gmane.science.physics.fabric-of-reality

Re: Joining post: 20 Years of a Science of Consciousness

Bruno Marchal — 2012-05-25T09:37:16


If the idea "I am conscious" is false, and asserted by the machine, it  
is not a first person idea.  But consciousness is a first person  
notion, so here such machine does not acquire the idea that he is  
conscious. It might utter "I am conscious", but without giving any  
meaning to the sentence. The first person idea of consciousness cannot  
be said, in that case, to have been acquired.

Conscious people, on the contrary, are incorrigible on consciousness  
(like on knowledge if you accept the standard definition of knowledge,  
where only a truth can be known).





We might believe that, but it is not knowledge (with the standard  
sense). It is conjectural 'knowledge'. (= belief, in the standard  
sense).




For creating belief. For knowledge, my feeling is that true  
referential (or self-referential) beliefs might be responsible for  
(self) consciousness.




My current feeling on this, is that all universal number are  
conscious. It is the starting initial point of the differentiating  
consciousness flux in arithmetic. I think that this might be a simple  
consequence of the computationalist hypothesis. But I am not sure  
about this. This comes from drug reports with total amnesia, including  
report of total amnesia about time, space, everything, yet completely  
conscious state. I need more data and work on this.




It was just a counter-example of the idea just proposed in the thread.



Why?
Augustin said, about time, that it was one of the clearest idea he  
got, but that he was completely unable to explain what it is to another.
Children have clear idea about line and plane, but cannot usually  
convey them.
In psychology, many things are clear yet non-communicable. Are you  
able to doubt that you are conscious right now, without lying to  
yourself? I have never been able to do that, neither when awake, nor  
in dreams.
Of course I can doubt that I am awake, but I cannot doubt that I am  
conscious, if only because for being doubting, I have to be conscious.  
If I am not conscious, I cannot doubt (I might fake it, but that is  
not a doubt).





We cannot define it precisely, but we can find terms which convey  
partially the idea, when asserted to people who we suppose are already  
conscious. We can point on may examples, like saying that if you have  
a headache, that is an example of conscious experience, or if you feel  
some pleasure when eating a fruit.




The problem of the role of consciousness,




No, we might not been able to do that.



That would.




Yes. Not self-conscious, plausibly, but conscious.
Octopi and jumping spider might already be self-conscious, imo.



You are reversing the inquiry. Consciousness is the fact that I want  
to explain. Not a tool to explain other things. Yet, it can do that.  
Indeed, consciousness, according to the theory comp, might explain why  
we believe that there is a physical universe, as the result of the  
consciousness flux which exist in arithmetic (by computationalism).




A proof made in a theory that you feel intuitively as very reasonably  
correct. It convinces you at first sight.




Nothing can be proved about reality, but proofs can be given in the  
frame of the conjecture itself. I can prove to you that there exist  
prime numbers, in the theory which assumes 0, successor, and the  
recursive laws of addition and multiplication.
Since Gödel, we know that proof is of the type belief (or conjecture).  
It automatically means that proving is always conditional or/and  
conjectural, like it is for any kind of communicable third person  
statement. So I am certainly not flying in face of that, quite the  
contrary. Proving just does not always entail truth.

- Bruno Marchal





http://iridia.ulb.ac.be/~marchal/





[Non-text portions of this message have been removed]



------------------------------------

Yahoo! Groups Links

<*> To visit your group on the web, go to:
    http://groups.yahoo.com/group/Fabric-of-Reality/

<*> Your email settings:
    Individual Email | Traditional

<*> To change settings online go to:
    http://groups.yahoo.com/group/Fabric-of-Reality/join
    (Yahoo! ID required)

<*> To change settings via email:
    Fabric-of-Reality-digest-hHKSG33TihhbjbujkaE4pw< at >public.gmane.org 
    Fabric-of-Reality-fullfeatured-hHKSG33TihhbjbujkaE4pw< at >public.gmane.org

<*> To unsubscribe from this group, send an email to:
    Fabric-of-Reality-unsubscribe-hHKSG33TihhbjbujkaE4pw< at >public.gmane.org

<*> Your use of Yahoo! Groups is subject to:
    http://docs.yahoo.com/info/terms/

Re: Joining post: 20 Years of a Science of Consciousness

2012-05-25T15:45:15

Citeren Henry Sturman <henry-wQxkM/xpHhjR7s880joybQ< at >public.gmane.org>:


That's a bit of a straw man argument, as you have put additional 
baggage in the person who knows that he doesn't understand Chinese. If 
you ignore this baggage and let the person reply in Chinese using the 
lookup table method, then when asked in Cinese if he understands 
Chinese, he'll answer that he does understand Chinese.

Of course, he really does not undersand any Chinese, but clearly the 
whole process in which the person only plays a trivial role, does 
understand Chinese.


When we say that a model is not the real thing, we have to consider if 
that distinction is relevant for AI or consciousness. I don't think it 
is, because the brain itself is known to use models of reality and that 
we experience that simulated relality and not the real thing. This is 
how Uri Geller can make you see him bending spoons  by mere touch. What 
you experience here is due to the brain inventing missing information 
based on what information it does have and the internal model of 
reality, and Uri Geller exploiting that to get an absurd result.

Another example: Phantom pain is caused by the brain using an outdated 
model of the body in which the amputated limb is still there. But 
because visual information is missing, the position of the amputated 
limb is determined using other information (cumulative effect of 
motions etc.) which is less accurate. So, basically you have a virtual 
limb, and it will start to drift in awkward positions after a while, 
causing pain.

No surprise then that an effective cure involves fooling the brain 
using mirrors that the amputated limb is the mirror image of the limb 
that is still there. This resets the information about the position of 
the virtual limb, and the phantom pain is gone.


Saibal

Re: Joining post: 20 Years of a Science of Consciousness

Brian Scurfield — 2012-05-25T07:46:47


On 25 May 2012, at 1:08 AM, Bruno Marchal wrote:



I don't see why an unconscious knowledge-generating computer program could not acquire a false idea that is it conscious. Like any knowledge creator, it would be fallible. What you have not explained is the relationship between creating knowledge and consciousness. We know that there are knowledge-creating processes that are not conscious -- for example, biological evolution -- so consciousness is not a necessity for creating knowledge. So, what is it about human knowledge creation processes that requires consciousness? I'm not saying there couldn't be a relationship, but at this point in time we don't know enough.
 
 
What problems does it solve saying that the subconscious mind has its own consciousness?


If it were a clear concept, you should be able to explain it to others.


You just said before that we cannot define it. What problem are you trying to solve in defining it like that?


But then one could do the same computations and create the same knowledge and ideas without it, right? Doesn't that then contradict what Brett said?


A fly moves in a real-time-space. Are you saying a fly is conscious?


You haven't explained what problems consciousness solves in epistemology.



What is a genuine proof? Our best epistemology tells use that nothing can be proved, all knowledge is conjectural. You seem to be flying in the face of that.

Re: Joining post: 20 Years of a Science of Consciousness

Elliot Temple — 2012-05-24T23:25:58

On May 24, 2012, at 11:22 AM, Henry Sturman wrote:


This is the wrong way to look at computer programs.

Programs with the same outputs for the same input are **not the same program**.

For example they may have different run times, be differently hard to re-use components of in other programs, be differently hard to add features too, and many other differences. They may be arriving at these same outputs by different internal algorithms (e.g. recursion vs iteration).

Re: The (scientific) observer is now a scientific target (for 20 years). Sorry!

Elliot Temple — 2012-05-24T23:21:38

On May 23, 2012, at 6:00 PM, Colin Geoffrey Hales wrote:


What's so difficult to believe about thousands of non-philosophers going down a metaphorical dead end road together, which is a dead end due to reasons they've never studied?

History is full of examples of more people being even more wrong together, including examples where those people were generally acknowledged as top experts who had dedicated a lot of their life to their work and were wrong in their field anyway.

Re: Joining post: 20 Years of a Science of Consciousness

Elliot Temple — 2012-05-24T23:17:13

On May 23, 2012, at 6:34 PM, Brett Hall wrote:


That is an implication of *your view of consciousness*. But to assume your view of consciousness is *true*, in a reply to someone saying your conception of consciousness may be false, is a mistake.

a nice illustration of what chapter 2 of the beginning of infinity talks about

Henry Sturman — 2012-05-24T18:11:24

http://www.htwins.net/scale2/

Re: Joining post: 20 Years of a Science of Consciousness

Bruno Marchal — 2012-05-24T13:08:43

On 24 May 2012, at 10:53, Brian Scurfield wrote:

I agree with Brett. Consciousness cannot be a false idea (of the  
conscious person) in the sense that if you are conscious you cannot be  
deluded on this, because to be deluded, you need to be conscious.

Also, the subconscious mind can be conscious, without us being aware  
of it.

"having an idea" is not a very clear expression. It can meant many  
things, involving or not consciousness.




I don't understand this. Each of us know very clearly what it is,  
despite we cannot define it. It is like truth, or time. I don't think  
it is vague. Mysterious perhaps, and undecidable when it concerns  
others, but it is not vague at all. I would say it is clearer than  
anything we might be conscious of.




If you agree to define (approximatively) consciousness by' belief in a  
reality', then you can relate it to betting on our self-consistency,  
(by the completeness theorem) and this makes consciousness speeding up  
our computational ability. That suggests that consciousness has  
developed with the self-moving ability, which needs to anticipate your  
position in a "real-time-space" environment.
Also, human understanding does not require the concept of  
consciousness, but it requires consciousness, a bit like a genuine  
proof does not require the concept of consistency, but it requires  
consistency.

There are many relationships between consciousness and consistency,  
but consciousness has also relation with truth, that consistency has  
not.

Bruno




http://iridia.ulb.ac.be/~marchal/





[Non-text portions of this message have been removed]

Re: Joining post: 20 Years of a Science of Consciousness

Henry Sturman — 2012-05-24T18:22:30

 calls it a duplication)
 n the computer on.
if you look only at information output, and information output happens 
to be all we want from a computer, then there is no difference
and similarly there is no difference in *behaviour* between a real 
chinese person, or a person who does not know chinese but sits in a 
chinese room manipulating symbols in order to simulate a chinese person
but the internal feeling of consciousness is different, as the real 
chinese person really *understands* chinese, while the non-chinese 
speaking person in the chinese room does *not* understand chinese and 
will tell you so if you ask him
and so too, the *information output* of a pizza oven might be the same 
as the *information output* of a simulation of a pizza oven
but the real oven can bake a *real pizza* while the pizza simulation 
machine *can not*
again, see the article 
http://machineslikeus.com/interviews/machines-us-interviews-john-searle-0

Re: The (scientific) observer is now a scientific target (for 20 years). Sorry!

Alan Forrester — 2012-05-24T08:19:21

On 24 May 2012, at 02:00, Colin Geoffrey Hales wrote:

Popperian epistemology and the Turing principle aren't based on assumptions, nor are they purported to be based on anything. Rather, both ideas are explicitly acknowledged as conjectures.

They have not had any success because their ideas about what to look for and how to look for it are crippled by ideas like inductivism. That is, they take for granted that people learn by induction, which is false. They also go looking for stuff without having much idea of what to look for or why.

Could you explain this assertion?

It's possible to observe stuff without doing science: that's what people did for most of human history.

They have no explanation of what the neurons are doing that would lead to thoughts or feelings or whatever.

Part of the problem is that nobody has clearly identified what they want to explain, and they are using junky epistemology, so their failure is not a surprise. If they don't know what algorithm they want to run, they can't get a computer to run it.

Human beings are physical systems and our most distinctive feature is our ability to create explanatory knowledge, i.e. - a particular kind of information processing. According to the Turing principle, which is a consequence of the quantum theory of computation, any physical system can be simulated by a universal computer. So if there is some feature of a human being that can't be simulated by a universal computer, the Turing principle is false. Furthermore, since all of the distinctive stuff about people is a kind of information processing, the simulation is a person according to the Turing principle. If you're going to say a simulation is not a person, then your position contradicts the Turing principle.

Alan

Re: Joining post: 20 Years of a Science of Consciousness

Brian Scurfield — 2012-05-24T08:53:15

On 24 May 2012, at 13:34, Brett Hall <brhalluk-PkbjNfxxIARBDgjK7y7TUQ< at >public.gmane.org> wrote:


No, our subconscious mind has ideas.


Consciousness is a vague concept and right now, as far as I'm aware, nobody knows the relationship between knowledge creation and consciousness. What problems in our understanding of knowledge creation does consciousness solve? I don't know. Do you?

Re: Turing/AI issues

David Deutsch — 2012-05-24T09:20:03


It's false. For instance, one program might be exponentially less efficient than the other. That is a meaningful characteristic.

That's why the principle does not refer to 'meaningful characteristics', but only to 'computations' or 'information processing'.

Turing/AI issues

hibbsa — 2012-05-24T08:11:32

There are two issues facing both A.I. and the Turing Principle:

Firstly...forget consciousness for a moment and rewrite his principle
more simply "if a computer program accurately emulates another computer
program then any meaningful characteristic that program has, the
emulation also has".

No particular problem with that.

But now substitue "a cup of coffee" for the other program. If a computer
program accurately emulates a cup of coffee...does it become a cup of
coffee?

The answer could be yes or no. It could be yes,but with different
hardware and much more complex software. Or it could be no because
Turing's principle is only computational not physical.

Turning back to consciousness, what this shows is that embedded into the
Turing Principle is the assumption that consciousness is analogous to a
computer program, and not at all physical/structural like a cup of
coffee. That may be true...but it's a big assumption, and for that
reason, all this talk about "if the Turing Principle is wrong universal
computation is falsified" is totally flawed, for the simple reason the
Universal Computation law is not falsified by the problems created by
that cup of coffee, and the Turing Principle - if it fails on
consciousness - may fail for the same sort of reason.

The second problem relates to the assumptions about what is accurate
emulation of consciousness and what sort of interface through which
should this be ascertained. But come to the problem from a different
direction. Let's say a computer program suddenly becomes conscious and
says to itself "Wow...I'm here".

That thought/experience, even if it is pure 'software' still has to be
facilitated by the computer hardware. This means that processing time
needs to be allocated to it. It means registry and disk space needs to
be allocated and pointers and references lined up. All of these
processes are hardwired at lower levels of the computer architecture. In
order that software can access those levels it must be put through a
process of compilation, first into - usually - an intermediate language
and then into machine code. From there the hardware processes can be
accessed.

But this is a problem for the A.I. concept. If that thought "Wow, I'm
here" happens, then....does it get written, and compiled first, in which
case...does a new thread get created? In which case what causes that to
happen? And when does the thought "Wow, I'm here" really happen...before
the compilation, or after?

Alternatively, if the consciousness somehow controls the hardwire in a
new way...that is very complex and technical...and would probably have
to happen at lower levels before higher levels. So how would that
happen..how would low level machine code become sentient. It is
incredible inflexible and hardwired stuff. In humans and animals, lower
level controls evolved way back when the hardware was also lower level.
As the hardware evolved...this triggered and was symbiotic with further
evolution of control/software processes, resulting in new layers,
essentially performing new higher level functions while receiving some
other services from pre-existing lower level functioins, which continued
to do a lot of the lower level work.

There is nothing like this happening in computers as they currently
stand. There is nothing instrinsically like this in the very
conceptualization of a hardware/software divide. There is no
evolutionary engine available from a purely software conception of
consciousness.

Re: Joining post: 20 Years of a Science of Consciousness

Brian Scurfield — 2012-05-24T07:53:49

On 24 May 2012, at 02:15, Henry Sturman <henry-wQxkM/xpHhjR7s880joybQ< at >public.gmane.org> wrote:


So, what is the difference between a virtual Windows PC instantiated on an iMac and an actual Windows PC?

RE: The (scientific) observer is now a scientific target (for 20 years). Sorry!

Colin Geoffrey Hales — 2012-05-24T01:00:50

-----Original Message-----
From: Fabric-of-Reality-hHKSG33TihhbjbujkaE4pw< at >public.gmane.org [mailto:Fabric-of-Reality-hHKSG33TihhbjbujkaE4pw< at >public.gmane.org] On Behalf Of Alan Forrester
Sent: Thursday, 24 May 2012 8:21 AM
To: Fabric-of-Reality-hHKSG33TihhbjbujkaE4pw< at >public.gmane.org
Subject: Re: Joining post: 20 Years of a Science of Consciousness

On 23 May 2012, at 06:10, Colin Geoffrey Hales wrote:

http://theconversation.edu.au/learning-experience-lets-take-consciousness-in-from-the-cold-6739

Hi all,

The points raised are not relevant anymore because they were all set up nearly a century ago, with the standard 300 year old presuppositions all through (In particular the assumed observer), in an era where the word 'neuroscience' was barely coined. 

The fact is that there are _thousands_ of scientists chasing 'Neural Correlates of P-consciousness'. All of them think they are doing science on a valid target = P-consciousness, no matter what anyone else thinks. Either they are all deluded or the classic view of the activity is wrong.

The epistemic cat is out of the bag. That cat is quite ill. Explaining the observer is now a 20 year old science. Now, I know the NCC cannot and never will explain the  observer! That's a whole other story. However, what I do know is that sorting out why not will lead to an explanation of the observer and the explicit inclusion of an explanation of our capacity to do science/observation. That explanation will change science itself.

We have a huge inconsistency in science:

Only neuroscientists are trying to explain P-consciousness. Only neuroscientists experience the neurons and then experience the report of the experiences, create a correlate and _then_ are expected that it has explained something _extra_ (the subjective life of the reportee).

In physics: F = MA. The force-correlates of acceleration. Nobody asks "what is it like to be M?". Why not? The neural correlates of consciousness asks 'what it is like'. Why not chemists (W.I.I.L. to be a benzene ring)? Why not engineers (W.I.I.L to be a bridge)? The expected absence of a first-person perspective is _not_ a scientific account of that lack. Equally invalid is the expected non-zero presence of 'what it is like to be a computer' or 'what it is like to be a Turing Machine' is _not_ a scientific account of that presence.

Science is now inconsistently handling scientific evidence on a _massive_ scale. Presuppositions are egregiously interfering with progress.

I merely make a scientific observation about scientists and I report "the weird 'law of nature' correlate of inconsistent scientific explananda by scientists". 

I'm just the messenger. Sometimes your world-view gets a kick in the butt. This is one of those times. :-)

Cheers
Colin

PS1. The Turing principle is perfectly right and completely irrelevant to the process of explaining an observer. Computing a model of a thing is not the thing for the same reason a computed model of flight is not flight and a computed model of combustion does not burn. By extension: A computed model of cognition (e.g. the scientific observer) is not cognition.

PS2. Here's an example of the taboo of the 300 year old view. In 1975, Wilder Penfield helped pave the way to a science of consciousness as follows:

Quote: <Lord Adrian, who shared the Nobel Prize with Sherrington, spoke as a neurophysiologist in 1966 when he said: "As soon as we let ourselves contemplate our own place in the picture, we seem to be stepping outside of the boundaries of natural science." I agree with him; nevertheless, we must step across that boundary from time to time, and there is no reason to assume that critical judgment does not go with us.> End quote.

Ref: Penfield, W. 1975 The mystery of the mind : a critical study of consciousness and the human brain, Princeton University Press. (Chapter 1)

Over 30 years later, "the scientist" = "scientific observation" is definitely being included in scientific activity. Lord Adrian 'seemed' wrong.

Re: Joining post: 20 Years of a Science of Consciousness

Brett Hall — 2012-05-24T01:34:08




On 24/05/2012, at 7:42 AM, "brian_scurfield" <briankscurfield-/E1597aS9LT10XsdtD+oqA< at >public.gmane.org> wrote:


It seems to me that this is a strict contradiction. Anything able to have a "false idea" is necessarily conscious. To have an idea at all is to presuppose consciousness.


As above. Once we can describe what it's like for a computer to have an idea...we will know it is conscious.

Brett


[Non-text portions of this message have been removed]

Re: Joining post: 20 Years of a Science of Consciousness

Alan Forrester — 2012-05-23T22:20:38

Other people have pointed out that you have not explained how you plan to solve the problem that your theory apparently clashes with the Turing principle, but there is another problem.

Your article is epistemologically bad (epistemology is the theory of the growth of knowledge). Saying that science has neglected the observer makes it sound like the most important feature of human beings is that we observe the world. But observation is a relatively small part of our knowledge and of how we make progress. Observation is an example of our general ability to create explanatory knowledge. Knowledge can't be created by observation because observations don't imply explanations or expectations of any kind, which invoke unseen events to explain or predict the things we see. Rather, explanations and expectations lead us to try to make some observations and not others. Arguments like this paragraph in your article:

http://theconversation.edu.au/learning-experience-lets-take-consciousness-in-from-the-cold-6739

consist of subjectivism and inductivism, which were refuted by Popper using arguments similar to the ones above, see Realism and the Aim of Science. Subjective experience is going to be explained by properly understanding the growth of knowledge, not the other way around.

Alan

------------------------------------

Yahoo! Groups Links

<*> To visit your group on the web, go to:
http://groups.yahoo.com/group/Fabric-of-Reality/

<*> Your email settings:
Individual Email | Traditional

<*> To change settings online go to:
http://groups.yahoo.com/group/Fabric-of-Reality/join
(Yahoo! ID required)

<*> To change settings via email:
Fabric-of-Reality-digest-hHKSG33TihhbjbujkaE4pw< at >public.gmane.org
Fabric-of-Reality-fullfeatured-hHKSG33TihhbjbujkaE4pw< at >public.gmane.org

<*> To unsubscribe from this group, send an email to:
Fabric-of-Reality-unsubscribe-hHKSG33TihhbjbujkaE4pw< at >public.gmane.org

<*> Your use of Yahoo! Groups is subject to:
http://docs.yahoo.com/info/terms/

Re: Joining post: 20 Years of a Science of Consciousness

brian_scurfield — 2012-05-23T21:29:42

--- In Fabric-of-Reality-hHKSG33TihhbjbujkaE4pw< at >public.gmane.org, Elliot Temple <curi< at >...> wrote:

One cannot formulate a clear concept without formulating an understanding of the problems to be solved. The question is designed to make the point that the problems are philosophical.

Now it may turn out that consciousness is a false idea we have, in which case one could not have a computer program that emulates a conscious process. The only thing that would be possible is for a computer program to acquire an equivalent false idea that it is conscious. However, the problem of why brains and computer programs acquire the false idea would still be a matter of philosophy and not a matter one could understand by studying the hardware.

Re: Joining post: 20 Years of a Science of Consciousness

Bruno Marchal — 2012-05-23T17:54:11

On 23 May 2012, at 19:16, Elliot Temple wrote:

It is the belief in a reality.


Bruno


http://iridia.ulb.ac.be/~marchal/

Re: Joining post: 20 Years of a Science of Consciousness

Elliot Temple — 2012-05-23T17:16:25


Telling *me* that *we* have a clear concept of consciousness -- and not even saying *what it is* -- is kind of ridiculous.

Especially after I just said anyone with a clear concept, who wants to talk about it, ought to say what it is.

Re: Joining post: 20 Years of a Science of Consciousness

Bruno Marchal — 2012-05-23T15:37:53

On 23 May 2012, at 13:41, Elliot Temple wrote:


It seems to me that we do have a sufficiently clear concept of  
consciousness to be able to discuss it. At least as clear than the  
concept of time, or line, plane, etc. We can agree on some principle  
and reason from that, without any need to find a definition. This is  
the axiomatic, or semi-axiomatic method. For example, I derived a lot  
from the idea that there is a level of substitution of myself such  
that consciousness is experientially unchanged for a functional  
digital substitution made at that level. And you assessed the first  
sixth step of the reasoning (but never answered the seventh step).



We don't need "clear concepts" to make reasoning. We need only to  
agree on clear axioms or principles about them.
But in the case of consciousness, it seems to me we have a clear idea  
of what it is, independently that we cannot define it. It is similar  
for the notion of actual truth, subjective time.

Bruno


http://iridia.ulb.ac.be/~marchal/