Re: Problem for physicalist evolutionists

On 27 Mar, 13:35, someone2 <glenn.spig...@xxxxxxxxxxxxxx> wrote:
On 27 Mar, 03:46, Nic <harrisonda...@xxxxxxxxxxx> wrote:

On 27 Mar, 03:03, someone2 <glenn.spig...@xxxxxxxxxxxxxx> wrote:

On 27 Mar, 02:38, Nic <harrisonda...@xxxxxxxxxxx> wrote:

On 27 Mar, 01:45, someone2 <glenn.spig...@xxxxxxxxxxxxxx> wrote:

On 27 Mar, 01:26, Nic <harrisonda...@xxxxxxxxxxx> wrote:

On 26 Mar, 23:48, someone2 <glenn.spig...@xxxxxxxxxxxxxx> wrote:

<snip>

That any consciousness would have to be epiphenomenal is only argued
in the article with regards to the following physicalist story:

In principle, if the physical state of reality were known, and any
rules in terms of the physical which governed interactions within
reality were also known, then the change in state of reality over time
(the behaviour) could be explained.

Yes you do argue the necessity of epiphenomenalism from that.  Well
that plus one other premise: that a behaviourally adequate thing (say
a robot) may or may not be conscious - so something which is not
causally efficacious must account for the difference.

What you seem to miss is that the difference you are explaining isn't
a testable one, or noticable in any way.  Your conviction that there
is a difference, I guess, can only come from the imaginability to you
of zombies.

What several people are pointing out to you is that functionalists
assert that consciousness is causally efficacious, and deny zombies.
That is why functionalists don't see a problem.

Functionalists may have a bit of a problem when they come to your
fairy lights example, and that was the problem I was trying to
sidestep in my original post last week.

The predictions of the robot's behaviour are the same whether theory
F, theory R, or theory C, as they aren't in disagreement about the
build of the robot, or the known laws of physics which the robot isn't
deviating from. Therefore though one of the theories might be right,
none are suggesting a difference in expected behaviour in the robot
which would allow for an experiment, all believing the build following
the known laws of physics is sufficient to explain the behaviour.

I am glad you understood the representationalism problem illustrated
with the fairy lights.

While functionalist can regard consciousness linked to function, and
therefore there could not be a zombie, as if it had the function, they
would claim that it would have the consciousness or first person
perspective. There is still the issue of what we are talking about
with regards to a first person perspective, as illustrated in the
Terminator movie for example,

I think I know the scene you mean.  You are saying you'd be able to
tell if the robot was colour blind or not from that.  It is fiction,
and so it is allowed to switch between first person and third person.
What you saw in the film was an authorial device, not a fictional
robot mind-reading scene.  The latter would look completely different
- e.g. they might have to bring in a fictional monitor and put a
fictional jack plug into his fictional ear.

and there is still the issue that
functionalism isn't the only theory, and how they would expect the
robot to behave if they were to be considering the null hypothesis to
their theory. As the behaviour is the same with their hypothesis and
the null hypothesis, it still illustrates epiphenominalism for whether
there was a first person perspective or not.

No sneaking it back in again with this 'null hypothesis' talk!  What
exactly would that be?  It would be the hypothesis that implementing a
particular function does not implement a quale.  Under that
hypothesis, there isn't such a thing as a first person perspective.
Which if you recall, is my 4th option, in which we disregard the only
evidence we have: that from Nic's own case.

Point taken though that
the proponents of functionalism don't like to consider how the
expected behaviour if their theory was wrong, I guess they might
consider such reflections to undermine their authority their theory
has, as it would highlight, at best the lack of evidence backing it up
(none whatsoever), and at worst highlight flaws in the conception.
Certainly a lot of people that have been writing here claiming to
agree with functionalism, are also certain that their having a first
person perspective affects their behaviour.

Well though the film was a fiction, it illustrated what we would be
considering when considering whether the robot had a first person
perspective or not. Yes the null hypothesis would be that it didn't
have a first person perspective at all. The functionalist might claim
that the first person perspective is linked to function, and that if
the robot had certain functions, that it would have it. The null
hypothesis is that it wouldn't. The expected behaviour for the robot
would be the same, for both the hypothesis, and the null hypothesis.

Ah - you are seeing functionalism as a *scientific theory*.  It isn't
- it is an analysis.

An example of a scientific theory is what Crick and Koch do.  They
make claims like certain firings in layer V1 of the mammalian visual
cortex are the correlate of conscious visual perception.  That (they
hope) is a falsifiable claim.

Thus with no behavioural difference predicted, whether or not there
was a "what it is to be like", or a first person perspective would be
shown to be irrelevant to behaviour, even from a functionalists
perspective. I can understand them not liking to contemplate their
theory being wrong, though actually out in the three theories outlined
in the paper, the objections in the other two, were objections to
functionalism that I don't see how it can escape from. Either way,
they can hardly say it is forbidden to ask what they would be
expecting the behaviour to be like if their theory wasn't the right
one.

Functionalism, if you want to treat it as a theory, is a theory of
what certain third person ascriptions mean.  Its falsification would
not come from the field of neuroscience, rather it would have to come
from how ordinary people (who don't know any neuroscience) actually
talk about each other's inner thingies.

It is better not to treat it as a theory but as a framework of
definitions mapping everyday terms to computerese ones.

[The point about the colour blind test comes in after this has been
accepted. As the implications above are that "what it is like" to be
you would have to be epiphenomenal to your behaviour, yet your
behaviour is reporting understandings that are only understood in
terms of what it is like to be you. Suggesting that "what is is like"
to be you, is affecting your behaviour. Thus the metaphysical
assumption where, in principle, if the physical state of reality were
known, and any rules in terms of the physical which governed
interactions within reality were also known, then the change in state
of reality over time (the behaviour) could be explained, would be
wrong. As the theories that sit on it, such as Theory R, C, and F
predict "what it is like" to be you to be epiphenomenal,

It is you who has sat R and C on F.  R and C are naive - which is why
I felt the need to go and check, as we can't have Crick marching
behind the functionalist banner if he believed what you seem to think.

and it
clearly isn't. Given a thousand different suggestions of what it might
be like to be the human, you can point out which one it is, and know
it to be true, and say so]

The reason functionalism wouldn't be regarded as a scientific theory
though is that as you say it wouldn't be falsifiable. It wouldn't be
falsifiable though, as the behaviour would be expected to be the same
whether it was true or not. Which is my point.

Eric Rowley has just explained a lot better what I was trying to say.

Therefore, as the suggestions would be that having a first person
perspective would not influence behaviour, to be considering that
people could talk about what "it was like to be them" would be to be
considering the metaphysical assumption:

In principle, if the physical state of reality were known, and any
rules in terms of the physical which governed interactions within
reality were also known, then the change in state of reality over time
(the behaviour) could be explained.

Which again, is my point.

(By the way Theories R and C don't sit on F, they object to it. All
are sitting on the metaphysical assumption mentioned above)


.

Relevant Pages

  • Re: Problem for physicalist evolutionists
    ... reality were also known, then the change in state of reality over time ... build of the robot, or the known laws of physics which the robot isn't ... the known laws of physics is sufficient to explain the behaviour. ... functionalism isn't the only theory, and how they would expect the ...
    (talk.origins)
  • Re: Problem for physicalist evolutionists
    ... build of the robot, or the known laws of physics which the robot isn't ... the known laws of physics is sufficient to explain the behaviour. ... functionalism isn't the only theory, and how they would expect the ... Well though the film was a fiction, it illustrated what we would be ...
    (talk.origins)
  • Re: Problem for physicalist evolutionists
    ... reality were also known, then the change in state of reality over time ... build of the robot, or the known laws of physics which the robot isn't ... the known laws of physics is sufficient to explain the behaviour. ... functionalism isn't the only theory, and how they would expect the ...
    (talk.origins)
  • Re: Problem for physicalist evolutionists
    ... how would you show that a robot didn't have ... terms of the physical which governed interactions within reality ... has a FPP is that it acts like it has a FPP? ... is no dispute about the known laws of physics, ...
    (talk.origins)
  • Re: Problem for physicalist evolutionists
    ... principle, the physical state of reality were known, and any ... knowledge of all physical laws): ... could suggest that there is a "what it is like" to be the robot ... the known laws of physics, or how the robot is built, how would ...
    (talk.origins)
Loading