Of course, it only remains a hard problem if it is assumed that consciousness is an emergent property. That is a premise, and the hard problem of consciousness demonstrates that, for the time being at least, it is an unprovable premise.
To put it another way, it could very well be that it is a hard problem because emergence is not true. There is no evidence that excludes the possibility that consciousness is transcendent, rather than emergent. However, that may well be trying to prove a negative, which is mostly fruitless. So, the fact that it cannot presently be proven that consciousness is an emergent property, does not necessarily mean that it is not.
A model for consciousness that separates perception from sensory processing itself would be somewhat akin to saying that the soul is wired into the brain, and receives all of its perceptions from the brain, but that the soul is not the brain. I don't think that this can ever be proven or disproven, because it requires introspection, which is at its very nature subjective.
But what do I know? Let's see where the research goes...
That doesn't show in that Wikipedia article, though. All it contains is some self-referential links. Ie I didn't follow the "qualia" link, but I am pretty sure if I go there, it will just point back to the "hard problem of consciousness".
I thought it said plenty. Admittedly I only read the first paragraph and went "oh, I get it". They're just things that are subjective, even if we nominally agree on an objective definition.
Like, we can all agree that a ketchup bottle is red, but what it really looks like to each of us might be totally different. My red might be your blue, but since your blue things are called red, and my red things are called red, we pretend it's just a single color, instead of a potentially infinite number of colors, a different one for each person.
Significantly, this hypothetical becomes real when we talk to, say, a colorblind friend. A ketchup bottle is still red, but it's definitely not the same red-- since it seems "closer" to colors that the non-colorblind would say.
Nice writeup. I bought David's book on consciousness after having a long talk with him at a Consciousness and Quantum Mechanics meeting about 10 years ago. I think that this is interesting stuff. Really, why should we have evolved with the kind of inner life (qualia) consciousness? What evolutionary advantages did it give us?
Not only is this problem of great philosophical interest, I always find it sociologically interesting as well: people who don't see why there is a "hard problem" are typically entrapped believers in the materialist/deterministic paradigm. I know because I was. :)
Put another way, if free will is an illusion, who is it an illusion to?
I think I know what you mean. A while back a saw a "why am I me and not someone else?" type question on Ask Reddit. And everyone had these pre-canned answers they thought were obvious. I got really frustrated that no one seemed to think it was an interesting question.
Or am I the only one who thinks that's a legimate question?
For one thing, I am skeptic about the notion of "mirror neurons". Or maybe the name is badly chosen. I think what happens is that certain neurons (or regions of the brain) associate with certain abstract concepts. It is not surprising that the same neurons light up no matter if I kick a stone with my foot myself, or see somebody else kick a stone with their foot. Both times "abstraction neurons" for things like stones, feet, kicking, etc. would light up (and by that I don't mean there is a genetically encoded brain region to encode "stones", I don't think there is). Why not? As an engineer, the concept of introducing specific "mirror neurons" seems unnecessary and too wasteful to make sense. I am not a specialist for his mirror neurons theory, so maybe I am doing him an injustice. As I said, the name could be badly chosen. And what I could imagine is that the brain is prone to make certain abstractions - just as it is prone to learn a language (it seems likely to me that parts of the brain are structured to facilitate learning languages, or recognizing faces - a bit like there being general purpose CPUs and specialized digitial signal processors in electronics). So in that sense they could be on to something, although I think their way of describing it would be rather misleading.
The main thing is, though, that the question "what is consciousness" does not make sense, therefore if he gives an answer to that, it has to be nonsense.
Now I only skimmed the article, and it might be that he is not actually pretending to give an answer. Rather, he is redefining the problem in terms of his mirror neurons theory. So it would only be marketing (it sure makes for good marketing to pretend to answer the "hard question"), which would be a bit more forgivable, but ultimately still be rubbish.
why asking questions like "what is consciousness"/"why does consciousness exist" doesn't make sense ? it seem like many questions being asked in evolutionary biology/biology/neurology.
Because nobody can say what they mean by "consciousness". "Why does consciousness exist" implies that "it" (what?) exists, which likely isn't the case.
To put it another way, it could very well be that it is a hard problem because emergence is not true. There is no evidence that excludes the possibility that consciousness is transcendent, rather than emergent. However, that may well be trying to prove a negative, which is mostly fruitless. So, the fact that it cannot presently be proven that consciousness is an emergent property, does not necessarily mean that it is not.
A model for consciousness that separates perception from sensory processing itself would be somewhat akin to saying that the soul is wired into the brain, and receives all of its perceptions from the brain, but that the soul is not the brain. I don't think that this can ever be proven or disproven, because it requires introspection, which is at its very nature subjective.
But what do I know? Let's see where the research goes...