Discussion:
  1. David J. Chalmers (1995). Absent Qualia, Fading Qualia, Dancing Qualia. In Thomas Metzinger (ed.), Conscious Experience. Ferdinand Schoningh. 309--328.
    It is widely accepted that conscious experience has a physical basis. That is, the properties of experience (phenomenal properties, or qualia) systematically depend on physical properties according to some lawful relation. There are two key questions about this relation. The first concerns the strength of the laws: are they logically or metaphysically necessary, so that consciousness is nothing "over and above" the underlying physical process, or are they merely contingent laws like the law of gravity? This question about the strength (...)
    Direct download (9 more)  
     
    My bibliography  
     
    Export citation  
Back    All discussions

2011-04-18
role of memory in the dancing qualia scenario
What is the role of memory in the dancing qualia scenario?

It strikes me that i cannot perform direct comparisons between my conscious experiences at different points in time - no more than i can directly compare my experiences to those of others.

In claiming that my experience of a red apple has remained the same "redness" over time, i must be comparing a perceptual experience *now* against the experience *now* of a memory of a previous experience.

The reductio asks us to imagine there being a difference in experience just due to differences in the material substrate of cognition. It seems plausible to me that when an experience is serialized while running on one substrate and deserialized while on another, the difference should go unnoticed. For example, the red experience of a neural system could be remembered as a blue experience when invoked on a silicon circuit, so that the comparison always succeeds.

Put differently, i wonder in what way the following scenario is not analogous to dancing qualia:

Consider two silicon circuits, one using binary representations of numbers, and the other using ternary. Merging the circuits in the way Chalmers describes must involve the circuit switching mechanism transforming between representations.

If these circuits were carrying out a process which repeatedly performed a computation (say pi to the nth place) and then compared against previous results, i wouldn't expect dancing between bases to perturb the result.

How does the role of qualia in a red color experience differ from the role of base-representation in a pi-to-the-nth numerical result?

2011-05-01
role of memory in the dancing qualia scenario
Reply to Jeremy Awon
I think you might have a point. 

As Joe is having his brain steadily replaced with silicon-based hardware, which ex hypothesi does not support conscious experience, we have to ask the following question: "As Joe's red qualia fade out, must he notice the change?" If he must, then his functional profile must change accordingly, and this is what Chalmers denies. As far as I can tell, his reason for the denial is that surely it would always be possible to maintain Joe's functional profile, including his noticing his bright red qualia, no matter what. The absurdity is supposed to be the scenario in which Joe continues to notice and enjoy his vivid red qualia even when they are no longer there. 

So yes, one way to respond to this 'reductio' would be to find ways to believe that the said scenario is not absurd after all.  Your way might work. 

But instead of finding ways in which the siliconised Joe could maintain his original functional profile (including his qualia beliefs), why not take a simpler route? That is, why not challenge the claim that the envisaged absurdity could arise?  If, in the normal course of events (if, indeed, any of these events could be regarded as normal) Joe's qualia-beliefs will closely track the fading, or dancing, why not just allow that as an assumption? His qualia-beliefs will closely track his qualia, unless extraordinary efforts are contrived to peel the two apart (thus maintaining his original beliefs throughout). And if such efforts are made, so what? There is no absurdity in his qualia-beliefs having been artificially maintained as his qualia disappear, since to artificially maintain them is to break the usual connection. 

My ultimate reason for rejecting Chalmers' argument is that while I agree that I must allow faithful qualia-tracking in a conscious being, because I claim that silicon does not support consciousness, and any changes in qualia must be noticed, this entails nothing whatever for a permanently, unchangingly, unconscious functional simulacrum of the original Joe. It is the changing of Joe's qualia that prompts the tracking, and for the imagined robot there is no change. 

So yes, I concede that a conscious Joe will notice every change in his qualia, and his functional profile will change to follow suit. Ultimately he will be reporting that they have almost disappeared. If by some technical means his beliefs were forced to remain constant, that would be at the expense of the normal reliability of his qualia-beliefs, so would not present an absurdity. But having granted all that, I can still freely claim that a silicon-based functional simulacrum of Joe might be entirely unconscious. The principle of organizational invariance has not been established. 






2011-05-03
role of memory in the dancing qualia scenario
Reply to Brian Crabb
Hi Brian,

In this view then (with apologies if i've misunderstood),

- purely neural-Joe reports experiences for which he actually has qualia.

- purely silicon-Joe makes indistinguishable reports, yet is divorced from reality in that he does not actually have qualia.

- transitional-Joe is behaviorally distinct from both, accurately reporting that his qualia are fading.

"If by some technical means his beliefs were forced to remain constant, that would be at the expense of the normal reliability of his qualia-beliefs, so would not present an absurdity."

I think such means are in place, but as an uncontrived feature of substituting one system with another which provides the same input-output mappings.

Does this view deny that such mapping is possible for components of neural-Joe's "circuitry" - except that it suddenly does not matter (behaviorally) when all components have been re-implemented in silicon?

2011-05-04
role of memory in the dancing qualia scenario
Reply to Jeremy Awon
Hi Jeremy -

What I am suggesting is an account of what I would expect to happen as Joe has his brain progressively 'siliconised'. I am suggesting that, unless unusual technical devices are deployed to keep his beliefs and reports constant, he will notice that his qualia are fading, and his functional profile will track that change. So when he gets to the point where his qualia have all but disappeared, he will notice this and accurately report it. So in that event I accept the principle of organisational invariance; particular experiential states determine particular functional profiles. 

If, on the other hand, unusual technical devices are deployed to keep his beliefs constant, this is only because the usual felicity of his reports has been disrupted. There is no absurdity. If we take Joe's career to the point where he is completely siliconised, I can even accept that he will end up reporting vivid qualia when he has none at all. 

So it really doesn't matter. Either he notices and reports his fading qualia to the end, or he is in some way forced to believe and report falsely that his qualia remain constant. No absurd consequences occur in either case.

That is the main point: I have presented a common-sense account of what would happen, and I think the burden is on Chalmers to say why it could not happen. Specifically, I am asking him to explain why he thinks that any circumstance in which Joe is simultaneously experiencing fading qualia and yet continuing to report them as being normal must be absurd. I don't think he has provided that explanation.

The second point is that whatever we conclude about Joe, as an originally fully conscious person, has no bearing on what mental-functional correlation laws must apply to a robot which has never been conscious. Joe's qualia reports might well track his fading qualia even to the point of extinction, but they do so in virtue of the change. A permanently mindless robot suffers no such constraints. 

I suppose Chalmers could argue that if Joe's functional profile can be made to remain constant without disrupting the felicity of his qualia reports (which he has not shown to be the case), then it follows that silicon can support consciousness. Even then, however, he hasn't shown that silicon must support consciousness. He will only have shown that under the specific conditions which enable Joe to maintain his original functional profile while his brain is replaced with silicon will he experience constant qualia. Again, the permanently mindless robot which is an exact functional simulacrum of the original Joe remains possible.







2011-05-04
role of memory in the dancing qualia scenario
Reply to Jeremy Awon
>>I think such means are in place, but as an uncontrived feature of substituting one system with another which provides the same input-output mappings.

Does this view deny that such mapping is possible for components of neural-Joe's "circuitry" - except that it suddenly does not matter (behaviorally) when all components have been re-implemented in silicon?>>

Sorry - I didn't address your question directly. My answer is:


It might be possible to maintain Joe's functional organisation throughout the silicon substitution while at the same time maintaining the veracity of his qualia beliefs and reports. If it is possible, then logically it follows that Joe's experiences must have been preserved too. And if that turns out to be the case, it will have been shown that Joe's functional organisation is sufficient to secure, or indicate, his state of mind. No fading qualia. That would be a demonstration of Chalmers' organisational invariance. 


The problem is that we don't yet know that all of this is possible. To assume that it is; that is, to assume that whilst preserving Joe's functional profile we at the same time maintain the veracity of his beliefs and reports, simply begs the question we are trying to answer. We could assume it if we already knew that Joe's functional profile is a sufficient condition for the veracity of his qualia beliefs, but we don't already know that. In fact, it is what we are trying to find out. 


Regarding the permanently mindless robot, I concede that if the above had been established as a mental-functional law of correlation we would then have to face the implication that the mindless robot is a nomological impossibility, but ... the above has not been established. 
 
I hope this is a bit clearer!