meekerdb wrote:
On 3/27/2015 4:54 PM, Bruce Kellett wrote:
I understand counterfactual correctness, but I think the concept is
misapplied -- even to the extent of making a category error.
Counterfactual correctness can be ascribed to a computer/calculator
but not to a calculation. A calculator would not be counterfactually
correct if it gave the same output for every input, but a calculation
is a calculation! It is a single thing -- one output from one input.
If you change the input, in general you would get different output.
But then that would be different calculation. It is a category error
to ask for counterfactual correctness from an individual calculation.
If I do a calculation with pencil and paper, writing out the steps of
my calculation, that is still a calculation even after I have
finished. It is still the same calculation 10 years later (if the
paper is intact). IIt is not counterfactually correct because I do
different calculations on different pieces of paper, leaving the
original recoded calculation intact. But it is still a calculation --
what else would you call it?
A fair point. But the MGA tries to link consciousness to computation.
I would argue that this is where the first, and probably the most
important, error creeps in. Why should you call a sequence of brain
states a computation or calculation? If you want to simulate that
sequence of brain states, the computer has to do a lot of calculations
to mimic synapse potentials, ion flows and all the rest of it. The end
result of these calculations is a simulation of the original brain
states. At no point does it have to be assumed that this sequence of
brain states is actually a calculation of anything.
Most people intuit that a certain sequence of brain states instantiates
some conscious thought. And further that the particular brain material
is not necessary to this instantiation, rather it is something about the
computation.
So what is your intuition about the relation between
computation and consciousness. Is it just the calculation instantiated
in the brain that creates the consciousness and could that same
calculation then create the same conscious experience when written out
on paper or realized by a one-entry lookup table. That seems wrong too.
No, as I said, I do not think it is helpful to describe the sequence of
brain states as a calculation. If you simulate the actual brain states
by doing a lot of calculations on a computer, then you will reproduce
the original conscious moment. But the conscious moment itself does not
calculate anything. The simulation of brain states could be written out
on paper, or use any number of look-up tables (as efficient programs
tend to do). It is still a simulation of the original brain states, and
if accurate, the conscious experience will be recreated.
Another possibility is that all those neurons that /*didn't*/ fire in
the calculation were just as necessary to the experience as the one's
that did. That seems quite plausible to me.
I find the notion quite bizarre. It is the actual sequence of actual
brain states that is important. If some neuron didn't fire, then they
did not contribute to /that/ conscious moment, no matter that they might
be crucial to other, /different/, moments of consciousness.
Translating that into what
it would mean in terms of an AI is that the transistors that didn't
switch were necessary, not to the calculation, but to the
computation/conscious experience instantiation. Counterfactual
correctness is the Platonia version of this - I think.
As I said, conterfactual correctness has very little to do with the
actual conscious moment. That is given simply by the sequence of actual
brain states -- this sequence does not really calculate anything.
Computationalism ultimately rests on a confusion between a simulation
and the calculations necessary to produce that simulation.
Brent
I think this basic confusion between the calculator and the
calculation renders the MGA toothless. It does not establish that the
recording cannot be conscious. The recording is as much a calculation
as the original. If you degrade the film/recording, then you finally
lose consciousness, but that is beside the point. It is just like
rubbing out or burning your original paper calculation. It is only if
you insist that your computing mechanism is counterfactually correct
that you can say that a recording cannot reconstitute consciousness,
but the computing mechanism is not the calculation that corresponds to
consciousness.
Bruce
--
You received this message because you are subscribed to the Google Groups
"Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to everything-list+unsubscr...@googlegroups.com.
To post to this group, send email to everything-list@googlegroups.com.
Visit this group at http://groups.google.com/group/everything-list.
For more options, visit https://groups.google.com/d/optout.