I have to say that I'm shocked that some of the early comments here complete dismiss what I consider to be the greatest and possibly most important of mysteries we are confronted with. If not great and important, at least the subject of who are "you" is a deeply and profoundly intimate question to ponder.
I distinctly remember first learning some basic laws of Newtonian physics in high school, and leaving school that day dumbfounded. If all particles behave in a deterministic fashion, how is it that these particles bouncing around in my head can lead to the perception of free will and agency? It was a question that would pursue off an on for most of the rest of my life. Arriving at college, I dove into psychology with a passion, hoping that understanding the biomechanics of the brain would give me some clues. I turned to computer science and was instantly drawn to AI. I read everything I could get my hands on on the subject, with Daniel Dennett and Douglas Hofstadter being favorites of course (I still revisit Godel Escher Bach from time to time). I even experimented with various psychedelics looking for that elusive insight that would explain things.
Now I'm in my 40's and of course I still don't have "the answer". I'm not as consumed by the question anymore, focusing instead on just enjoying my days on earth here with my family and friends.
But how on earth anyone can belittle the question of consciousness blows my mind, so to speak. :)
You'll have to be a bit more specific in terms of what exactly people are saying that "belittles the question of consciousness", I haven't see anything in this thread that I took as such.
As to your other statement, Newtonian physics was wrong. It broke down under certain circumstances which are not (largely) encountered under daily live, so our intuition was useless to deceive the faults in it's laws, yet this did not stop people of the time from thinking it was the end-all-be-all set of rules that described physical reality. Newtonian physics is certainly not the branch of physical that would help you understanding free will and agency.
Then Einstein came around, and amended (some of) Newtons problems. Then Quantum Mechanics came around and filled in some gaps in Relativity. QM and Relativity still don't play nice together so in the future, someone else will most likely make further amendments and get us closer to the answer.
But anyways, what I'm getting at, is there seems to be a reaction, perhaps innate to the human psyche to ... hmm... blow a fuse, when faced with an problem perceived to be intellectually intractable. But progress comes, inevitably. Sometimes slowly, and sometimes quick, but always inevitable.
My own thoughts on consciousness is that, of course its generated by the brain, of course it is caused by things that remain solely in the realm of physical reality. Of course (assuming we don't blow ourselves up or the like, as this may take quite some time) we'll not only understand what's going on in the brain, but will be able to build conscious machines, machines which we will quickly find resemble us more closely than machines and at this point will see that the amazing phenomenon we call consciousness is the not the sole domain of humanity, but just the sole example(let's exclude other animals from this thought for simplicity) of such up until that point. It might take awhile though, evolution took a billions of years to get consciousness to this point, I think it's only fair to give humanity a bit more time before we call it a lost cause.
> You'll have to be a bit more specific in terms of what exactly people are saying that "belittles the question of consciousness", I haven't see anything in this thread that I took as such.
When I first opened the comments, the first few comments I saw were:
dominotw 58 minutes ago | link[-]
> i can't take anyone who uses the word "consciousness" seriously, given noone has any idea what it actually means.
Marazan 57 minutes ago | link[-]
> Predicated on the concept that the Hard Problem of Consciousness is actually a problem.
crimsonalucard 44 minutes ago | link[-]
> How can we solve a problem about a word that isn't clearly defined?
Regarding my other comments, I focused on deterministic Newtonian physics because that is what first sparked my interest in the question. Yes, since that time I have learned a bit of quantum mechanics and relativity. Around the time that I found myself trying to grasp string theory is when I gave up the physics thread (pun intended). I'm not a physicist, and frankly I'm not smart enough to fully grasp modern physics.
For what it's worth, I'm in the "emergent property" camp. I certainly don't think there is anything non-physical necessary to explain consciousness. I just think it's a fascinating topic and was shocked at the early dismissiveness I saw. Even in the original article, there's at times an undertone of "consciousness is an illusion and therefore not interesting". Just because we don't need to resort to the supernatural to explain consciousness doesn't make it less interesting. To me, that's exactly why it is interesting.
I think the derision is not from pondering the actual question of consciousness (or strong AI) but the nebulous, fuzzy definition of consciousness by some in the philosophy camp. I think that better work is being done by those focusing on well-defined problems and holding off on the "hard" question until we are in a better position.
Also, this article is based heavily on Koch's pansychism which I find very unconvincing. There is talk about consciousness relating somehow to mental simulations and I believe this is where the answer lies (and an explanation that was not included in the article.)
> fuzzy definition of consciousness by some in the philosophy camp
When discussing consciousness I like to focus on the feeling of pain, which is much more clearly understood by people. So the question is, why do I feel pain when my brain is in a certain physical state?
I think that's pretty easy to explain. The change of state in your brain modulates the thing that it both creates and controls (consciousness) state to "feel" pain.
Consciousness is a emergent property of the incredible complex organ we call a brain. At first it may seem too odd to think that something as incredibly surreal and deep as consciousness can spring forth some arrangement of atoms, atoms identical to those make up the inanimate objects around us, it because less weird when you think about a car just being made up of nuts, bolts, and a bunch of metal, or a space shuttle out of ceramic, composites and what have you. Computers are just made up of transistors, capacitors and the like, and have tremendous capabilities.
Humans are just the most extreme example we have of this phenomenon of complexity through composition, but that makes a certain kind of sense as humans are the product of billions of years of evolution. We've only been seriously studying consciousness for a good hundred years or so (people have thought about it for quite some time, but I think it's fair to say no clear progress was made on the subject until the 20th century).
> I think that's pretty easy to explain. The change of state in your brain modulates the thing that it both creates and controls (consciousness) state to "feel" pain.
It's certainly not easy to explain. I have no idea what that means. You're basically saying that change of physical state causes pain. Why? Is that some physical law I haven't heard of?
> Consciousness is a emergent property of the incredible complex organ we call a brain. At first it may seem too odd to think that something as incredibly surreal and deep as consciousness can spring forth some arrangement of atoms, atoms identical to those make up the inanimate objects around us, it because less weird when you think about a car just being made up of nuts, bolts, and a bunch of metal, or a space shuttle out of ceramic, composites and what have you. Computers are just made up of transistors, capacitors and the like, and have tremendous capabilities.
The big difference here is that computers or cars are physical concepts, the definition of a car could be that it's a certain arangement of atoms.
I think it is an emergent property too but it turns out that philosophers use of the word emergent is not the same as other scientists use of the word. Philosophers use emergent to mean irreducible which isn't very useful.
> Of course (assuming we don't blow ourselves up or the like, as this may take quite some time) we'll not only understand what's going on in the brain, but will be able to build conscious machines, machines which we will quickly find resemble us more closely than machines and at this point will see that the amazing phenomenon we call consciousness is the not the sole domain of humanity, but just the sole example(let's exclude other animals from this thought for simplicity) of such up until that point.
Why of course? I completely disagree, I've come to the conclusion that consciousness is not a normal physical phenomenon. Let's say that we can create consciousness on a computer. We can simulate a computer with pen and paper. Can we create pain just by scribbling something on paper? We can also simulate a computer by moving sand grains. And there's practically an infinite way of encoding the computer state into configuration of sand grains. So when we just toss sand grains around, are we creating 100% real pain?
What is a "normal physical phenomenon" and what is a "non-normal physical phenomenon"? It would seem to me that something is either physical or metaphysical.
Not once in the entire history of human inquiry has something turned out to be provably "because magic". Lightning was not Zeus having a fit, plagues were not the vengeance of the gods, birth defects are not the disdain of our ancestors, and humanity did not get created from clay.
The scenario you put forward is the "Chinese Room" argument, and it's incredibly flawed but is a great example of flawed philosophical thinking regarding consciousness. It's main problem is that it was directed at Alan Turing's "Turing Test" which was also flawed. The Turing test boils down to, "If a machine can make you think it's a human, it's true artificial intelligence". The Chinese Room argument rebutted and put forward the idea that if you have some in a rule with a algorithm to associate symbols from one language to another(lets say english to chinese), and another room with somebody who actually speaks Chinese and English, externally they would be indistinguishable but these is clearly a difference between the two. I, for one, totally agree with this sentiment, but it doesn't really speak to consciousness, but just how bad the Turing Test was as defining artificial intelligence. The problem is that the philosophers then applied to Chinese Room Argument to not just the very narrow (and completely wrong) definition of AI outlined by the Turing Test, not just to artificial intelligence (which would be incorrect), but consciousness in general (which is a horrendous error), no matter how defined, it's just a plain logical incongruence.
You ask questions such as "are we creating 100% real pain", while not describing exactly what IS pain? I think it's pretty hard to say that thing A(let's say an artificial intelligence living in some super computer cluster), is not equal to B(A fully conscious entity), when we can not yet even define what B is. While I think it's pretty safe to say that such a AI would never know what it is like to be you, and you it, that hardly means it's not (fully, real, actual, insert whatever word you want to call it here) consciousness. Once or if total mastery of artificial is achieved it seems reasonable that you would be able to create a artificial intelligence with capabilities very similar to the humans. With the same sensors, neuronal network capabilities, even an artificial human body and at this point the differences between the way a human experiences reality this theoretical human-like robot experiences reality would more resemble the differences between the way I perceive reality/exercise consciousness than the differences between a pile-of-sand computer and a human.
> It would seem to me that something is either physical or metaphysical.
Yes, I meant that it's a metaphysical thing.
> Not once in the entire history of human inquiry has something turned out to be provably "because magic". Lightning was not Zeus having a fit, plagues were not the vengeance of the gods, birth defects are not the disdain of our ancestors, and humanity did not get created from clay.
Which is why people find it hard to accept the idea of consciousness not being physical.
> The scenario you put forward is the "Chinese Room" argument
Perhaps there are some similarities, but it's a different argument.
> You ask questions such as "are we creating 100% real pain", while not describing exactly what IS pain?
Good point. How would you define consciousness? I define it simply as my current feelings, perceptions, thoughts, etc. By this definition, only one consciousness exists. Consciousness is the only thing one can be sure about. When I feel pain, it's a non-disputable fact that I have this unpleasant feeling. But I can't be sure about anything else, for example the physical laws (they can change tomorrow). Or you can wake up from a Matrix - actually absolutely anything can happen.
How would I define pain – it's sort of a axiom (not sure how to express myself here) and it's subjective. Objective definition that would allow you to test if someone is feeling pain doesn't exist (or it would describe something else than the feeling of pain). Similarly, physics is simply unable to define consciusnes, it's out of the scope of physics.
For one, quantum physics are not deterministic. The world you live in, and you, are predictable, but not quite deterministic.
I personally think that if we create robots with a sufficient understanding of the world, they will necessarily be self-aware and "conscious". Those are just emergent properties of a sufficient understanding of the world, it seems to me. You might think that's not true. Asimov's robots are purely "logical" (but not reasonable) and don't seem conscious, they don't have free will... But robots like that might never exist. They'd be too rigid to deal with the complexities of this world in any kind of useful way. For robots to be sufficiently capable, they'll need a more fluid, more "natural" and less "predictable" kind of intelligence.
I agree, but you don't need thermal noise to simulate the effects quantum randomness. You can use actual quantum randomness! The number of photons absorbed in a photodetector is determined by quantum randomness, because some of them are absorbed and some of them pass trough. IIRC a camera with a CCD is enough to measure individual photons.
(Nevertheless, in my opinion neither of this is necessary to simulate consciousness, a good PRNG would be enough, but we should wait a few years until this is confirmed.)
I don't know if this helps the argument for free will, but physics still has randomness. Radiation, for example, is to my knowledge completely random. So at least from a mile-high view, you can't actually perfectly predict the outcome of everything, even if you could model perfectly.
The universe is not deterministic. However, this was used to say that "there is still free will!!!" (e.g. Penrose). So, there is "hard" and "soft" determinism—the universe may not be determinstic, but that doesn't mean we're any less determined by physical processes.
But quantum uncertainty is not the same as me having free will, because my will does not control the quantum uncertainty. Quantum noise may influence the atoms that control the neurons that control my brain, but that just makes me not a perfectly-determined machine. There's noise at the bottom layer of the machine, but that noise is not the same as me having free will.
It's at least a step away from "everything can be predetermined with physics". If I could perfectly predict your actions (with a physics model), most people agree free will is impossible.
It doesn't mean free will exists, of course. It's just less improbable?
>It's at least a step away from "everything can be predetermined with physics".
You'll never get away from that. We are material beings, made out of the same material as everything else in the universe and hence subject to the same fundamentals laws.
>If I could perfectly predict your actions (with a physics model), most people agree free will is impossible.
If that's the case, there should not be a problem. Even in a fully deterministic universe we'll never be able to run a simulation to predict the actions for anyone anyway. You're a product of not only the actions of the trillions of particles that make up you, but also of the countless of particles in the environment over your entire life-span, that of your parents, your grandparents, all the way to the big bang. We'll never be able to simulate the trillions of particles that make up any specific body, much less that body's environment using fundamental equations. So if predictability of particles is how you measure free will, then we have free will.
Things get muddled a little because we may be able to get a 'good enough' prediction of someone's behaviour just by analyzing his/her neurons with a fMRI or something. This is akin to the difference between knowing tomorrow's weather EXACTLY by simulating every particle of the atmosphere vs. knowing the weather tomorrow to some degree of confidence by analyzing higher-order patterns like air pressure, cloud coverage, temperature etc.
So your definition is too weak. Most people would say we don't have free will if we live in a universe with deterministic laws. Whether anyone can actually use those laws to predict your actions is not that relevant.
>It doesn't mean free will exists, of course. It's just less improbable?
Not really. Think about, let's say every decision you made was determined by a (quantum, full random, totally unpredictable) coin toss. That doesn't get you one step closer towards a libertarian-type of free will. Sure your actions are, in principle, much less, predictable to those around you, but that's not the yard-stick we use to measure free will by anyway. Your actions are already unpredictable in a deterministic universe anyway. You merely traded being a slave to deterministic laws of physics to being a slave of random coin toss.
Second, it's an open question how much quantum randomness actually affects your day-to-day actions. Quantum effects are only pronounced on very small scales. At the scale of groups of atoms, molecules and cells, Newtonian physics, chemistry, and biology, and neurology can be substituted without loss of precision.
1. The fact that we cannot scientifically determine how something functions does not necessary imply a non-physical cause.
2. Thought experiments are inherently limited to human conceptualizations which--while based in reality--have no authority over reality. Here's a famous thought experiment: Descartes believed that clear and distinct perceptions can be certain. I can think about a horse and a dog separately, so they most be different things, right? Well, I can also think about time and space as distinct entities but guess what they're not. Thought experiments are entirely limited by human intuition and human intuition is not guaranteed to properly reflect reality. The fact that you can imagine a zombie of yourself has no bearing on reality. I can imagine traveling faster than the speed of light but that doesn't make it possible.
3. When systems reach sufficient enough complexity it is impossible to understand them by merely analyzing the individual components and understanding the rules that govern them. Even if I had full knowledge of the human brain and how it worked, there would be no way for me to understand another person's first person perspective because there would be no way for me to simulate his brain. Mary's Room, qualia, and Mooreland's argument from first person perspective are all dependent upon the assumption that if you can understand the components of such a system, you ought be able to understand the whole. This premise is false.
4. There are plenty of "nonphysical" entities that we presume to have an entirely physical basis. Take the operating system on your computer. Is your OS a physical object? A physical property? It's real. You can interact with it. But if you pry about your hard drive you will not find a little desktop with folders and icons. Even if you could read the one's and zero's off the magnetic disk that wouldn't be considered your operating system. It only exists when all the pieces are connected and running and has no value (or existence) without such animation. Nonphysical entity. Physical cause.
Why do you equate the Hard Problem with dualism? I've met Chalmers, and read much of his oeuvre, and he's most emphatically not a dualist. No-one in the field takes "strong" (or even most varieties of "weak") dualism remotely seriously any more. Frankly, I feel like you're straw-manning a bit here, by attacking a position that's not even under discussion.
Where did I equate the Hard Problem with dualism? I'm contending with a view that this article surveyed. Namely,
"Nonetheless, just occasionally, science has dropped tantalising hints that this spooky extra ingredient might be real."
Furthermore, Point #2 attacks the use of thought experiments as a means of understand the Hard Problem. Point #3 relates to why the Hard Problem is so hard. Point #4 gives an analogy supporting Dennett's position.
As far as I understand it, one of the main things that makes the Hard Problem so hard is "epistemic asymmetry": the only consciousness you can even know exists is your own. That's the point of p-zombies, not to be a clever thought experiment (about which you're mostly right, btw), but to very pointedly illustrate the very real, and very frustrating fact that (thus far) there is no proof whatsoever that the notional "lights are on" anywhere else in the universe but in one's own skull.
As for point #4, and Dennett, I'm going to have to beg off discussing, because I don't want to have to resort to that kind of language before lunch. (Read: really, really not a fan.)
You don't know your own consciousness exists. You can just be experiencing something that is not consciousness but that you think is. That's the ludicrous banality of the Hard Problem.
You can just be experiencing something that is not consciousness...
That's what consciousness is: to borrow Thomas Nagel's phrase (which Chalmers also uses frequently), "the thing that it feels like" to be something. By every reasonable definition I've ever heard — and I've studied and read about this stuff for decades — that's pretty much a constant: consciousness is the experiential part of ... you know, experience. Yes, it's entirely possible — even common — to be mistaken or misled about the content of your experience, but not the fact that you are experiencing.
In a discussion already absurdly full of sophistry, your comment should win a very large, gaudy prize.
> When systems reach sufficient enough complexity it is impossible to understand them by merely analyzing the individual components and understanding the rules that govern them.
Instead of "impossible" I'd rather use "unhelpful" or "pointless". In case of OS (or wind for a simpler example) we are focusing on the observable results of many small interactions. Most of the time looking at these "phenomena" from the higher level is sufficient for us - there's no point in analysing single atoms or 0's and 1's. To some extent it maight be similar in case of consciousness. One day we will be able to precisely monitor all chemical and electrical activity in neural network but it won't be necessary to make suffciently accurate predictions etc.
The Hard Problem of Consciousness(tm) can't be answered because it isn't a question at all; it's just a confusion. People who like to worship mystery are attracted to it, because it sucks to have your sacred mystery pulled out from under you by an explanation.
Hand-waving and dismissing something as a confused non-question isn't an explanation. People who can't handle the state of uncertainty are attracted to it, because its uncomfortable to find your premises aren't grounded in certainty.
Then explain how it is that meat can feel. There is absolutely nothing inherent in the nature of meat, as far as we can quantify, that entails feeling.
So, please, yank my sacred mystery out from beneath me: what's the explanation?
> There is absolutely nothing inherent in the nature of meat...
This is like declaring there is absolutely nothing inherent in the nature of matter to have temperature. Anyone making such a declaration would be looked at strangely if they wanted to enter a discussion on the nature of temperature.
Declaring from the outset that there is anything particularly weird about biological systems being conscious, when one of the most basic, widely-accepted facts about biological systems is that they can be conscious, is not a good move. It sets up the problem as a Big Mystery, which is what the OP is pointing out.
There are some "problems" that many people are never going to accept the solution to. Every few years someone publishes "the definitive answer" on the Mona Lisa's smile, and I still see people asking, "does nature or nurture cause behaviour?" as if it was a serious or interesting question.
Likewise, there is a very significant chunk of the population dedicated to the proposition that "consciousness is fundamentally mysterious", and one day when we have it worked out pretty well why some complex systems have this emergent property, those people will still be ending articles on the subject with coy gibberish about how "...in the end, you can't reduce the magic of conscious experience to an equation."
Sure you can. And we will, in the same way we came to understand temperature as an emergent property of systems of interacting particles in closed systems, without recourse of phlogiston or caloric or any other thermodynamic secret sauce.
That, right there, is why I reject this kind of argument. It's not parsimonious. Temperature isn't an "emergent property". It's a name we've given to a measurable phenomenon; in fact, it's the name of the measurement, itself.
Another un-parsimonious notion: "one of the most basic, widely-accepted facts about biological systems is that they can be conscious". Okay, if they can be conscious, what makes one conscious and another not? Why am I blessed with the gift of consciousness, but my cat isn't? Or if she is, why her, but not the spider in my bathroom she loves to torment? Or if it is, why not the bacterium on my toilet seat?
Drawing imaginary lines somewhere in the middle of reality and saying "things on this side of the line have squooblith, but things on that side don't" is going to require a bit of explanation as to why the line is there, and not somewhere else.
For myself, the most parsimonious answer I can find to all this is that consciousness is an intrinsic property of matter, itself. So, sure, a rock (let alone a molecule of gypsum) is going to have a markedly different kind of experience than I do, what's to say that it doesn't have some kind of experience? It also nicely eliminates this "mystery" everyone seems to be down on: what could possibly be mysterious about something that is literally everywhere?
But that's just me, and I readily admit that I'm rather anomalous, have studied this stuff for far too long, and taken a nontrivial amount of psychedelics. So, you know, grain of salt, &c.
Consider Conway's Game of Life. A few basic rules and you end up with remarkable, emergent complexity. The game of Go is another beautiful example.
The same is true of any number of phenomenon. Break them down to their basic elements and it's not clear how the emergent properties arise, and yet they do.
So is the brain just "meat"? Maybe. But that doesn't preclude incredible emergent complexity.
I wonder sometimes if we grant undue weight to "feeling" and "consciousness". A plant reacts to external stimuli in easy-to-understand ways. Sure, we are more complicated than that, but why are phototropisms just a curiosity while "feeling" is special? What if "feeling" and "consciousness" is nothing special at all, just extremely intricate and sophisticated clockwork that, like any sufficiently advanced technology, appears to be magic?
I don't agree with your conclusion - and I say this as someone who has fully accepted the ideas that human existence is a random occurrence due to the recursive emergence of complexity, that life has no intrinsic purpose and that we are on our own, so to speak.
Even imagining consciousness as an emergent property of simple brain functions doesn't fully explain it; the particular experience of various qualia, the sensation of being...these are hard to quantify.
Or perhaps you're an unconscious robot, so you'll never be able to understand. For me, it's absolutely clear and obvious that the hard problem of consciousness is real (and FTR I'm a very rational and scientifically-minded person).
> Philosophers and scientists have been at war for decades over the question of what makes human beings more than complex robots
[Citation needed]
It's the same problem that AI has. AI is whatever a computer can't do.
A few decades ago the computer couldn't play chess, so chess was a good example of AI. Once the computers could play chess and easily wind a match against a grand master, the we can see behind the curtain and understand the tricks, and now chess is only a circus trick and no real AI.
For now, computers can't simulate consciousness, so it's a mystery and a good AI research topic.
Not sure if I agree with this. My computer can't do a lot of things and only a small set of that would really be considered AI.
The chess example isn't great. We have known chess is a deterministic game and a good enough algorithm on powerful enough hardware should beat or stalemate a grandmaster. The state space is big but finite. Its just a more complex form of checkers. This seemed to be more of a research issue of a known solvable problem that needs its state space shrunk to something we can calculate than a true AI breakthrough.
I think AI like described in this article is a functioning, self-determined, self-learning, HAL-like system. This doesn't exist. It probably can with the right research and funding. Its not a problem of perception or arguing about what may or may not be AI. Although the AI-effect can be annoying.
I don't see the AI effect explaining why AI hasn't gotten farther. I think not understanding consciousness or the brain on a certain level is probably our biggest hurdle. Unlike chess, we don't know what we're trying to solve exactly.
Or can they? I suppose we'd need a quality definition of consciousness first - one that is specific, unambiguous, and able to be tested. It seems everyone has skipped over that part to prematurely claim that whatever pops into their head is a legitimate theory.
It's not surprising consciousness research is such a mess - it is a topic that is too close to how people define themselves. If somebody had a nice, simple, practical, hypothesis it would probably be attacked and buried. Civilization had a hard time adjusting to the sun being the center of the solar system, and many still haven't come to terms with the fact that we evolved from common ancestors of other living things. I doubt many people want to hear that our thoughts, our feelings and experiences can be simply modeled on a computer. Or worse, that filthy animals are conscious too.
The reason the problem is "hard" is not scientific, it is emotional.
> If somebody had a nice, simple, practical, hypothesis it would probably be attacked and buried.
The problem is that consciousness is a metaphysical concept, you can't scrutinize it with the scientific method, it requires a complete change of perspective to understand consciousness.
I'm a layman but I don't think that's fair. When I think about "good AI", that we haven't achieved yet, I think about Data from Star Trek. I'm pretty sure we'll know when we've reached that level of AI. We're not there yet.
What's they're trying to say is that the goal for what AI is supposed to be/achieve is always raising as when we actually reach the goal.
The general the goal of AI is to do things only Humans can do (in the past this included chess or holding a semi-coherent conversation). Now its more focused on human consciousness like Data from Star Trek so their statement is fair enough.
If we had that, I suspect lots of people would say "ah, well, that's just a clever bit of programming - it's not really thinking/feeling/alive/conscious/whatever, so I'd hardly call that good AI".
Also I'd bet the body helps - if you had a 'cloud' (not android) version of Data, people would be even harder to convince.
The entertainment industry has set a pretty strange standard for A.I. with the general public. What A.I. will eventually become will probably be wildly different than a limited personification like Data in Star Trek.
The first automobiles looked like horse-drawn carriages, so maybe it's not right to fault people for expecting A.I. to look like humans.
Exactly. "Philosophy is the discipline in which you kick up a lot of dust and then complain you can't see" - attributed to Leibniz. People were fussing over this issue long before the 1994 paper mentioned. Aristotle fussed over it; he was trying to pontificate on what makes humans different from animals. He decided it was the ability to do arithmetic. We now know that arithmetic is, in an absolute sense, one of the easiest things to build out of logic elements.
The big practical problem in AI is not "consciousness", but "common sense", the ability to predict what's probably going to happen from an act and deciding whether or not to do it. After we solve that problem and are able to build robots that can get through a day of diverse tasks without help, we can look at consciousness again. As a practical matter, it's useful to view "consciousness" as a background task with the job of adjusting the parameters used to formulate sequences of actions.
Humans can't figure out how consciousness works by introspection because brains don't have enough internal access to their own processing. Other methods will be necessary, and we don't have the analysis tools yet.
It is funny to watch that no one have mentioned Indian Philosophy, which could be studied in Oxford and other places. It has lots to say about the nature of consciousness.
Warning: The following may sound like mumbo-jumbo until you've finished reading.
As someone in another comment here mentioned, the last time something that was not understood was nigh declared magic and unphysical or ignored even by educated men was in the 17th century, and the phenomenon was light.
If you accept panpsychism, the idea that consciousness is not limited to brains (examples: i feel and react to pain but my dog reacts to pain as well, and even certain trees are now known to release signal chemicals into the soil which cause nearby trees to harden their bark when they are damaged), and also accept that consciousness is a field/fundamental force that couples(interacts) with the other fields, and with matter, it becomes plausible to say that an interaction between two particles will also cause an interaction with the 'consciousness field'.
Of course all of this is complete speculation because it seems to be impossible to measure the properties of said field, as the proposed interaction would be one-way (examples: when i burn my hand with scalding hot water, a circuit in my brain/body which relies on known forces and interactions causes my hand to move away, yet i feel pain).
I don't see how wave function collapse, which has frequently been mentioned in connection with this, could be related, as it's not even known what wave function collapse actually is, much like our current problem with consciousness.
My own view over the philosophy of mind is closest to the identity theory - "states and processes of the mind are identical to states and processes of the brain." [http://plato.stanford.edu/entries/mind-identity/] I also think that mind is nothing but the states and processes, so another way to put it, mind is the ongoing states and processes of the brain.
Using computer systems as an analogy, the brain is the hardware, the mind is the software code and run-time states of this system. Some of the software code was written by genetics but the majority of them are written after birth by the environment. Now there are two major differences between computer system and a human:
a: computer systems are Turing machines, humans are not. Human uses intuition to perform induction which Turing machine can not (because induction is not logic). And induction is such an important foundation of human understanding of the world that I believe a Turing machine's understanding of the world based on pure logic would be very different.
b: human minds is aware of himself, he can think about his own mind, and think about the thinking about his own mind. It is like if the software can model the software itself, without entering an endless recursive loop. That is fundamental property of consciousness.
My guess is that the fact that human mind relies on intuition rather than logic (non Turing machine) is the reason he can recursively think about himself while computer can't, after all he intuitively knows when to stop and jump out of the recursive loop.
Panpsychism... really? We're going to reject God as an explanation, because the physical universe is all there is, and then say that the universe is conscious? Seriously? That's not an explanation, that's a "just so" story. In fact, it's a very short step from there to pantheism (everything is God).
And in fact, God (the theistic version) would explain this perfectly. Drop your materialistic assumptions for a moment, and hear me out. If God exists, as a being with consciousness and personality, and if he created humans "in his own image", then we are not faced with the Hard Problem of Consciousness. Humans can then be truly conscious, and have true personality, without those things being just a configuration of atoms.
The pantheistic version of God doesn't work as well, because it's ultimately impersonal (despite the "theistic" part of the word). There's no way in pantheism for personality to be anything more than an illusion, which ironically is the same problem that materialism has with consciousness.
Koch has been pushing panpsychism in the media a lot and it is really unfortunate because it is poor science. If I were Paul Allen, I'd find somebody less publicly embarrassing to head my Brain Science Institute.
Its possible the confusion is not understanding what consciousness is.
The common accepted notion is that reality is made of small pieces of matter (ie. atoms), which in turn when then combined sufficiently gives rise to consciousness. Therefore its matter at the core, and consciousness arises from it.
Its quite possible that the order is wrong: Namely the universe may actually be composed on a foundation of consciousness.
Specifically, the idea is that consciousness is a force like all other forces (ie. magnetism, gravity, etc). However, this universal force of consciousness causes quantum-waves to collapse, hence why there is something instead of nothing.
The brain evolved in much like all your other sensory organs (eyes, ears, touch, taste) - in that it detects and can presumably interact with these naturally occurring 'force of consciousness'.
This explanation also solves many of the various quantum-physics implied paradoxes.
> However, this universal force of consciousness causes quantum-waves to collapse, hence way there is something instead of nothing.
The phenomenon described as consciousness has nothing to do with wave-function collapse, it is the act of measurement that is responsible for the collapse, whether conscious or otherwise.
I'm afraid it is. The issue stems from confusing the popular usage of "observed" (i.e. by a person) with how the term is used by quantum physicists (i.e. the interaction of the quantum particles with any other particles). That is to say: For the purposes of quantum physics, it's the hardware of the LHC that's doing the "observing," long before the scientists observe its output on the monitor.
I think you are misinterpreting what is meant by "force of consciousness", in that it doesn't necessarily imply an observer in the colloquial sense. Its not meant to imply any supernatural or otherwise mystical existence - not any more than the 'force of gravity' or whatever.
To state it another way, why does the universe exist at all? Why is it not just a jumbled mess of superposition quantum states? We know at the deepest levels there is a wave-particle paradox to most of what exist, so what causes the collapse to happen so that particles can exist?
The standard answer is that the question itself is a form of survivorship-bias, meaning because our universe is one in which superposition states collapsed into regular particles, is what allowed humans to evolve to even pose the question.
However, there is an alternative, which is that the universe exist because the 'force of consciousness' was present to cause those collapses, which then in turn provided an environment which animals could evolve and eventually detect / interact with that underlying force.
And what independently testable, otherwise non-reproducible effect on the physical world would this "underlying force" have? This explanation reeks of the same long discredited arguments used by Creationists.
What independently testable, reproducible effects can you ascribe to "hyperstring theory" or any other postulations of modern theoretical physics? Is even "big bang theory" (although imminently logical), actually testable?
It all becomes philosophy and metaphysics after a certain threshold.
This problem has been understood and resolved by a number of thinkers. The Buddha, lots of buddhist masters, then we also have people like Eckhart Tolle who clearly realise it.
The problem is that consciousness is goes beyond conventional logic and descriptions. Interestingly, if it could be grasped by conventional logic it would lose the characteristics of consciousness.
A similar (possibly related) question is why anything exists at all (including why even the concept of something existing exists!)
Unfortunately, having no intuitive insight, I cannot lay claim to an answer, but then notably other thinkers seem to agree that consciousness lies beyond what can be directly explained anyway.
There is an odd branch of logic which allows contradictory statements, possibly such a tool could be used to formally
analyse some of the properties of consciousness.
A lot of people seem to be dismissing this as "not a problem" or "not well defined". Can you expand on this so I can understand it?
It seems to me that there is a disconnect between:
- the standard scientific view of reality as consisting of fundamental particles and the laws relating them - and,
- the experience of being somebody, an experience that all of us have.
The standard view may well end up providing a good explanation of the experience (and some people seem optimistic that it will), but I don't understand how it can be dismissed. That's like saying that the EM spectrum of hydrogen cannot be explained by classical electrodynamics, and so is a non-problem.
No one seems to address the very germane question of whether consciousness is separable from language, or rather communication more broadly. Is it possible to have language, and not be conscious? Is it possible to be conscious and not have language?
Though no one takes Julian Jaynes seriously these days, but I think his ideas of the development of language preceding consciousness are as good as any other theories I've read.
This is all quite well discussed in Scientology, a subject so controversial that people rarely make the effort to dig deeper than the understanding of the subject offered by tabloid article editors and people interested in destroying the organization, while ignoring the subject.
You are a spiritual being, aware of being aware. Consciousness is the state of being aware of being aware. It can be attenuated, and it can be enhanced, by the unit of spirit itself, represented in Scientology by the mathematical symbol: theta, which is used because it represents infinity.
Anyone wanting to know more can study the subject without getting into too much controversy - all it takes is to read the books about the subject made available in any library. This is the very thing that the subject is intended to explore and define, and it is a very thorough investigation indeed - covering everything from the most basic axioms of life all the way to applying the properties of theta in the modern, present time world.
But of course, the hubris and hysteria about the organization and its activities precludes an honest look at the real, very well defined, details of the subject. Get past the hysteria, and you will find a perfectly workable, perfectly acceptable definition of consciousness. However, that's not as easy as it sounds, alas, in this modern hubristic world..
Perhaps the answer to the question of "what is consciousness" can't be reached by thinking. Maybe there's not a logical answer in the same way we find reasons to explain the world around us. That would explain why the greatest minds can't solve the mystery.
Either that, or our species is just not evolved enough to comprehend the problem.
It was a puzzle so bewildering that, in the months after his talk, people started dignifying it with capital letters – the Hard Problem of Consciousness – and it’s this: why on earth should all those complicated brain processes feel like anything from the inside? Why aren’t we just brilliant robots, capable of retaining information, of responding to noises and smells and hot saucepans, but dark inside, lacking an inner life? And how does the brain manage it? How could the 1.4kg lump of moist, pinkish-beige tissue inside your skull give rise to something as mysterious as the experience of being that pinkish-beige lump, and the body to which it is attached?
To analyse its contents, I'll ask this: what's to say that a computer doesn't experience some very primitive form of consciousness? If we unplugged everything except the power cord, but left a complicated simulation running, it would still have something like "a rich inner life." Its peripherals and sensors give it a sense of a body and, with abstract drivers, a degree of conceptual separation from said hardware. Doubly so in the case of virtual machines. After all, we can't "truly" experience the same thing that a computer might from the inside, so who are we to doubt "computer consciousness?"
If the proponents of "The Hard Problem of Consciousness" can't give a quantified explanation of how to distinguish a theoretical computer consciousness from a human one, that raises the question of whether the problem actually exists.
For my part, I don't believe in consciousness as a concrete thing, only a label we use to group together quite a few disparate systems and phenomenon. It's the same way that I don't believe in "Ruby" as a concrete thing, but only in the unit tests, the sample code, the docs, and the thoughts in Matz's head that we subconsciously conflate.
Is there anything for which you cannot ask a similar question of the form: Why does xyz exist? Why are things not just [this simpler thing]? For those types of questions, I imagine one would answer: because if it were not so then things would be nonsense.
If this question is asked of consciousness without saying what consciousness is doing such that if you took it away then you would have nonsense, then I do not see how anyone can answer it.
The way the question is framed up, consciousness accomplishes nothing as far as survival goes. "evolution could have produced zombies instead of conscious creatures – and it didn’t" Zombie seems to be defined as: phenomena that you think has consciousness but does not.
If consciousness has been defined as some phenomena where you have an intuitive understanding of what is being communicated by that word "consciousness" but also it (consciousness) does not do anything, then how can anyone explain why consciousness should be? Anything you can think of as a function of consciousness, can then be hand-waved as able to be accomplished by zombies.
Here's a more concrete and less "woo woo" (although only less, not necessarily "woo woo" free): Suppose for the moment that consciousness exists on some sort of scale (i.e., I'm not presuming binary have/have not, and I'm not really presuming total ordering of "quality" either). Humans have "more" and a dog has "less" and a fly has "hardly any". We are not too far out into the water, here, really. There is some sort of meaningful something that humans have lots of, dogs have less of, and flies have hardly any, even from a strictly physical perspective.
Now, suppose we assume that the end of the article is correct and "consciousness" is really just a form of integrated information processing, and their device for measuring the interconnected nature of a given brain is measuring something meaningful.
(And let me take a moment to observe that when I say "assume", I'm serious. I'm not asking for agreement, I'm saying, work with me and assume for the sake of a discussion.)
Now, we are still not too far out into the waters here. We have a device that is measuring something real, and fairly non-mysterious. It would not be hard to create an equivalent for a computer system by considering the system as a set of graph nodes, then probing the connectivity of the system. We're still standing on fairly concrete ground.
Now, consider that in the near future we will be able to build computer systems that by any basic connectivity metric (average number of neighbors, various stats of connectivity shape, etc.) would match even human brains, but can be demonstrably doing something very simple like just copying data from here to there, something that is clearly not "conscious" behavior.
Characterize the difference between my hypothetical computer system that matches the human brain on $RELEVANT_CONNECTIVITY_METRIC and the actual human brain. In theory, as you refine the connectivity metric you should also be describing to me how to build a human-equivalent brain, too, or at least telling me exactly how I'm falling short.
Note this is a sort of mathematical argument, where I'm asking you to bring your own $CONNECTIVITY_METRIC to the party. It seems pretty clear that it can't be as simple as whether or not it is or is not of a certain size and happens to follow a power law in the connectivity graph, because it's really easy to produce a program that would have that level of connectivity and still not be conscious. (A naively-written non-trivial program naturally has a power-law distribution to it.)
This phrases the problem in a fairly practical way ("how do I build a human-class AI?"), demonstrates the difficulty of answering it even given a powerful and potentially incorrect assumption, and is phrased virtually entirely in physical terms such that a sufficiently concrete and accurate answer would very likely lead to the ability to build AI. And we have no clue about that $CONNECTIVITY_METRIC. All we've got is some people asserting that the answer lies in this direction and others asserting it doesn't and very little hard data to help us answer that question.
And I'd also observe that I'm not even worried about the answer, this generalizes trivially to the "set" of conscious behaviors, and also your choice of whatever test you like to apply for consciousness up to and including "I'll know it when I see it", since right now we can't answer any of these variants.
To expand on my earlier dismissal. The hard problem states
1 Consciousness has a special property
2 Physical process are insufficient to have the special property
3 Zomg what is consciousness
4 We are totally not Dualists, honest
The answer is either
A) Conciousness is the result of a physical process
Or
B) You are a Dualist, stop lying to yourself.
I strongly encourage people to read the book The Origin of Consciousness in the Breakdown of the Bicameral Mind by Julian Jaynes. He makes some rather solid and defensible statements about consciousness (included falsifiable predictions that were later validated with brain imaging), and then makes some rather fantastical inferences. It's thought-provoking.
One of the very first things he does in that book is provide a definition of consciousness that he uses throughout the book. It's a more concrete definition than I've seen in most other places, and in part because it's very small. Like most others, I tend to roll my eyes a bit when philosophers start talking about how consciousness is the act of feeling and experiencing, and is the core of being, and is the sum totality of human experience, and start getting all mystic about it. Jaynes takes the opposite approach: it's not all and everything of human experience, it's exactly one specific thing about human experience.
Specifically, introspection. The ability to not just observe the real world, but to create a simulated world in our minds and observe that. Replay memories, or test hypothetical present or future situations. Anything else unique about the human mind is taken to be a consequence of consciousness, rather than a part of it.
Another big thing he does is show that activities that we normally think of as conscious and rational/intelligent are actually not taking place in the consciousness (these are backed by experimental evidence). For example, chess players get a lot of their performance from perception that filters information before it hits the consciousness. For another, decision-making is shown to be a pre-conscious activity, and a person's supposed line of reasoning is actually a post-hoc rationalization of the decision that the sub-conscious mind dumped into the consciousness (I forget the details of how this is demonstrated). The latter result is... disturbing, because it challenges your notion of free will. But the more I observe myself and others, I can't help but think it's true.
One major upshot of this: I think there's sort of an implicit assumption that consciousness is some sort of "higher form" of intelligence. Jayne's work casts doubt on that. It seems that consciousness interacts with intelligence, but most of our "intelligence" does not come from (what he considers) the faculty of consciousness.
Observationally, this seems to agree with the state of AI research, and the distinction between weak AI and strong AI. In fact, if one accepts his theories, then weak AI is actually just unqualified Artificial Intelligence, and what we currently call strong AI is not AI at all, but Artificial Consciousness, which should rightly be a separate field.
Great recommendation of a great book—you made me wish I could read it again for the first time. I love that book: grand title, compelling style, and fascinating ideas that linger for years.
There is no unambiguous definition of consciousness because there's a segment of the population that will not accept a definition unless it includes spooky woo-woo magic.
I think that there are two possible definitions, and they get confused. This confusion does not help us to think clearly about the question.
Definition 1: Consciousness is awareness of your external environment. By that definition, my dog is conscious, insects are, and cars are becoming so. And if cars are conscious, then this must be a mechanistic property - it just takes enough sensors (and perhaps complex enough wiring).
Definition 2: Consciousness is awareness of your own awareness - being able to watch yourself think, and to think about thinking. By that definition, humans are conscious, but I'm not sure about dogs, or even apes. It is much less clear that this is purely mechanistic.
But when we have these discussions, it's going to be helpful to specify which version we're talking about.
Have they considered that it is only in consciousness that the notion of solving the mystery of consciousness can arise?
It is consciousness asking itself how it came into existence, but existence itself means arising in the field of consciousness, as it is only in consciousness that the notion of existence can arise.
So today I read the article on the Guardian website. It was about how the world's smartest minds could not seem to figure out consciousness. In it, they mentioned the problem of how there seems to be something more than just the physical stuff making up the body. But else could this other thing be?
All I could think of was that these guys did not do much computer programming. Everybody who is doing programming has heard of the binary search algorithm. At school, you first are taught a computer language or two. Next you take a data structures and algorithms class. You can not pass unless you know and can write from memory the binary search algorithm.
In English, it goes something like this. Assume an ordered array of values. Also assume two indices's, initially at the beginning and end of the array. Pick the location at the mid-point of the two indices's. Is it the one we are looking for? If so, quit and return the location index. Otherwise if the value at the mid-point
is greater than the search key, move the upper index to the mid-point. Otherwise, the mid-point value must be less than the search key and we move the lower index to the mid-point. Repeat until either the search key is found or the upper and lower indices's meet.
In essence, this is roughly how you search a phone book for a name. Everyone above the age of ten knows this algorithm.
What is interesting is that the algorithm is the same, no matter the language in which it is described. English, French, Algol or C.
The algorithm is recognizable as essentially the same in the unoptimized version described above or in an improved version where it is noticed that having once examined the mid-point, it is no longer necessary to consider it again and the indices's can be moved one location before or after the mid-point.
The algorithm is the same even in the human version wherein the pages of the phone book are searched without exactly choosing a mid-point. Instead, a rough estimate of the mid-point is chosen and that works just about as well.
So is the algorithm a thing? Yes, of course. It is recognizable. You read some unknown code and all of a sudden, the algorithm pops out of the confusion of its coding and you see the idea that the writer was attempting to reach. You can then look at the code in a new light and determine if the expression of the idea was accurate and free of mistake.
Is the algorithm a physical thing? Not at all. There can be expressions of the idea which somewhat become physical in marks of ink on paper or in coded holes in paper or even the the arrangement of a linear encoding of magnetic flux, displayed in the pulsing of electrons encountering a glowing phosphor screen.
As was alluded to above, even the representations of the algorithm are somehow at the limits of physicality.
One can even speak of the intelligence of the algorithm or in other words, the appropriateness of the algorithm to the problem domain.
For example, the algorithm depends on the amount of time required to access the array elements being roughly the same. The algorithm could become self-aware if the access to the array elements were abstracted into a function and the execution of the access function timed. If the algorithm were to be enhanced to account for this time, it could notice a violation of the basic constraints if the access times varied by much from the norm. It could even raise an alert and signal alarm at the discrepancy.
Suppose that some bone-headed programmer were to implement the binary search algorithm in layers of abstraction. But instead of representing the data in an array, the dunderhead was to layer the binary search function over a linked-list data representation. Recall that to find an item in a linked list, one must start at the first element and then examine additional elements until either the search key is located or the end of the list is reached. This form of layering obviously violates the equal access provision and would be a "stupid" algorithm. Or a “ugly” composition of otherwise decent ideas.
As a parenthetical note, I take this previous example from an actual discovery in examining some purchased code for which my company had paid something like $50,000 dollars. When I mentioned this to my counterpart, the support engineer of the seller, he acknowledged the error but countered with the observation that the stupidity in the code was not significant as the number of items in the list
was never large enough to matter. Needless to say, my expectations regarding the code, the engineer, and the company were reset in a downwards direction.
So in conclusion, an instrumented algorithm with an awareness of time, can be aware of deviations. Deviations can produce an alarm which can be sensed by other parts of the system. This kind of a system is at some level, "self-aware". And it is based on ideas which are non-physical and in some sense eternal or outside of time. Ideas which can be composited or layered.
So the question now is this. Can ideas be composited naturally or do they require a programmer?
To argue the first case is to argue for evolution and the second, for a god.
But in the first case, evolution, you still reach the point of the existence of a programmer.
Which in the limit, becomes a god, provided that communications costs are held sufficiently low and the efforts of all programmers can combine in a non-stupid manner. Probably not with money being the motivating force but rather a love of beauty and elegance at the core of their efforts.
I distinctly remember first learning some basic laws of Newtonian physics in high school, and leaving school that day dumbfounded. If all particles behave in a deterministic fashion, how is it that these particles bouncing around in my head can lead to the perception of free will and agency? It was a question that would pursue off an on for most of the rest of my life. Arriving at college, I dove into psychology with a passion, hoping that understanding the biomechanics of the brain would give me some clues. I turned to computer science and was instantly drawn to AI. I read everything I could get my hands on on the subject, with Daniel Dennett and Douglas Hofstadter being favorites of course (I still revisit Godel Escher Bach from time to time). I even experimented with various psychedelics looking for that elusive insight that would explain things.
Now I'm in my 40's and of course I still don't have "the answer". I'm not as consumed by the question anymore, focusing instead on just enjoying my days on earth here with my family and friends.
But how on earth anyone can belittle the question of consciousness blows my mind, so to speak. :)