo-RNG, but presumably there are, or could be, machines with quantum-RNGs that would be true RNGs. In fact, it would probably be sufficient for an RNG hooked to the internet to check some indexical fact like "current temperature at a given location" in order to count as truly random. The key idea is that a machine with an RNG can mimic the unpredictable nature of life-like behavior without necessarily having any actual subjective/qualitative sensations, desires, etc. A machine with an RNG can seem to "consciously choose" options when, in fact, it has no inner feelings about its choices one way or another.
I say that sentient creatures (or sentient machines - if we succeed in building them) will not be algorithmically-driven. Sentient machines will need to engage in "agent causation" - which is a concept that current physics is not able to handle. This "agent causation" is the reason why I say that current physics will eventually fail to reductively explain the activity of micro-elements in a sentient brain. In effect, I'm predicting that neurons (or, perhaps, sub-neural components?) will manifest patterns that ultimately violate quantum statistics. One possible way in which this might work: "Local hidden variables" would violate Bells Theorem, but quantum systems don't violate Bell's theorem, thus we conclude that there are generally no local hidden variables at the quantum level. But we currently have absolutely zero data showing whether or not quantum systems in living brains violate Bell's theorem. My prediction is that if/when we ever have such data, we will discover that Bell's theorem is violated in these systems. The source of this violation will be what I'm calling "agent causation."
Algorithmically-driven machines could, perhaps, mimic the behaviors of sentient agents, but mimicking sentient behavior, or modeling sentient behavior, is not the same as being sentient. According to my proposal, being sentient requires violations of Bells theorem in physical systems where, according to current theory, there should be no such violations.
I'm going to jump in on this and say I agree - although, of course, I've also been saying that our current concept of matter is inadequate insofar as it provides no logical basis for explaining how qualitative experience emerges from the structure/activities of matter.
That - to me - is something of a leap. You are leaping from "We do not currently understand consciousness is produced by matter" - to "therefore our concept of matter must be flawed or incomplete".
This is too much of a leap for me. We could get to the point where we understand everything there is to understand about matter - and still not understand how it produces consciousness.
Understanding the parts of something - and understanding how those parts work together in a given system - are two different things. And it is perfectly possible to understand all there is to know about the parts - and still fail to understand the workings of the whole.
Quote:
Originally Posted by Gaylenwoof
could all agree on specifically what we mean by "robot" and "machine."
I will stick to the meanings (that is I will steal them) that have been used by myself and others in the thread - which is to describe the brain as a biological machine that produces consciousness - and a "robot" or "AI" as a non-biological machine constructed to achieve the same thing.
Thus far mystic has failed miserably to do anything to support the notion that a non-biological machine is not every bit as capable of producing consciousness as a biological machine. The machine does the same thing - even if that parts used to make it are different.
The basis for claiming it is not possible is still clouded in dodge rhetoric alas.
So wait a sec. If current science as we know it has no problem dealing with the fact that subjective beings exist, what's the problem again?
The problem is that, according to materialism, the qualia dealt with at the level of neuroscience and psychology ought to emerge from the elements and principles of fundamental physics. Until we can explain this emergence, the term "emergence" does not really mean much of anything. It amounts to saying "somehow it happens." This leaves you with little more that faith in science. We've explained lots of really cool stuff in the past, so somehow we should be able to explain the emergence of qualitative experience. I have faith that science can, to some interesting extent, explain this emergence, but I'm proposing that the success of science in this endeavor will require at least one fundamental tweak.
I've used the analogy of "units of measure" to get at the logical outline of the problem facing science. It's just an analogy (I'm not sure if qualitative "units of measure" are literally needed at the level of physics), but if you can see the logical problem with trying to derive "kilometers" from "grams" without some bridge term that relates kilometers to grams, then you can get at least a rough idea of what I'm saying needs to happen in physics. If, for example, you want to calculate how many kilometers your car can go on 20 grams of gas, you need a grams/kilometer conversion factor. Mathematically, you are just plain stuck until you figure out this conversion factor. Also, logically, you know that you can't have a grams/kilometer conversion fact that fails to somehow reference distance.
In qualitative terms: We know that we are composed of matter that behaves in complex patterned ways according to the laws of physics, and we know that we experience subjective/qualitative phenomena, so we need a conversion factor that helps us relate processes of type X to qualitative experiences of type Q. Logically, we know that we can't have an X/Q conversion factor that fails to reference Q-like properties. The theoretical elements and laws of current physics fail to reference anything "Q-like", thus we know that, to find an X/Q conversion factor, we need to find some way to incorporate some "Q-like" factors.
One possibility is brute-fact high-level neural correlations. Maybe that's the best we can rationally hope for. I'm trying to figure out if, maybe, we can rationally hope for a bit more than this. Perhaps with the right proto-Q-like factors at the level of physics, we can model the emergence of Q at higher levels - rather than simply proposing brute-fact correlations at the neural level.
If you can find any factors in fundamental physics (electron volts, units of mass, etc.) that could, in principle, play the role of proto-qualitative factors out of which the qualitative aspects of experience could emerge, then I will gladly eat my hat - or, at least, lick it or nibble on it.
Last edited by Gaylenwoof; 01-12-2015 at 09:28 AM..
I'm going to jump in on this and say I agree - although, of course, I've also been saying that our current concept of matter is inadequate insofar as it provides no logical basis for explaining how qualitative experience emerges from the structure/activities of matter. Current models of physics can account for recursive/self-referencing material activity - which can seemingly account for the sorts of physical behavior that we attribute to conscious creatures - but there is still nothing in our current concept of matter, or the laws of physics, that gives any clue as to why material activity of this sort (or any sort) should qualitatively experience anything like pain or the redness of a sunset. We can account for wincing and rubbing a wound and, generally, engaging in "pain behavior," but no one has yet shown any logical basis in current physics for explaining why material systems engaged in this type of behavior should subjectively feel pain. The reality of self-reference gives us a good basis for explaining how there can be unique perspectives (i.e., subjectivity), but we still have no basis for explaining the qualitative feel of being a system adopting a unique perspective.
The best anyone has offered is a brute-fact correlation: "Material activity of type P just is pain, and that's the end of the story." That might be all that science can ever do, but I think that, with the right kind of conceptual breakthrough, science might be able to do better. (I acknowledge that, no matter how good your explanation is, someone can always ask another "why" question, and, realistically, we need to accept "brute facts" at some point, but I propose that physics has not yet hit the explanatory rock bottom with regards to qualia. With a better conception of matter, we can go at least one step further.)
Now, concerning the recent exchange between Mystic and monumentus, I'm wondering if there is any chance that we could all agree on specifically what we mean by "robot" and "machine." I'd like to propose the following:
Robot = An algorithmically-driven computer housed in a mechanical body that is capable of interacting with the world (e.g., wash dishes, play the role of a butler, engage in verbal conversation, etc.)
Machine = A physical system engineered and built by intelligent creatures (i.e., people).
By these definitions, all robots are machines but, of course, not all machines are robots. I've left the term "machine" extremely wide-open. A machine could be a self-replicating, self-maintaining, self-referencing system that can grow/evolve beyond whatever initial "programming" was build into it. A machine, for example, designed in accordance with the principles of a "neural network" could learn behaviors and display types of goal-seeking behaviors that the engineers/builders could never have predicted.
With these definitions, I'd say that a robot could not be conscious, but some types of machines, perhaps, could be. (Commander Data on Star Trek would be a sci-fi example of this type of machine.)
A key sticky point will be the concept of "algorithmic". An algorithm is a step-by-step procedure for calculating some more or less determinate output from a set of initial conditions/inputs. I say "more or less" determinate because an algorithm can incorporate a random number generator (RNG). Your laptop has a pseudo-RNG, but presumably there are, or could be, machines with quantum-RNGs that would be true RNGs. In fact, it would probably be sufficient for an RNG hooked to the internet to check some indexical fact like "current temperature at a given location" in order to count as truly random. The key idea is that a machine with an RNG can mimic the unpredictable nature of life-like behavior without necessarily having any actual subjective/qualitative sensations, desires, etc. A machine with an RNG can seem to "consciously choose" options when, in fact, it has no inner feelings about its choices one way or another.
I say that sentient creatures (or sentient machines - if we succeed in building them) will not be algorithmically-driven. Sentient machines will need to engage in "agent causation" - which is a concept that current physics is not able to handle. This "agent causation" is the reason why I say that current physics will eventually fail to reductively explain the activity of micro-elements in a sentient brain. In effect, I'm predicting that neurons (or, perhaps, sub-neural components?) will manifest patterns that ultimately violate quantum statistics. One possible way in which this might work: "Local hidden variables" would violate Bells Theorem, but quantum systems don't violate Bell's theorem, thus we conclude that there are generally no local hidden variables at the quantum level. But we currently have absolutely zero data showing whether or not quantum systems in living brains violate Bell's theorem. My prediction is that if/when we ever have such data, we will discover that Bell's theorem is violated in these systems. The source of this violation will be what I'm calling "agent causation."
Algorithmically-driven machines could, perhaps, mimic the behaviors of sentient agents, but mimicking sentient behavior, or modeling sentient behavior, is not the same as being sentient. According to my proposal, being sentient requires violations of Bells theorem in physical systems where, according to current theory, there should be no such violations.
no Christ ... for the 1000's time and why I dislike philosophy so.
You are an input devices and a processor.
"awareness" is not 'special". it may be fundamental, that is true .... So ... machine inputs have been made to "resonate" with many other fields more completely/better than human input devices. (like e.m. fields) We will probably make a processor "better" than the human brain thus it will "resonate" more completely than us too with this qualia thing.
if a machine or robot (you need definitions because you are stuck in philosophy) mimics you in every manor it is you. end of discussion.
If it mimics awareness in every way, it is aware. nothing you can or have said changes that.
there are "levels" to awareness. peroid
yes, nobody knows what "matter" "is" thus the descriptions are incomplete.
now what?
making stuff up is unacceptable and for some people doing it, it is dishonest.
although I must say you don't seem to dishonest it's just that I don't think you grasp what is going on past reading someone else's stuff. And many times that is good because where I see no possible way you may breakthrough. I don't think so because this quala thing is too far off base. But maybe parts will be right.
I am sorry but I must repeat it because people that don't know may think its on equal terms with what is known. and it is not.
Frankly I am bemused by the fact that you can read my detailed posts and NOT understand them . . . or pretend you do not. Your continued pretense that you do not know or understand what the difference is between living and non-living matter is specious. Non-living material can NOT in any way experience what living matter can. Any arguments or pretense to not know why there is a difference borders on trolling. It is a transparent attempt to place the burden of explaining what life IS on your opponents. When you can tell me what is missing from a rock that makes it not be alive . . . I will entertain your pretended ignorance, taunts and sophistry.
not totaly true. The problem is that you have parts of it right so it gets down to what we anchor our belief in.
A rock does interact with its surroundings and react to changes in them. Just not like you do. But it uses the exact same "rules" as you do. Only the arrangement of the materials different. Well, a "dirty" coal rock has many of the same materials not granite, but you get it.
It comes down to how you define a 'thing". If it is just by the "rules" then everything looks the same. If it is just by the "matter" then everything looks different. I kinda use both and lean one way or another depending on the convo. Just like the periodic table does.
the periodic table most certainly is not complete ... it is just all we have right now to use as a base to make predictions..
I will stick to the meanings (that is I will steal them) that have been used by myself and others in the thread - which is to describe the brain as a biological machine that produces consciousness - and a "robot" or "AI" as a non-biological machine constructed to achieve the same thing.
Ok, I can live with that. But then I'm not to sure what you mean by a "machine". Would a bunch of water molecules floating down a river constitute a machine? Also if you use the term 'machine' in such a way that the brain is a machine, then obviously - using your definition - Mystic won't object to saying that some machines are conscious. But, realistically, I would expect Mystic to insist that "The brain is not a machine." But this just means that there is some other concept that we need to define more clearly in order to isolate the source of the disagreement between you and Mystic. What exactly is it that makes the brain a "machine" or "not a machine"? Without more clarity on the meaning of 'machine' there is no useful communication going on when one says the brain is or is not a machine.
And, of course, if you define 'robot' as a non-biological machine designed to achieve consciousness, then you've defined your conclusion into the terms of your argument. If you define robots as conscious, then of course you can conclude that robots are conscious. Now, to be fair, you did say "designed to achieve consciousness" which seems to leave room for failure of design, but I don't think this is a realistic definition of 'robot' because most of the people who design robots are not designing them "to be conscious." Most robotic engineers would say they don't have a clue how to design a conscious machine. They are simply trying to design a machine that displays intelligent - or, in some cases merely useful - behavior.
I'd like to offer an amendment to my definition of 'machine.'
Machine = A physical system that is designed by people, or "as-if-designed" by natural selection, in such a way as to exhibit intelligent/purposeful (or, at least, useful) behaviors or functions.
Here I will expect Mystic to say that natural selection (or self-organization, etc.) are not sufficient for the creation of intelligent behavior. I'm pretty sure he will say that prior intelligence is required in order to achieve the creation of an intelligent system.
Last edited by Gaylenwoof; 01-12-2015 at 11:05 AM..
The problem is that, according to materialism, the qualia dealt with at the level of neuroscience and psychology ought to emerge from the elements and principles of fundamental physics. Until we can explain this emergence, the term "emergence" does not really mean much of anything. It amounts to saying "somehow it happens." This leaves you with little more that faith in science. We've explained lots of really cool stuff in the past, so somehow we should be able to explain the emergence of qualitative experience. I have faith that science can, to some interesting extent, explain this emergence, but I'm proposing that the success of science in this endeavor will require at least one fundamental tweak.
It may or may not be just one, though, and it might or might not be fundamental. As you've said, this claim is just a placeholder for a whatever thingy of some undefined sort. No need to assume anything about it at this point, other than we don't have the complete picture yet.
Quote:
I've used the analogy of "units of measure" to get at the logical outline of the problem facing science.
What fundamental units let us logically derive running from fundamental physics? A justice system? Weather? If you can't list them, then we'll have to follow your approach and add one or more fundamental tweaks for each of those.
Or we could realize the inherent weakness in this whole approach of trying to figure out how basic physics is going to "logically" imply the behavior of complex systems before we even understand what science is at work in those systems.
Can you point to an example where your approach has ever been useful in the past? I've pointed out several examples of times in the past where the idea of adding some fundamental something or other to science to explain something that feels special to us has been an abject failure. Is there any reason at all to think your approach isn't a repeat of people who insisted that organic chemistry as we know it today is impossible?
Quote:
In qualitative terms: We know that we are composed of matter that behaves in complex patterned ways according to the laws of physics, and we know that we experience subjective/qualitative phenomena, so we need a conversion factor that helps us relate processes of type X to qualitative experiences of type Q.
Yes, people will develop models of brain function consistent with known laws of physics and this whole problem will go away. Or maybe they won't and something will have to change. But pretending you know what the answer is going to be just because some philosophers can play cute word games is not going to get you very far.
No need to assume anything about it at this point, other than we don't have the complete picture yet.
The first step toward solving a problem is to recognize that there is a problem. Noticing logical gaps or inconsistencies in theories is generally a good clue to the existence of a problem (even if a lot of nay-sayers insist that they don't see a gap or inconsistency). Once the problem is recognized, it can be helpful to seek careful definitions and lay out the logical structures of various options to see if you can narrow down the range of possibilities to some of the most likely candidates. A lot of people have said that they have no clue what an answer to the hard problem could possibly look like. I'm offering some tentative clues. My proposals are vague, I admit, but I think they are as-good or better-than other proposals that have been made.
My suggestion, in case you've missed it, is to analyze the self-referencing functions of the brain and look for anomalies in the causal sequences of events. Or, from the other direction: Analyze the neural correlates of qualitative reports from subjects in neuroscience experiments. If my proposal is correct, each time you are able to isolate a neural correlate of a qualitative report, you will be able to identify a self-referencing function. This is just the first tentative step in what will probably be a long and complicated journey, but it's hard to get anywhere without taking a first step, and it is helpful if that first step is at least roughly in the right general direction.
What fundamental units let us logically derive running from fundamental physics? A justice system? Weather?
Running: Mass, distance, ergs, etc. I don't see any reason to list all of them or tell the entire story in detail because there is no fundamental mystery here. Conceptually, there is no reason that a collection of atoms, as described by current physics, could not model a human-shaped mass moving through space at 15 mile/hr., etc. If you want to explain the feeling of wanting to run in order to lose weight, or win a gold metal, then according to my proposal some of the fundamental units are currently missing.
A justice system: To the extent that you focus on objective behaviors (human lips moving, a guy holding up a bloody glove in front of a bunch of other people, etc.), there is no more mystery here than there is in the case of running. Basically all you need are units of mass, trajectories of motion, principle of self-organization, etc. But if you want to explain the specific behavior of specific individuals in terms of qualitative sensations, emotions, etc. of the victims who are hurt or pissed off, or the zeal of the prosecutor who is striving to keep his conviction rate above 80%, or what it's like to think about think about the definition of 'fairness', or what it is like to feel that something is or is not fair, then you are talking about precisely the realm where I say that current physics falls short.
Weather: Same as running. A collection of asymmetrically interconnected units (which are modeled by current physics in terms of atomic structures, electron shells, gravity, etc.) are sufficient, in principle, to model whether patterns. The principles of dynamical systems will be especially helpful here.
The basic idea is that so long as you are talking about descriptions of objective behaviors, the current units are sufficient, in principle, to explain all sorts of complex patterns of behavior.
Ok, I can live with that. But then I'm not to sure what you mean by a "machine".
I do apologise as I thought it was self evident from what I already said. By "machine" I mean any artificial construct produced by humans with the intention of producing consciousness. As such mystic has continued to fail to give us any argument as to why a non-biological human artificial construct designed to achieve the same things the biological machine in our head does - can or should be less able to attain consciousness as we are.
I see no point in obsessing over the word "machine". Use the word "thing" for all I care. The "thing" in your head appears to be a biological "thing" which produced consciousness. I see no argument on this thread to suggest a non-biological "thing" designed to do the same thing - should be expected to be ever unable to achieve it.
Please register to post and access all features of our very popular forum. It is free and quick. Over $68,000 in prizes has already been given out to active posters on our forum. Additional giveaways are planned.
Detailed information about all U.S. cities, counties, and zip codes on our site: City-data.com.
Please register to participate in our discussions with 2 million other members - it's free and quick! Some forums can only be seen by registered members. After you create your account, you'll be able to customize options and access all our 15,000 new posts/day with fewer ads.
