Re: The view from the inside
- From: curt@xxxxxxxx (Curt Welch)
- Date: 30 Sep 2008 23:47:47 GMT
casey <jgkjcasey@xxxxxxxxxxxx> wrote:
On Sep 28, 2:26=A0pm, c...@xxxxxxxx (Curt Welch) wrote:
It's this issue of whether we believe the universe
out there exists, and has properties, independent
of our ability to think about it. Do you place
thought as the foundation of all reality, or do
you place the physical universe as the foundation
of all reality?
It is taken as a working assumption in science that
there is a Reality "out there" that exist independently
of any thoughts about it. And indeed our thoughts are
determined by part of the Reality out there.
To me, the model is clear. Humans are just machines
built out of the stuff of the universe. We are
exactly like computers in that what we can do, is
defined by, and limited by, what you can build from
the stuff of this universe.
Nothing extra added. Got that.
Some people clearly choose other beliefs - like the
mind first view of the universe, or dualistic views.
But I don't.
Dualism has no explanatory power. Either this "extra
something" has an effect, in which case it is physical,
or it is unverifiable make believe which, even if it
was true as a lucky guess, having no effect means it
doesn't matter if it is true or not.
As for the idea of the mind first view it doesn't
really fit in with the fact our own mind's world has
a brain perspective. If the mind was the only reality
what point is there in a brain?
Yeah, based on how well the rest of physical reality has been explained by
reduction, you have to answer the question of why we have a brain if the
mind is something more than just a brain. And if the brain is taking
advantage of some physical feature of the universe we don't know about, we
would have to also answer why science doesn't know about it since we have
gone to so great levels of abstraction below the biology of the brain
already and we don't seem to have any indications of "stuff missing".
Because we are humans, we are trapped in the middle
of this loop. Our understanding is limited to what
we can think about, and with our power to think, we
create an understanding of the universe, and in that
universe, we find a brain, which generates our power
to think. Being stuck in the middle here, it's hard
to figure out the fine line between mind and matter.
Of everything we understand, how much of the nature
of reality which we understand, is a produce of or
powers to think, and how much is a nature of the
reality itself? Is red in the mind, or out in the
universe for example.
It's not clear where the line is, or which is, or
should be, on the "bottom" of the reduction stack,
or if it even makes sense to consider one the
bottom, instead of the other, or whether we should
just see it as a circle with no bottom, etc.
However, I believe there is a physical reality out
there which is what everything reduces to. All
our thoughts, and our power to think, reduces to
properties of the physical universe, which means
they all reduce to atoms (and below) as well.
They may reduce to atoms and below but they are to be
understood at a higher descriptive level.
Yes, we work with the higher levels of abstraction to understand what is
happening at that level. But it's not because the lower level fails to
explain the higher level, it's simply because the higher level abstractions
simplifies the problem to only the aspects we care about at the highest
level. The high level abstraction is just a short cut. It's not a
requirement.
We can talk about the temperature of a gas, which is just an average
momentum of all the gas particles. But whatever we calculate using the
high level idea of average momentum, we could also have concluded by
looking at the behavior of all the parts using the lower level
abstractions.
We do have the problem that Alpha loves to deal with in that our low level
reductions tend at times not to be complete. And if they aren't complete,
then important stuff is missing, which can at times show up at higher
levels.
For example, if there was extra energy in the atoms other than it's
momentum which added to the overall temperature of the gas, then a
reduction based only on momentum wouldn't correctly explain what we say the
temperature of the gas doing. But this doesn't make momentum and invalid
reduction. It just means it's only a partial description of the lower
level. And when our lower level reductions are only partially correct,
effects can emerge that the low level reductions did not predict.
And if what emerges is something more than random noise, then we might be
able to use the high level effect. But in this case, that high level
effect is the lowest level reduction of the aspect of the universe.
So, if you have valid and complete reductions, all your higher level
concepts will be 100% explained by the low level reductions. You don't
have to work at the higher levels because all high level effects are 100%
explained by the low level reduction. We only use the high level effects
as short cuts to deduce the complexity of the problem.
But, if the reductions are not 100% complete, then there can be some
understand which exists at the high level, which currently has no reduction
to explain. That high level is then our lowest level understanding of the
effect.
Human behavior is full of this. We understand human behavior and give it
labels (folk psychology) at the high level, and we don't yet have a 100%
complete reduction to explain it. We couldn't predict what a human would,
even in theory, by simulating the neurons in their brain because our
reduction of human behaviors to the action of neurons is not complete yet.
However, these high level effects which have no current reduction to
explain them is what drives science. Science is the search for reductions
to explain everything that doesn't already have a reduction to explain it.
So any high level effect (War and Peace) which is not 100% explained, will
keep science busy until it is 100% explained (with the big gaps in that
reduction being the human brain and the complete history of evolution on
the Earth).
When you talk about things, anything, I always, try
to figure out what you are talking about, and map
your words, back to the part of the physical
universe I think you are talking about.
I take the physical Universe as a given. So when I talk
about the view or a Self the physical part is understood.
It is the configurations and behaviors that are of
interest. When you have thoughts they are not about
the parts that must exist for you to have those thoughts
it is about some higher level pattern or action.
Yes, but the actions of the parts and "the parts" are one and the same
thing.
The part is nothing more than the collection of all it's actions grouped
together under one name (the dog).
We can talk about a single action of a part separate from its other
actions, but the action is not something separate from the part. It's
separate from all the other actions of the part. But the collection of all
the actions a part can produce, is what the part _IS_. It has no other
form of existence separate from its actions.
You can call them actions, or properties, or attributes, but it all boils
down to the same thing. It's a collection of physical effects we have the
power to sense in the universe which we group together and give a single
name for.
Some attributes (aka actions) are long time and very persistent, and some
change quickly and easily. But all the attributes can change so there is
no single attribute which is the "thing" which is separate from "what it
does".
For a gear, it's rotational position and velocity are example of attributes
that change quickly and easily. Their persistence is very short term so we
call them actions. If you apply a little energy, the rotation changes in
an instant.
It's shape and volume (the space it consumers) is far more consistent so we
think of that as one of the properties of the "thing" vs "something it
does". But in fact, "occupy space in the shape of a gear" is just a
behavior. It's a behavior that changes in an instant with the application
of a little energy (melt it with a little heat which makes all the atoms
move to new locations).
With more energy, the atoms fly apart, and the properties which made it a
brass vanish and now it's just a cloud of vapor.
Apply some more energy and the sub-atomic particles break apart.
In the universe, there is no dividing line between what a think is, and
what a think does. That concept is yet another fabrication by man to
simply allow us to break a continuous spectrum of behaviors into rough
categories of long term vs short term behaviors.
Repeating again what you wrote above:
When you have thoughts they are not about
the parts that must exist for you to have those thoughts
it is about some higher level pattern or action.
I certainly relate to this type of thinking in that is how I think about
the operation of computers. I use high level abstractions that talk about
the behavior of the software. I talk about how a software module will
receive data and process it and produce data as a result. And though we
talk about "data" as if it were a long term persistent entity (like atoms
our for our level of existence) we know it's just the _behavior_ of the
computer we aer talking about.
The keys on my keyboard have long term persistence, but the keystrokes I
make to "enter some data" are short term behaviors. It's very easy for us
to say the keys are "things" and the "keystrokes" are actions. But in
fact, the keys are just the behaviors of atoms. And if our eyes could see
what was happening at the atomic level, the "keys" would look more like
complex machines with billions of parts all in motion, then they would look
like "a thing".
To claim as you did above, that the thoughts are "about" the actions, but
"not about" the "parts" would only be a valid concept if "part" and
"action" were not one and the same thing - but they are. So I can't agree
that it's valid to try and insist that "thoughts" are "actions" and not
"things" at the same time.
When we look at the operation o fa computer, we do pick out _some_ of the
actions as being special and interesting, and ignore the rest. If I tap on
a key but don't push it hard enough it won't cause a signal to be sent to
the computer. So "key taping" is not "entering data" even though it very
much is a behavior. But if I push it far enough, then the behavior
suddenly becomes "data".
If the electrical signal on a wire in the computer is jumping up and down
at a low level due to radio frequency interference but it doesn't go high
enough to cause the transistor circuit to flip output states, then we don't
call the voltage changes "data". But if the electrons more far enough, we
suddenly call it "data", just like with the keyboard.
All behavior is not "data" behavior. But this is only because of the high
level abstractions we use to describe the behaviors of the computer we are
interested in - the behaviors that help us predict the future behaviors of
the computer.
So even though we use high level abstractions to talk about some of the
behaviors of the machine we are interested in, that does not mean that all
the behavior is the "data" and all the "things making behavior" are not
data.
This distinction between "thing" and "action" is not something that exists
in the machine, it only exists in our how we like to think about the
machine. It's an artificial view we apply to the computer which is useful
in helping us predict what it will do in the future, but it's not a very
accurate description of what it is actually doing.
And as such, if you try to apply the same abstraction to explain the
oddness of consciousness, you are getting sucked into an illusion that
things, and actions, are separate, when they aren't in fact separate in the
physical world.
I find it hard to follow you, because I don't know if
you are talking about brain behavior when we think
about rocks, or if you are talking about the actual
rocks.
The context should make that clear.
When we talk about putting a collection of gears together
and making a clock, we can say a new behavior emerges.
The set as a whole can keep time. But that doesn't mean
the parts are doing something new. They are not. They
work the same way whether they are in the clock formation
or whether they are in another formation.
It is the whole that does something new. There was never
any suggestion the parts did anything beyond what they have
always done.
It can also depend on how those parts are connected.
Yes, but that's not something it can _sometimes_ depend on.
The behavior of the gear is _always_ a function of it's
relation to the rest of the universe.
You seem to be unnecessarily pedantic on these things. What
I was alluding to was a collection of clock parts don't do
"clock behaviors" until they are connected up the right way
and then it is the whole not the parts that is doing the
"clock behavior".
Well, the entire concept of "the whole" is just a high level abstraction.
When we switch to thinking about something at a higher level of
abstraction, it's true that we use concepts of "whole" as we think.
What I have problems with, is that you you seem to want to deny the fact
that when we say "whole" we are still talking about the behavior of the
parts. We aren't talking about "something new" that "didn't exist" before.
We are just using a short hand notation to make reference to multiple
parts.
It's this whole debate about Strong emergence Don and I have been having
with Alpha here and the subtle but important implications it all has to
ideas like consciousness (which is where all this gets important). When I
use the high level abstractions of software to help me program a computer,
none of these semantic details are in the least bit important or
interesting. But it's when we try to explain and talk about consciousness
where the subtle meaning of this becomes highly important.
When I use the words and ideas of computer programming, am I talking about
something that doesn't exist in the atoms that make up the computer or not?
My claim is that I'm not. I'm just using the idea of "program" and
"subroutine" and "data" as high level concept to talking about a lot of
atoms at once.
When we use the high level concept of temperature are we talking about
something that doesn't exist in the atoms? Of course not. The temperature
of a collection of atoms is just the average momentum of the atoms. The
word "temperature" is just short hand notation for "The sum of the momentum
of each atom at time T divide by the number of atoms multiplied by a unit
adjustment factor" - or some such thing. Temperature fully reduces to
lower level concepts about atoms. When we use the high level abstraction
of temperature, we are just talking about atoms.
There is no validity to the idea that when we talk about the temperature we
are not talking about atoms.
And there is no validity to the idea that when we talk about a clock, that
we are not just talking about the parts that make up the clock.
There is no way possible to take away the parts of a clock, and still have
something left. To assume there is "something else there" that doesn't
exist in the parts, would require that we could take the parts away, and
still have something else left to talk about.
When we talk about the behavior of crowd, we are _still_ talking about the
behavior of people. The fact that people, and clock parts, are able to
produce new types of behaviors when they are in a crowd, doesn't mean we
are no longer talking about the behavior of people, or clock parts.
You keep witting things that seem to imply that these new behaviors are
something more than just the behaviors of the parts. And that, again, fits
in with what I see as the error of conscious - the believe that something
"new" emerges from the parts other than the behavior of the parts.
This is exactly what Alpha and other people that think strong Emergence is
a valid concept are clinging to. They are looking for a way to explain why
consciousness seems to "emerge" as something new and different from the
brain when it didn't exist in the parts.
A ball by itself, floating in space, can only travel in a straight line.
That's the only behavior it can do on it's own. But put two balls in
space, and all sorts of new "behaviors" show up - they can orbit each other
for example. They can work together to produce new behaviors. But it's
still the _ball_ producing the behavior. It's still the ball which is
orbiting.
If you put a lot of atoms together to make a human body with a brain, even
more new effects show up. The fingers can now type Usenet messages on the
keyboard of a computer.
But there is nothing new there other than the fact that the path though
space the atoms are taking, are now more complex than when the atoms were
on their own.
In other words, if you use the word "consciousness" to describe some aspect
of the "whole" then the only aspect that is there to be described, is the
path the atoms that make up the body are taking as they move through space.
And as such, you are still talking about the atoms and what they are
currently doing and nothing more. The reduction of matter to atoms is a
100% complete reduction. There is nothing else about matter which is lost
by that reduction - at least not at our level of existence here on Earth.
The problem with consciousness as I've explained many times, is that it's
an idea that comes from people's incorrect self image of themselves. They
hear this voice in their head, the voice doesn't seem to be a phsyical
effect, and they identify this voice as being a part of themselves. But
since this part of them doesn't seem to be phsyical, it creates the natural
belief in most people that they exist as body combined with this odd
floating voice that's not part of their body but somehow associated with
it.
From this error, comes a never ending list of made up stories to justifywhy our conscious self (the floating voice in our head which isn't a
physical effect), somehow "shows up" as the result of brain behaviors.
It's why Alpha and his friends cling strongly to this Strong Emergence
crap. It's because the think they need something like to to explain
consciousness. But you don't.
The fact that the "voice in our head" doesn't seem to be physical, is an
illusion. The non-physical nature of the voice is an illusion created by
the simple fact that brain can't out how to built the correct associations
to effects in the physical world.
We know the correct association however. It's brain activity. But knowing
the association, isn't enough to cause the low level brain hardware to
actually wire those associations into our thinking.
When we hear a dog bark, the brain has associations wired into it that make
us think about what a dog looks like, and where it might be (standing on
the ground instead of flying through the ear), and about how large it
probably is (dog sized instead of elephant sized). All these phsyical
attributes suddenly activate in our mind as a result of the stimulation of
hearing a dog bark because the brain has wired those associations between
those attributes.
When we hear the voice in our head, there are no associations like that.
How big is the thing making that voice? Where is it located? What does it
look like? How does it move through space? Nothing. Nothing like that
pops into our mind when we hear the voice in our head talking about
something like it does when we hear a dog bark.
Because the brain hasn't wired such associations, it leaves us with no
phsyical grounding of that voice in our head. It floats in some undefined
space not connected to all the stuff we think of as the phsyical world.
And this is why the voice seems to be separate from, the body.
However, the voice is phsyical brain behavior. When we hear that voice in
our head, we should be having thoughts of specific neurons firing in some
specific part of our brain. If we knew how our own brain actual worked,
that's exactly what we would be thinking. If we had the option of using a
high resolution brain scanner, we would get to know all the different parts
of the brain that are associated with that voice in my head. We would be
able to look at the specific neurons and understand what part they play in
the "voice in our head". If there was one neuron on the left side of my
head in the temporal lobe that activated whenever I had a thought of "dog",
then every time I heard the voice in my head talk about "dog" the
associations that got wired in my head would make me think about the "dog"
neuron in my head which I new was firing.
With enough exposure like this, all our "mental activity" would become the
actions of neurons, and NOT something "new" that emerged.
Alpha is just totally lost. He is clueless. He knows lots of facts, but
understands almost nothing. He's using the wrong abstractions, and instead
of fixing the abstractions, he just piles on more and more errors on top of
the previous errors making things endlessly more complex when they were
never complex to start with.
When you keep trying to cling to these ideas that the stuff that exists at
the high level doesn't exist at the low level as well, I just sense it's
more attempts to justify the illusion of conscious - how consciousness
seems to "emerge" as something separate from the phsyical body.
The only thing that changes when you put a lot of atoms together to make a
body, is that the atoms can do things together that they couldn't do alone.
But other than that, nothing "else" is emerging. The effect that makes
consciousness seem to "emerge" is not a physical effect at all. It's
nothing something we explain as a "new" phsyical effect. It's not new at
all.
To mystery of consciousness is really just the mystery of why the voice in
our head doesn't seem to be a physical effect, like hand waving does. And
the answer to that, is not Alpha best answer which is "Strong Emerges"
(which is total bull shit). The answer is far simpler. It's just that he
brain hasn't wired up the correct cross connections between the brain
activity that represents that voice in our head, with the brain activity
that represents physical events. Strong emergence is a stupid way of
trying to explain a missing neural connection - but that's exactly what it
is and why it's so completely stupid.
Your talk of "thoughts are about behavior but not about the "thing"" is
just more grasping at straws in my book to create a explanation of that
missing neural connection.
All meaning translates back to the physical universe.
This is a fact of physicalism. If you believe, or just
understand physicalism, you understand this.
I don't like ISMs of any kind. Either something has an
effect on you or it doesn't, end of story. As for "meaning"
I see that as something you give to something else, it is
a relationship between you and the object. Thus an apple
is as you say a collection of atoms but to you it might
mean food.
Sure. But what is it in you that gives you the power to "assign meaning".
To say it's something we assign, just doges the question of what meaning
is. It doesn't explain what it is.
What is the "view from the inside" in terms of the
physical universe? What do you think that means when
translated down to atoms? (meaning, explain it to me
by using atoms).
Atoms are the wrong explanatory level.
My point here is that they are not the _wrong_ level. They are simply a
level which hard to work because the details of teh behavior of each atom
creates create total complexity. It's not that you can't explain it, it
just that it's harder to explain if you include all the detail at the
atomic level. But if the reduction is correct, we should be able to
explain everything at that level - and I think we can.
Don and I have both shown Alpha how to explain W&P using the reduction of
atoms. But it goes in one ear and out the other because Alpha believes
there is something else that needs to be explained that we aren't touching
on. It's more than just the fact that we can't fill in every little
detail, he thinks we have failed to explain where the conscious meaning
that exists in a book like that comes from.
And in case you don't get it, I pick "atoms" not
because I think that's some "bottom" of the reduction
chain but because I think that's a level low enough to
make it clear I'm not talking about large structures
like neurons or humans brains or computers or rocks or
clocks. I'm talking about a level which is below all
those things in the reduction chain.
I think your whole emphasis on it is "nothing but atoms"
simply started out as it isn't non physical.
Yeah, that's basically it. The "nothing bot atoms" means you _can_ explain
it all at the level of atoms, but we don't simply because it's too much
detail.
And I have
never disagreed with that. You can always reduce the
behavior of a computer to the behavior of logic gates
but for most high level behaviors that would be a
pointless excess in understanding the behavior.
Right, it would be too much detail. It's not that something is missing (as
Alpha argues), it's that it is just too much detail. So we use higher
level abstractions that leave out most the details and only touches on the
issues we care about at the high level.
He [Glen] has never said one thing in any of his posts
that struck me as dualist in nature. But yet, you
can hardly write a single post on subjects related to
consciousness without me suspecting you are making
dualistic errors of thought without even realizing you
are doing it.
There IS a dual view as you admitted yourself when you
explained it correctly as thoughts seeming to be without
any physical basis but in fact are the physical activity
of a neural network.
Sure, there's the illusion of duality which I've talked about many times.
But that's just missing neural connections, not a fundamental aspect of
nature.
If you had a computer program capable of philosophizing
but not having knowledge of its own logic gates I would
suggest it would also wonder about the nonphysical nature
of its own thoughts.
Well, it would depend on those missing connections. If it was built much
like us, and likewise, didn't have the ability to probe it's own brain with
physical tools, then the same illusion would develop in it, sure.
But, there's something else here that you might not grasp. The illusion is
almost completely gone from me, and I didn't have any such tools to work
with. I only had logic and reason to guide me. But with that, I was able
to figure out what the truth was, and the more time you spend thinking
about your thoughts as being physical actions, the more sense it makes, and
the more of those "missing connections" get filled in. The result is that
you stop thinking of yourself as having this non-physical consciousness. I
have actually been able to condition myself to see myself differently.
Though I didn't realize it until people here made me think about it, I used
to think of myself as this conscious "self" located somewhere "inside" a
physical body. I had this distinct self image of being these two things at
once. I was both a body, and this "conscious self" which existed inside
the body.
For example, when I get dressed in the morning, there's this chair I sit in
to put my socks and shoes one. And all too often, after putting my socks
and shoes on, I get distracted and start thinking about things. 15 minutes
later, I realize I really should get on with my day and get up.
But in the past, when I would do something like sit in a quite room in a
chair like this, the "I" which was doing all this "thinking to myself" was
the thing _inside_ the body. It's like "I" would "let my body relax" while
I went of talking to myself about something. But the "I" doing the "taking
to myself" was distinctly something separate from my body. That was the
natrual self image that had developed in my.
But the other day, I found myself in that chair, and I was thinking about
these AI debates, and I realized, I no longer thought of myself as this
"conscious self" separate from the body. I was simply a body sitting in a
chair with it's brain busy doing stuff. And I had the thought, "Gee I used
to think of myself as a conscious self inside a body - what happened to
that"? I had to focus on that idea for a while until suddenly, I got my
mind to flip back to that frame of reference again, and remind me what it
was like when I used to think of myself that way.
It worked exactly like those optical illusions where it can take one two
forms, and your mind will flip back and forth from seeing it one way, or
the other, but you can't really see it both ways at the same time. One
instant, my self image was of a body in a chair, with nothing else there,
the next instant, it split into two again, and I was a "conscious self"
located "inside" the body, then it flipped back to just being a body. I
now have to mentally concentrate to turn my self back into a "dualistic"
person.
This illusion of duality isn't just "like an illusion". It's an honest to
god real illusion. Most people think of themselves as this odd "split
being" where their conscious self is separate from the physical self. But
like with many illusions, you can train yourself to see things differently.
The fact that I've trained myself to no longer see myself as this "split
being" shows how this is all just a question of perception and not some
fundamental property of the universe.
I simply don't see myself as a mind and a body anymore. I'm just a
phsyical body interacting with my physical environment.
The people who are still confused about consciousness (like Alpha) are the
ones that not only see themselves as a split body, but they have also
chosen to believe that this split nature of our self image is some
mysterious attribute of the universe, instead of understanding it's only
the result of a few missing neural connections in the brain.
The brain doesn't "create consciousness" by some magic advanced biological
function we don't understand yet. It "creates consciousness" by _NOT_
making any neural links between the neurons that represent "self talk in
the head" to "location in space" neurons.
In a computer, this illusion would be created by changing the value of a
few pointers in memory. That's all the mystery of consciousness is. It's
just a few missing pointers in the database of associations that represent
the nature of reality for us.
What the program is aware of, what the program reacts to,
is not the i/o of the logic gates that the program IS, but
rather the objects (or data) of the program. You confuse
the reaction of the program at a high level with the
reaction of its parts.
Bingo! You do it again. I can't tell if you know what you
are saying, or if you are just confused. Never once has
Glen left me confused, but when I read the words you write,
like the ones above, it sure the hell leaves me feeling you
don't really understand what you are saying.
Do you understand that when we say "the program" we are in
fact talking about the behavior of a big pile of atoms
known as a computer?
Computers are made up of atoms sure. The atom's behaviors are
independent of the computer however.
Well, again, this is more examples of you talking total nonsense in my
eyes.
The atoms are not separate from the computer. If you take away all the
atoms, what do you have left? NOTHING. If you take away the _behavior_ of
all the atoms, where is the computer? Again, NOWHERE. The only behavior
that exists in the computer, is the behavior of the atoms that make up the
computer.
All the words we use to talk about the behavior of computers are simply
words we use to talk about the behavior of a lot of atoms at the same time.
The computer's behavior
is dependent on the behavior of atoms but we really don't have
to reduce our explanation to that level.
Yes, we don't HAVE TO, but we can't do what you just did above, which was
to try and declare "the atoms behavior are independent of the computer".
They are not in any sense what so ever independent of the computer's
behavior.
It's as silly, and misguided, as trying to declare that the behavior of the
solar system is independent from the behavior of the planets. It's just a
stupid thing to say or think. The behavior of the solar system _IS_ the
behavior of the planets. The behavior of the computer _IS_ the behavior of
the atoms of the computers.
The program is a high level description. You can understand
how a program works without reference to logic gates or machine
code and indeed such a reduction, although possible, would
not add one bit to the understanding of the program itself.
Right, because when we use a high level description, we are leaving out
most of the details. We aren't talking about something new that doesn't
exist at the low level, we are doing just the opposite. The high level
abstraction is one which ignores most the details, and only talks about the
ones we care about.
It's exactly the opposite of what Alpha argues. Nothing new "emerges" at
the high level, at the high level, we are actually just throughing out and
ignoring most of the details of what is really happening.
I can say something like, "the program sent a packet to the
server", or I can say, "the configuration of the atoms in
the computer caused the voltage to fluxate in the network
cable which led to the atoms in the server to be re-configured"
...
The only thing that changes in the picture is the words
flowing out of the human standing next to the computer.
Nothing about the computer changes.
I never suggested it wasn't anything more that a human's
(or an AI's) high level description of the process. It is
an appropriate level at which to describe the process.
Yes, but I would argue that it's not "appropriate" as much as it is simply
"useful for what humans need to deal with".
So, above, you said the very odd phrase "what the _program_
is aware of".
The word "program" is just a word we use to talk about how
a computer is configured and what sequence of events it's
likely to produce because of how it's configured.
What on earth does "awareness" means in this context?
The context is a high level description of a process which
behaves in a way we see as it being "aware". And what it is
aware of is also in the form of a high level description.
Ok. I guess that's good enough.
I know what I think it means, but from the above, I can't
grasp what you think it means.
What the atoms are reacting to, is all the other atoms.
End of story.
Paint, Word, Excel are all the reactions of logic gates.
End of story.
When I use words like "the computer is aware of the
packet received from the network" it means something
to me as a computer programmer, but that word (aware)
gets very dangerous in the context of the topics
of this thread.
And you as a computer programmer are nothing but the
activity of neurons which understands these things at
the level of "packet received from the network". The
activity of neurons is designed to work at that level.
That is the level of the awareness of the whole.
I'm simply using the word "the packet" as a label for
the behavior of a billion or so atoms.
But you only know that because you know about atoms.
You don't actually experience the atoms you are only
able to cogitate their existence from science.
True, I wouldn't know about the atoms if I wasn't taught that (or did the
science to discover it on my own). But the point is, if someone knows about
a solar system, they do in fat "know about what the planets are doing",
even if they didn't know the solar system was made up of planets.
That lake out there is a collection of atoms. I might not know it as
atoms, but atoms are what I'm aware of when I see the waves in the lake.
Humans or AI programs can know things by "working them
out" or by "experiencing" them as output from some lower
functioning unit which has done the "working out".
It makes no difference which words I'm using, I'm still
talking about what the atoms are doing in all cases.
That is fine in the sense I believe you go on about atoms
to emphasize there isn't anything extra added. But it makes
a big difference in our ability to understand how all those
atoms are working to get the results we observe.
Right, but only because our minds are far too weak to deal with all the
details. If we had "eyes" that could see at the atomic level, and a brain
big enough to understand what billions of atoms were doing at the same
time, then we would not need to throw most the data out and only use the
high level abstraction.
We get there is nothing extra added but there is a level
of explanation that can be followed and one that can't be
followed even if both explanations "nothing but atoms"
and "packets" refer to the same thing.
The only reason the "nothing but atoms" level can't be followed, is because
our brains are too small. We are too stupid to do it. It's not because the
answer doesn't exist at that level. We use the high level abstractions as
a way of reducing the amount of processing we have to do in order find
solutions to our problems.
It's the same reason why a chess program can evaluate millions of moves and
humans can only evaluate 50 in the same time. The comptuer is far better
at dealing with details than the human brain is. The only reason the brain
is as good as it is at playing chess is because it's so good at building
useful high level abstractions which allows the brain to do a lot more,
with far less processing power.
We don't use the high level abstractions because that is the only level the
answer can be found, we use it because our brain is too weak to make in
progress if we used the "check every move" approach.
It makes no difference if we use words like "computer
program", or "pain" or "running", we are always making
reference to atoms somewhere whether the person using
the words understand or believes that or not.
That may be true but what is the point? I understand your
need to say there is nothing extra but really the use of
high level words doesn't mean that the user believes there
is something extra.
Sure, but I've pointed out what, maybe 3 things you have said in this
reply, that I see as total nonsense. And it's the result of you trying to
argue the position that something "exists" at the high level which doesn't
exist at the low level - that we _MUST_ use the high level abstractions
because the answer doesn't exist at the "nothing but atoms" level of
abstraction.
Let me just quote one of the things you said above just ot make it clear
why I keep going on about. You wrote:
> The atom's behaviors are independent of the computer however.
That is a simple example of why I keep telling you don't fully get it yet.
If you did, you wouldn't make a mistake like that. It's nonsense to even
suggest the the atom's behavior are independent of the behavior of the
computer. The atoms ARE the computer and what they do is what the computer
does.
... you seem to want to believe in materialism.
Something effects you or it doesn't. Call it what you like.
what the program reacts to, is not the i/o of the logic
gates that the program IS
What the program reacts to is not only the I/O of the
logic gates, it's also the atoms that make up the logic
gates and the electrons flowing in the wires. It's
very much reacting to all those things.
It comes back to an appropriate level of explanation.
If you ask the program what it is reacting to and it
doesn't have a description of the subject at the level
of electrons or logic gates it will say it is reacting
to, being aware of, x, where x is some high level
description of what it is reacting to. The fact you
might have some deeper insight to the lower levels
doesn't change that fact.
Right. But, my issue with some of what you have written are not about
whether one level of abstraction is used, or useful. It's about the links
between the levels of abstraction that you seem to make the errors on - and
that's the basis for all the debate Don and I are having with Alpha on
emergence. It's how these levels all link together (or don't link
together) that we are debating here.
If "consciousness" was just some "high level abstraction" and we didn't
care about how it was linked to "atoms" and other levels of abstraction,
then we would have nothing to debate. But in trying to solve AI, we need
to know exactly how all these abstractions link together - and if they can
be linked together or if we might be forced to keep them all separate.
Alpha argues for lossy downward links along with strong emergence. This
allows us to believe that consciousness is something "new" that emerges
from the brain but which doesn't exist in the atoms. It allows us to
believe that when we talk about consciousness, we are not just talking
about atoms.
I reject those beliefs and argue that when we talk about what happens in a
human, we are in fact always talking about atoms. The fact that we might
use high level abstractions of "pain", and "desire" and the like doesn't
mean that "pain" and "desire" is something other than atoms.
For the most part, we don't have to deal with these connections between
levels of abstractions. For each problem we have to solve, we pick the
level of abstraction which works best to solve that problem for us and just
ignore all the other levels of abstraction.
But to get a good grip on complex AI questions such as "what is pain" and
"what is consciousness" we not only have to face these questions, we have
to get all the links between levels correct, or else we have little chance
of solving AI in any sense.
So I can't grasp your point here because at fact value,
it's inconsistent with materialism.
If you mean it is inconsistent with there being nothing
extra added I don't see why. We are talking about effects.
You can talk about a car hitting a brick wall or you can
talk about one bunch of atoms hitting another bunch of
atoms what ever takes your fancy. Which do you find the
most informative in those two descriptions?
Right. It's just a question of what is useful for whatever problem we as a
human are trying to solve, not a question of which one is a correct
description of what happens. The atoms are a more "correct" description
because it's a more complete description (leaves out less of the true
details). But for staying alive, "car hitting wall" is a level that often
works better for helping humans stay alive.
When we talk about what our computer programs are doing,
we totally ignore all the other levels.
Correct.
But in this thread, we are very much debating the true
nature of all of reality, which includes understanding
what it is we are actually making reference to when we
talk about our computer software and what it is doing.
I don't know what the "true" nature of Reality is. If it
effects you then it exists in some way. If it doesn't
effect you then it doesn't matter if it exists or not.
You just keep writing these things which strikes me as
odd and inconsistent with your other beliefs like you
did again in this message, when you wrote:
"what the program reacts to, is not the i/o of the
logic gates"
When in fact what the program reacts to IS the I/O of
the logic gates.
No, I would say at a low level description the gates react
to each other but at the high level description, a program,
is reacting to high level data.
We talk that way because it conveys understanding. If I
was to say the program is reacting to square shapes that
would mean something, but if I was to say it was reacting
to its logic gates that wouldn't convey any useful
information.
But if I were say the program is reacting to a pattern of logic gates where
gate A was high, gate B was high, gate C was log, etc.... Then it would
convey all the same information and more. the low level description
doesn't convey less information, it conveys more. Abstractions are data
filters, they don't add more information.
If the program was aware of the logic gates, of which it
depends to exist at all, it could talk about them.
Well, my point is that it is talking about them whether it understand it is
talking about them or not. And the same for you. If you think you are not
talking about atoms when you talk about what your computer software is
doing, you are in fact talking about atoms whether you understand that or
not.
And when someone talks about consciousness, they are also talking about
atoms, whether they understand that or not.
This will be my last post as I will not have access to
a computer for a few weeks.
Well, I hope you are doing something fun with those few weeks!
JC
--
Curt Welch http://CurtWelch.Com/
curt@xxxxxxxx http://NewsReader.Com/
.
- Follow-Ups:
- Re: The view from the inside
- From: Alpha
- Re: The view from the inside
- Next by Date: Re: Abiogenesis
- Next by thread: Re: The view from the inside
- Index(es):
Relevant Pages
|
