Will Solving The ‘Hard Problem’ of Consciousness Unweave the Rainbow?
Some say in fifty years or so we'll have enough neuro-scientific evidence to completely describe the functioning of the brain. The question is, will this mountain of evidence be enough to explain the emergence of human consciousness? Consciousness. This familiar yet indescribable experience we all have, an awareness, something we can't physically point to nor experience from another's viewpoint.
Being a hypothetical question about some future state of our knowledge, it has mainly been of academic interest to philosophers. But I actually think it's relevant to all of us because it accesses two fundamental questions about what it means to be human. First, on a practical level, is consciousness amenable to explanation? Second, on a mystical level, if consciousness can be explained, will its essence be lost?
Consciousness: 'easy' and 'hard' problems
An influential approach to consciousness splits the question into the 'easy problems of consciousness' and the 'hard problems of consciousness' (Chalmers, 1995). The so-called 'easy problems' are things like finding out how memory or attention actually work, the nuts and bolts of these functions. And even though these are the easy problems, scientists are still having considerable difficulty with them.
But, argue people like Chalmers, once we've described all these functions, we still won't fully understand consciousness. This is because we won't have addressed the so-called 'hard problem'. This is the feeling of what it is actually like to be you. That ineffable you-ness that no one else can share. Your experience.
Chalmers doesn't represent the most extreme example of this position, there are those who argue we can never truly understand consciousness. At least Chalmers acknowledges there are possibilities, although new conceptual techniques need developing.
There is no 'hard problem'
On the other side of the fence are those who argue the distinction between the 'hard problem' and the 'easy problem' is at best ill-advised and, at worst, plain dangerous. Just because we can't conceive of how consciousness can emerge from the description of the easy problems like attention and memory etc., doesn't mean it never will (Churchland, 1996). Just because we can set up complex philosophical arguments about what might be true in a thought experiment, doesn't mean it explains what is true here and now.
Philosophers of mind like Dennett argue that consciousness emerges from the physical processes of the brain (Dennett, 1996). Effectively he is saying there is no 'hard problem' to explain, some even argue he is saying there is no such thing as consciousness, rather he is redefining consciousness as 'reportability' (Chalmers, 1997).
Unweaving the rainbow
So there's a glance at two views on the so-called 'hard problem' of consciousness. But being a student of psychology, in addition to the actual arguments themselves, I'm naturally drawn to the motivations people might have for which they choose.
While there's a lot of complex discussion in this area, I think it partly comes down to whether you're relaxed about the idea that science might one day be able to explain the essence of human experience. For many people, I think this is an extremely uncomfortable thought. What Keats, talking of Newton's findings, refers to as a fear of 'unweaving the rainbow' - the fear that explaining something might somehow reduce the magic of it - is very real.
What's your view?
You can vote below on whether you think current scientific methods will ever be able to explain consciousness. Of course, it would be great to know your views, so go ahead and post a comment. I'd love to know what you think.
References - Click here to toggle visibility
Join 22301 readers
Hi anonymous, thanks for your comment. I see what you are getting at and I agree that from the persective that reasoned argument or evidence could, in some sense, resolve this problem, the discussion of psychological biases doesn't help.
But, imagine we had a problem that can't be 'resolved' by reasoned argument or by empirical evidence. Then, all we'd have left is the discussion of psychological biases - at which point this would become interesting as it tends to at least explain why people are making these arguments.
I think this hard/easy problem is exactly of this latter type. Perhaps you don't and that is why you are taking up a different position on this.
Interesting stuff - thanks for the comment.
For the record, I agree that trying to understand why people have pro- or anti-hard problem intuitions in the first place is fascinating. However, I would expect any sort of work on these matters to be conducted in a serious scientific fashion. The "unweaving the rainbow" explanation seems more like a speculative, hand waving, folkie explanation of what some of the anti-folks think must be motivating the pro-folks.
To the extent that psychological biases and intuitions play a role in views on the hard problem, I think it is more likely that such intuitions operate on the level of the key postulates of the typical hard problem kind of argument, rather than operating on the level of not liking the consequences of the rejection of the hard problem.
For instance, the philosophical discussion on the hard problem often turns on closely related matters like whether it is conceivable for any kind of neuronal process to take place in the absence of phenomenal consciousness, or whether structural and functional facts exhaust the complete set of facts regarding phenomenal consciousness. To the extent that intuitions play a strong role in forming one's stance on the hard problem, the relevant kinds of intuitions are probably about the degree to which one finds these key sorts of claims in the hard problem discussion plausible and compelling or not.
I take your point. Thanks again for your interesting and provoking comments.
I would like to pose a thought experiment.
Suppose someone came along with two identical boxes and explained that one box contained a mechanism (based perhaps on studies of the human brain) that was truly conscious. He then explained that the second box was designed to appear to be conscious, but was not. The first box could, for example, suffer pain, whereas the second could pretend to suffer pain but would not, in fact suffer anything because it was not even conscious! Each box had a USB port, and so could be interrogated - rather as in Turing's hypothetical experiment.
The problem with current 'emergence' theories of consciousness is that they can't explain how to distinguish between these two possibilities - even in principle! This is a vastly greater level of ignorance of - say- the nature of dark matter, where suitable experiments to probe the concept are often discussed. Science doesn't have any way of measuring consciousness, and some at least would claim that the two boxes would be in principle indistinguishable! That sounds potentially reasonable until you think that it means that concepts like torture are basically meaningless!
Some years ago, I had some peripheral involvement with Artificial Intelligence (AI) (I developed a PROLOG compiler), and I gradually realised that much AI research was really about faking a level of awareness that was not really present. For example, since much research started in the US, a common theme was to try to get an AI program to answer questions about a restaurant scene - "Why did John try to speak to the waitress?". If the program gave some reasonable answers, it was easy to slip into thinking that it 'understood' about restaurants, even though it knew nothing about hunger, human friendship, dating, dieting, or indeed much at all!
Although in a sense, a failed endeavour tells you nothing except how difficult it is, the essential failure of AI is extremely interesting in this context, and led me to think of the above thought experiment.
My best guess is that consciousness will be a very tough nut to crack and will involve some sort of revolution in science - not just an appeal to complex systems and emergence!
Jeremy,
Yours is the 'standard' answer - almost the only answer - that orthodox scientific opinion can give. Because we have become used to well researched, but weird theories in physics, such as Special/General Relativity, Quantum Mechanics, etc., we have become rather tolerant of superficially absurd scientific ideas. It is only in this context that your comment about us all being "black boxes pretending to feel pain" makes any sense at all. Others seem to use Libet's neural timing experiments to argue that consciousness does not exist at all (perhaps another way of expressing your views).
However, it seems to me that acceptance of these remarkable theories was hard won. They are represented by a full mathematical exposition, and have been tested in a large variety of ways. Even now, it is still considered worthwhile to test these theories in novel ways. Contrast this with the orthodox view of consciousness, which seems to be perilously close to, "It must work this way, because we can't think of anything else".
In about 1980, it was almost taken for granted that Artificial Intelligence would be comparatively easy to achieve with the increasing power of computers. This confidence seemed to follow directly from the idea that natural consciousness was effectively a form of information processing. It is well known that the nature of the hardware - if sufficiently powerful - has no bearing on the types of computation that are possible. Vast sums of money were pumped into AI research, and very, very little emerged. Subsequently, people tried making artificial neural nets (usually implemented on a computer), and again there was a huge burst of excitement, but very little real results.
My feeling is that the failure of AI, Libet's paradoxical experiments (and maybe the even more paradoxical Bierman/Radin experiment, which seems to show that the brain has some awareness of an emotionally disturbing event BEFORE it has even happened), point to the need for a re-think - particularly when the 'orthodox' view of consciousness has become so weird!
Hi David, just to be clear, I'm not arguing that consciousness doesn't exist, although I suspect we might have different definitions of consciousness, perhaps with mine being more 'stripped down'.
Interesting comments about AI. It seems hard to believe now that there was ever such huge optimism that high-end consciousness could be created in a machine. Whether this will ever be possible is an argument for another thread, but this optimism seems striking now.
Thanks for the pointer to the Bierman/Radin experiment, I hadn't come across that before. This is the most accessible description of their studies I can find at the moment (Evening Standard).
Jeremy,
I was quite excited about AI all those years ago, until I realised that, despite the hype, the subject was not going anywhere! I don't think that enthusiasm was so hard to understand. If the brain 'computed' then it should be easy to encapsulate that in a much simpler computer program which could benefit from being designed rather than evolved. Remember, Alan Turing already had this dream, back in the days when computers were absurdly slow.
You would normally expect that any complete specification of the neural basis of consciousness (as hypothesised at the start of this blog) would, of necessity, be stored on a computer, and would be executable by simulation. Such an executable description would constitute an artificial intelligence in the full sense.
A genuine AI computer program, that could genuinely feel emotion, would, I feel, be a deeply paradoxical entity. For example, imagine debugging such a program - repeatedly starting and stopping it, and re-running it with the same data! Would the program 'feel' its emotions while it was paused in the debugger, and would it 'feel' those emotions a second time if it got re-run on the same data?
Furthermore, a computer program can be thought of as a mathematical transformation of its data into the corresponding output. The actual computer is only needed as a crutch to help us plough through all its myriad steps. Thinking of a program as a purely mathematical entity, seems to cause any emotions it 'feel's to belong to the same timeless realm that contains mathematical theorems!
Suppose we say that the computer simulation would not actually be aware - feeling emotions. In this case, the simulation would have to be missing something that the physical brain possesses. What could that be? For example, all the relevant hormone levels could be stored as real numbers inside the computer simulation.
Roger Penrose, however, speculates that physical systems may be capable of intrinsically non-computable processes which could not, even in principle, be simulated on a computer. He does this because he claims that any computer program would be subject to Godel's theorem, whereas mathematicians seem to evade it. Although there is a lot of horribly obscure debate over this point (detailed in "Shadows of the Mind"), I feel he has a point. Of course, any conclusion regarding mathematical thought, would presumably apply equally to thought of any kind.
Science has always thrived on paradoxes - they are usually the harbingers of the next paradigm shift!
Very interesting David, thanks for your comment.
Hard Problem
Comprehensive survey of beliefs
We're working to create a concise comprehensive survey of beliefs of experts on the topic of the "Hard Problem of Consciousness" in everyone's own words. If you have ever thought about this in any way, we would love to know your thoughts on this issue, and how you think this may or may not be resolved in the future. We are using the new Canonizer system prototype at http://canonizer.com
The specific topic page in the prototype, including links to the beliefs summarized so far, is here:
http://test.canonizer.com/topic.asp?topic_num=23
Looking forward to hearing what you believe.
Thank You.
Brent Allsop
Hey all!
Apologies but I haven't read all or even half of the comments.
I'd like to challenge that the complete understanding of something requires you to be able to control, manipulate and alter it, not so?
If that is the case, does that not mean that understanding consciousness would mean that we'd be able to alter it? In a sense, wouldn't it also mean that one would be able to play with metaphysical(?) existence...?
For a good overview and understanding of why the 'hard problem' is a problem and why it’s a matter of limiting the domain of scientific methodology I would recommend an author called Ken Wilber. Empiricism didn't always mean only direct sensory (5 senses) experience, but originally referred to any direct experience at all (including interiorly ‘felt’ experience). And no, I’m not condoning wacky pseudo science etc. I would recommend reading some of Ken Wilber's writings, they are much better at describing these issues than I am able to in this short blog, with qualifications no less! He even goes into detail in "The Marriage of Sense and Soul" of the historic disassociation of the different strands of science, each equally valid with its corresponding validity claims, and falsifiability injunctions etc. and why they are all important and cannot be explained by or reduced to each other. Note: "The Marriage of Sense and Soul" is not his latest book, by all means consult this and others (e.g., "Es, Ecology, Spirituality", "A Brief History of Everything" among others) but for his up-to-date ideas refer to his later works which have been updated to integrate new ideas and previous criticisms.
thanx
"This is the feeling of what it is actually like to be you. That ineffable you-ness that no one else can share. Your experience."
the hard part. consciousness is perceived differently by you, your wife/husband, neighbor, and so on. from our first "conscious" moment on, our brains interpret and learn to interpret things independently of those around us. the way in which we handle situations throughout our lives is shaped by our brain function, and as no life is identical, no brain could possibly function in the same way.
but that's the beauty of life and the brain. unless there is a way to control an environment and the way in which a group of individuals are allowed/able to perceive that environment from the onset of consciousness, the task of mapping the functions of neurons in order to track an individuals thought process would seem fruitless to me.
would that not be considered mind control?
i can't quite express in words exactly what I'm trying to say as clearly as I would like to. I am just a business student with an interest that lies in the mind...i have a long way to go to fully understand as well as you all seem to but would appreciate feedback , if only to say I am not making any sense at all.
I think one issue which makes consciousness very different from other problems is that this problem revolves around the inner state of the brain. To give an example, compare it to vision. We can easily imagine a machine who can "see", simply because the ability to see can be confirmed with a "black box" experiment: we can give the machine some inputs, receive some outputs and conclude the machine can see. But for consciousness, I cannot imagine how can we build a similar experiment simply because consciousness is about how a system makes decisions rather than the decisions itself (for instance, we can all agree that a rule based system is not conscious even though theoretically we can simulate a person through a -very- lengthy list of rules).
From my understanding, this is the difficult part of the problem, that we somehow need to dig deep into the brain and observe how brain operates in a more detailed level. But until we have reached the point where we an incredible understanding of brain functions, we can all just give random philosophical conjectures about the nature of the consciousness.
Does Pandeism solve the hard problem of consciousness? See Intriguing Metaphysical Parallels between the Consciousness Debate and Pandeism for a discussion.
** Turning mind inside out **
'Mind.' That what's left of soul after the philosophers got through with it. ‘Body' and 'mind' are opposite sides of the same counterfeit $100 bill.
'Mind' is the last immaterial bolt hole for 'soul'. 'Body' is a misleading abstraction. Zombies are animated bodies. (Neuroscience deals only with zombies.) Persons are not animated bodies.
'Human being' is a misleading abstraction. A society consists of persons living in nature, sharing a common highly artificial, web of interactions called a culture. Outside of culture there are no persons. Inside of nature there are no machines, no plans, no purposes, no aims. Nature is silent.
Culture blabs incessantly. Whatever can be explained by language, including my experience of so-called "private" perceptions, belongs solely to culture, not to nature. Language is our abstract, interactive, collective consciousness. Conscious ‘mind' disappears into language.
'Body' is merely an abstraction from 'person.' Specifically, part of a failed attempt by ancient peoples to account for the difference between a "living" person and a "dead" body.
One of hellenistic xianity's darkest gifts is its focus on the so-called individual soul. Western philosophy gets nowhere by having been theology's lickspittle. Basically, we must jettison xianity's focus on the 'soul' of each individual, cut off, alone under the pitiless gaze of the eternal judge.
The irreducible locus of investigation must be culture -- a vital, shared, abstract reality which only we humans are able to inhabit. Explanations drawn from neuroscience, genetics, and evolutionary biology cannot be expected to provide a full account of culture. They were not designed to do so. They will not do so.
Persons, language, and culture arose simultaneously. (Do I need add without the assistance of any divine anthropomorphic pseudo-person?) In that sense culture is irreducible; it is the given. Science itself is derivative, a cultural creation. (A very peculiar one. But that's another matter.)
bipolar2
© 2008