I am completely on the side of Hammy on this one.
The “Looks like a duck” argument that says a physical system that behaves conscious must actually be conscious totally begs the question. Implicit in the argument is an already conscious knowing subject interpreting the behavior of a physical system and assigning it meaning. It’s no different than assigning numerical meaning to abacus beads. When we manipulate the beads manually according to an algorithm they are still just dead wood. They have no meaning in themselves. Likewise, electronic switches and lights have no meaning other than what they get assigned by a knowing subject. And by logical extension, the activity of neurons, firing and not firing, also have no inherent meaning. Yes, they correlate with mental properties. But beads and switches can also correlate with their assigned meanings. As such I see no justification for claiming that the brain, as a physical mechanism, does anything more that produce signs awaiting interpretation by a knowing subject. Signs themselves have no essential properties in common with the things they signify.
I think David Bentely Hart states the issue well...
"Computational models of the mind would make sense if what a computer actually does could be characterized as an elementary version of what the mind does, or at least as something remotely like thinking. In fact, though, there is not even a useful analogy to be drawn here. A computer does not even really compute. We compute, using it as a tool. We can set a program in motion to calculate the square root of pi, but the stream of digits that will appear on the screen will have mathematical content only because of our intentions, and because we—not the computer—are running algorithms. The computer, in itself, as an object or a series of physical events, does not contain or produce any symbols at all; its operations are not determined by any semantic content but only by binary sequences that mean nothing in themselves. The visible figures that appear on the computer’s screen are only the electronic traces of sets of binary correlates, and they serve as symbols only when we represent them as such, and assign them intelligible significances. The computer could just as well be programmed so that it would respond to the request for the square root of pi with the result ‘Rupert Bear’; nor would it be wrong to do so, because an ensemble of merely material components and purely physical events can be neither wrong nor right about anything—in fact, it cannot be about anything at all. Software no more ‘thinks’ than a minute hand knows the time or the printed word ‘pelican’ knows what a pelican is."
– David Bentley Hart
The “Looks like a duck” argument that says a physical system that behaves conscious must actually be conscious totally begs the question. Implicit in the argument is an already conscious knowing subject interpreting the behavior of a physical system and assigning it meaning. It’s no different than assigning numerical meaning to abacus beads. When we manipulate the beads manually according to an algorithm they are still just dead wood. They have no meaning in themselves. Likewise, electronic switches and lights have no meaning other than what they get assigned by a knowing subject. And by logical extension, the activity of neurons, firing and not firing, also have no inherent meaning. Yes, they correlate with mental properties. But beads and switches can also correlate with their assigned meanings. As such I see no justification for claiming that the brain, as a physical mechanism, does anything more that produce signs awaiting interpretation by a knowing subject. Signs themselves have no essential properties in common with the things they signify.
I think David Bentely Hart states the issue well...
"Computational models of the mind would make sense if what a computer actually does could be characterized as an elementary version of what the mind does, or at least as something remotely like thinking. In fact, though, there is not even a useful analogy to be drawn here. A computer does not even really compute. We compute, using it as a tool. We can set a program in motion to calculate the square root of pi, but the stream of digits that will appear on the screen will have mathematical content only because of our intentions, and because we—not the computer—are running algorithms. The computer, in itself, as an object or a series of physical events, does not contain or produce any symbols at all; its operations are not determined by any semantic content but only by binary sequences that mean nothing in themselves. The visible figures that appear on the computer’s screen are only the electronic traces of sets of binary correlates, and they serve as symbols only when we represent them as such, and assign them intelligible significances. The computer could just as well be programmed so that it would respond to the request for the square root of pi with the result ‘Rupert Bear’; nor would it be wrong to do so, because an ensemble of merely material components and purely physical events can be neither wrong nor right about anything—in fact, it cannot be about anything at all. Software no more ‘thinks’ than a minute hand knows the time or the printed word ‘pelican’ knows what a pelican is."
– David Bentley Hart
