Re:Daniel Dennett’s Science of the Soul
by TheOtherHobbes on Thursday March 23, 2017 @10:17PM
(#13946063)
That last sentence seems to be true as far as it goes, but I agree with the criticisms that it doesn't add much to the debate.
Suggesting that "consciousness" is some kind of smooth continuum is just plain wrong. Empirically, it's more like a series of discrete abilities that kick in as more and more sophisticated neurological processes become available. (We don't really know that for sure, and Chalmers and the animists/dualists may turn out to be right. But that kind of dualism is very fuzzy, and if you're trying to build intelligent systems it's hard to do anything useful with it.)
One tell-tale is the mirror test. Some animals understand they are the creature in the mirror, while others see a different animal - and respond accordingly, driven by instinct.
That's a binary test. You can't "sort of" see a reflection of yourself.
It's true you can get it wrong some of the time - as a human, you can be tired or confused or drunk or ill or simply in a poorly lit environment.
But the ability to make the distinction, assuming you're functioning normally, is either there or it isn't. You either have the self-abstractive processing needed to operate with a self symbol and to understand that you're experiencing it in the mirror, or you don't. And if you don't, you never will.
There are many other possible binary tests, and consciousness seems to exist as a sum total of the pass/fail profile for all of them.
That's why the lion question is irrelevant - it doesn't go far enough to make this point. It assumes - on the basis of no neurological theory - that being able to talk would transform the instinctual and perceptual mechanisms built into lions so completely they'd become irrelevant, and you'd have something that looked like a lion but thought like a philosopher and was no longer even slightly interested in chasing after antelopes.
This is clearly wrong, because being able to talk hasn't made us stop feeling and acting like primates, or - often - even dumber animals.
With those kinds of details in mind, a multidimensional consciousness scale could be defined quite precisely.
But Dennett's fuzzy argument about being sort-of conscious and sort-of not is - I think - hand-wavey vagueness. It lacks the precision needed to do that usefully.
Suggesting that "consciousness" is some kind of smooth continuum is just plain wrong. Empirically, it's more like a series of discrete abilities that kick in as more and more sophisticated neurological processes become available. (We don't really know that for sure, and Chalmers and the animists/dualists may turn out to be right. But that kind of dualism is very fuzzy, and if you're trying to build intelligent systems it's hard to do anything useful with it.)
One tell-tale is the mirror test. Some animals understand they are the creature in the mirror, while others see a different animal - and respond accordingly, driven by instinct.
That's a binary test. You can't "sort of" see a reflection of yourself.
It's true you can get it wrong some of the time - as a human, you can be tired or confused or drunk or ill or simply in a poorly lit environment.
But the ability to make the distinction, assuming you're functioning normally, is either there or it isn't. You either have the self-abstractive processing needed to operate with a self symbol and to understand that you're experiencing it in the mirror, or you don't. And if you don't, you never will.
There are many other possible binary tests, and consciousness seems to exist as a sum total of the pass/fail profile for all of them.
That's why the lion question is irrelevant - it doesn't go far enough to make this point. It assumes - on the basis of no neurological theory - that being able to talk would transform the instinctual and perceptual mechanisms built into lions so completely they'd become irrelevant, and you'd have something that looked like a lion but thought like a philosopher and was no longer even slightly interested in chasing after antelopes.
This is clearly wrong, because being able to talk hasn't made us stop feeling and acting like primates, or - often - even dumber animals.
With those kinds of details in mind, a multidimensional consciousness scale could be defined quite precisely.
But Dennett's fuzzy argument about being sort-of conscious and sort-of not is - I think - hand-wavey vagueness. It lacks the precision needed to do that usefully.