Consciousness is information?

[Kelly Harmon]

On Apr 27, 1:42 pm, Brent Meeker <meeke...@...> wrote:
>
> Are you thinking of something like a linked list in which each state, in
> it's inherent information, has a pointer to a previous (or future)
> state.  And the existence of this link constitutes the "feeling of flow"?

Hmmmm.  As a metaphor that works I think.  Though the pointer is
explicitly in the "subjective feeling" of the state, and only
implicitly in the information that underlies that subjective feeling.


> I think it also entails allowing that conscious states have some
> duration in time, or form a continuum.

If consciousness isn't something that exists at any given instant of
time, then when does it exist?  It exists "outside" of time?  What is
a good analogy for how this would work?


> My problem exactly.  But if we are no longer talking about information
> IN a conscious state, but rather information responsible for the
> conscious state then we have introduced the possibility of a whole
> physics (a brain, a world) that may be responsible for many things, only
> one of which is consciousness.  In particular, it may be responsible for
> limiting the conscious states and for fitting them together in succession.

Not if information exists platonically.  So the question is, what does
it mean for a physical system to "represent" a certain piece of
information?  With the correct "one-time pad", any desired information
can be extracted from any random block of data obtained by making any
desired measurement of any physical system.

If I take a randomly generated one-time pad and XOR it with some real
block of data, the result will still be random.  But somehow the
original information is there.  You have the same problem with
computational processes, as pointed out by Putnam and Searle.  The
molecular/atomic vibrations of the particles in my chair could be
interpreted, with the right mapping, as implementing any conceivable
computation.

So unambiguously connecting information to the "physical" is not so
easy, I think.

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups "Everything List" group.
To post to this group, send email to everything-list@...
To unsubscribe from this group, send email to everything-list+unsubscribe@...
For more options, visit this group at http://groups.google.com/group/everything-list?hl=en
-~----------~----~----~----~------~----~------~--~---

[Jesse Mazer]

Some parts of this message have been removed. Learn more about Nabble's security policy.

Kelly wrote:

>
> Not if information exists platonically. So the question is, what does
> it mean for a physical system to "represent" a certain piece of
> information? With the correct "one-time pad", any desired information
> can be extracted from any random block of data obtained by making any
> desired measurement of any physical system.
>
> If I take a randomly generated one-time pad and XOR it with some real
> block of data, the result will still be random. But somehow the
> original information is there. You have the same problem with
> computational processes, as pointed out by Putnam and Searle. The
> molecular/atomic vibrations of the particles in my chair could be
> interpreted, with the right mapping, as implementing any conceivable
> computation.
>
> So unambiguously connecting information to the "physical" is not so
> easy, I think.

This is essentially the problem discussed by Chalmers in "Does a Rock Implement Every Finite-State Automaton" at http://consc.net/papers/rock.html , and I think it's also the idea behind Maudlin's Olympia thought experiment as well. But for anyone who wants to imagine some set of "psychophysical laws" connecting physical states to the measure of OMs I think there may be ways around it. For example, instead of associating an OM with the passive idea of "information", can't you associate with the causal structure instantiated by a computer program that's actually running, as opposed to something like a mere static printout of its states? Of course you'd need a precise mathematical definition of the "causal structure" of a set of causally-related physical events, but I don't see any reason why it should be impossible to come up with a good definition. I think Chalmers attempts one based on counterfactuals in that paper, though I'm not sure if I like that approach.

Jesse

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups "Everything List" group.
To post to this group, send email to everything-list@...
To unsubscribe from this group, send email to everything-list+unsubscribe@...
For more options, visit this group at http://groups.google.com/group/everything-list?hl=en
-~----------~----~----~----~------~----~------~--~---

[russell standish-2]

On Sun, Apr 26, 2009 at 01:01:49PM -0500, Jason Resch wrote:

>
> On Sat, Apr 25, 2009 at 3:52 PM, Kelly <harmonk00@...> wrote:
> >
> > I don't say that they are rare, I say they don't make any sense.  A
> > big difference.
> >
> > I say that every possible event is perceived to happen, and so nothing
> > is more or less rare than anything else.  There are only things that
> > are rare in your experience.  They are not rare in an absolute sense.
> >
> > Why do I say this?  Because I think that platonism is the best
> > explanation for conscious experience, and the above view is (I think)
> > the logical conclusion of that platonic view of reality.
> >
> >
>
> I am not sure that the measure problem can be so easily
> abandoned/ignored.  Assuming every Observer Moment had has an equal
> measure, then the random/white-noise filled OMs should vastly
> outnumber the ordered and sensible OMs.  Though I ever only have one
> OM to go by, the fact I was able to maintain a
> non-random/non-white-noise filled OMs long enough to compose this post
> should serve as some level of evidence that all OMs are not weighted
> equally.
>
> Bruno has suggested that computationalism is a candidate for answering
> the measure problem in a testable way.  However there may be other
> ways to answer it by considering platonic objects, for example
> counting the umber of paths to a state, that is how often it reappears
> as a substructure of other platonic objects, etc.  Whether or not this
> is testable is another question, but whether the ultimate explanation
> of consciousness is computation or information, I feel that measure is
> important.
>
> Jason
>

What you are talking about is what I call the "Occam catastrophe" in
my book. The resolution of the paradox has to be that the
random/white-noise filled OMs are in fact unable to be observed. In
order for the Anthropic Principle to hold in a idealist theory
requires that the OM must contain a representation of the observer, ie
observers must be self-aware. Amongst such OMs containing observers,
ones that are the result of historically deep evolutionary processes
are by far the most common. And evolution of those observer moments
must also be constrained to be similar to those previously observed,
eliminating white rabbits, due to "robustness" of the observer.

Cheers

--

----------------------------------------------------------------------------
Prof Russell Standish                  Phone [phone] (mobile)
Mathematics                        
UNSW SYDNEY 2052                 hpcoder@...
Australia                                http://www.hpcoders.com.au
----------------------------------------------------------------------------

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups "Everything List" group.
To post to this group, send email to everything-list@...
To unsubscribe from this group, send email to everything-list+unsubscribe@...
For more options, visit this group at http://groups.google.com/group/everything-list?hl=en
-~----------~----~----~----~------~----~------~--~---

[russell standish-2]

On Sun, Apr 26, 2009 at 08:19:51PM -0700, Kelly wrote:

>
>
> On Sun, Apr 26, 2009 at 9:00 PM, Jason Resch <jasonresch@...>
> wrote:
> >
> > In fact I used that same argument with Russell
> > Standish when he said that ants aren't conscious because if they were
> > then we should expect to be experiencing life as ants and not humans.
>
> Did you win or lose that argument?
>
> I've heard that line of reasoning before also.  Doesn't it also
> conclude that we're living in the last days?  If there are more
> conscious beings in the future than in the present, then we should
> expect to live there and not here, so there must not be more conscious
> beings in the future?

I did a calculation based on historical population levels with
exponential population growth and concluded there's at least a few
centuries left, although the world population at the end of the 21st
century is likely to be less than at the start. I guess we'll have to
wait a bit to see if that one's right.

> And also it predicts that there are no
> significant number of (conscious) aliens?  Because if there were, we
> should expect to be one of them and not a human?

Remember the Chinese question discussed in the ant paper. There can be
alien planets with greater populations than the Earth, but they must
be relatively rarer than planets containing Earth population levels.

There are other reasons for suspecting intelligent life is rare in the
universe (eg Fermi's paradox).

>
> Sounds like over-use of a good idea.  In this case it ignores all
> other available information to just focus only on one narrow
> statistic.  Why should we ignore everything else we know and only
> credit this single argument from probability?  Surely, after studying
> ants and humans, the knowledge that we gain has to alter our initial
> expectations, right?  But that isn't taken into account here (at least
> not in your one line description of the discussion...ha!).
>
> I think the problem with Russell's ant argument stems from trying to
> use "a priori" reasoning in an "a posteriori" situation.  There is
> extra information available that he isn't taking into consideration.
>

What extra information do you have in mind? I'd gladly update my
priors with anything I can lay my hands on.


--

----------------------------------------------------------------------------
Prof Russell Standish                  Phone [phone] (mobile)
Mathematics                        
UNSW SYDNEY 2052                 hpcoder@...
Australia                                http://www.hpcoders.com.au
----------------------------------------------------------------------------

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups "Everything List" group.
To post to this group, send email to everything-list@...
To unsubscribe from this group, send email to everything-list+unsubscribe@...
For more options, visit this group at http://groups.google.com/group/everything-list?hl=en
-~----------~----~----~----~------~----~------~--~---

[Stathis Papaioannou-2]

2009/4/29 John Mikes <jamikes@...>:

> The Financial Crisis Explained
>
> Heidi is the proprietor of a bar in Berlin . In order to increase sales, she
> decides to allow her loyal customers - most of whom are unemployed
> alcoholics - to drink now but pay later. She keeps track of the drinks
> consumed on a ledger (thereby granting the customers loans).
> Word gets around and as a result increasing numbers of customers flood into
> Heidi's bar.
> Taking advantage of her customers' freedom from immediate payment
> constraints, Heidi increases her prices for wine and beer, the most-consumed
> beverages. Her sales volume increases massively.
> A young and dynamic customer service consultant at the local bank recognizes
> these customer debts as valuable future assets and increases Heidi's
> borrowing limit. He sees no reason for undue concern since he has the debts
> of the alcoholics as collateral.
> At the bank's corporate headquarters, expert bankers transform these
> customer assets into DRINKBONDS, ALKBONDS and PUKEBONDS. These securities
> are then traded on markets worldwide. No one really understands what these
> abbreviations mean and how the securities are guaranteed.
> Nevertheless, as their prices continuously climb, the securities become
> top-selling items.
> One day, although the prices are still climbing, a risk manager of the bank
> -- subsequently, of course,  fired due his negativity -- decides that the
> time has come to demand payment of the debts incurred by
> the drinkers at Heidi's bar.
> However they cannot pay back the debts.
> Heidi cannot fulfill her loan obligations and claims bankruptcy.
> DRINKBOND and ALKBOND drop in price by 95%. PUKEBOND performs better,
> stabilizing in price after dropping by 80%.
> The suppliers of Heidi's bar, having granted her generous payment due dates
> and having invested in the securities are faced with a new situation.
> Her wine supplier claims bankruptcy, her beer supplier is taken over by a
> competitor.
> The bank is saved by the government following dramatic round-the-clock
> consultations by leaders from the governing political parties.
> The funds required for this purpose are obtained by a tax levied against the
> non-drinkers.
>
> Finally an explanation I understand ...
>
> JohnM

Excellent story, worth the brief deviation from the thread topic!


--
Stathis Papaioannou

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups "Everything List" group.
To post to this group, send email to everything-list@...
To unsubscribe from this group, send email to everything-list+unsubscribe@...
For more options, visit this group at http://groups.google.com/group/everything-list?hl=en
-~----------~----~----~----~------~----~------~--~---

[Stathis Papaioannou-2]

2009/4/29 Brent Meeker <meekerdb@...>:

>> I'm not sure I understand. Are you saying that the information in most
>> physical processes, but not consciousness, can be discrete? I would
>> have said just the opposite: that even if it turns out that physics is
>> continuous and time is real, it would still be possible to chop up
>> consciousness into discrete parts (albeit of finite duration) and
>> there would still be continuity.
>
> I could buy that if the finite duration was long enough that the content
> of the conscious interval was sufficient to order the intervals.
> Otherwise you'd need some extrinsic variable to order them (e.g physical
> time, brain states).

It seems to me that if the seconds of my life were according to an
external clock being generated backwards or scrambled, I would have no
way of knowing this, nor any way of knowing how fast the clock was
running or if it was changing speed. So how would the external clock
be able to impose an order on moments of quasi-consciousness below the
critical minimal interval, when it has no subjective effect on
supercritical intervals?

>> In fact, I can't imagine how
>> consciousness could possibly be discontinuous if this was done, for
>> where would the information that tells you you've been chopped up
>> reside?
>
> In Bruno's Washington/Moscow thought experiment that information isn't
> in your consciousness, although it's available via third persons. My
> view of the experiment is that you would lose a bit of consciousness,
> that you can't slice consciousness arbitrarily finely in time.

Could the question be settled by actual experiment, i.e. asking the
subject if they noticed anything unusual?


--
Stathis Papaioannou

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups "Everything List" group.
To post to this group, send email to everything-list@...
To unsubscribe from this group, send email to everything-list+unsubscribe@...
For more options, visit this group at http://groups.google.com/group/everything-list?hl=en
-~----------~----~----~----~------~----~------~--~---

[Quentin Anciaux-2]

Hi,

2009/4/29 Stathis Papaioannou <stathisp@...>:

>
> 2009/4/29 Brent Meeker <meekerdb@...>:
>
>>> I'm not sure I understand. Are you saying that the information in most
>>> physical processes, but not consciousness, can be discrete? I would
>>> have said just the opposite: that even if it turns out that physics is
>>> continuous and time is real, it would still be possible to chop up
>>> consciousness into discrete parts (albeit of finite duration) and
>>> there would still be continuity.
>>
>> I could buy that if the finite duration was long enough that the content
>> of the conscious interval was sufficient to order the intervals.
>> Otherwise you'd need some extrinsic variable to order them (e.g physical
>> time, brain states).
>
> It seems to me that if the seconds of my life were according to an
> external clock being generated backwards or scrambled, I would have no
> way of knowing this, nor any way of knowing how fast the clock was
> running or if it was changing speed. So how would the external clock
> be able to impose an order on moments of quasi-consciousness below the
> critical minimal interval, when it has no subjective effect on
> supercritical intervals?
>
>>> In fact, I can't imagine how
>>> consciousness could possibly be discontinuous if this was done, for
>>> where would the information that tells you you've been chopped up
>>> reside?
>>
>> In Bruno's Washington/Moscow thought experiment that information isn't
>> in your consciousness, although it's available via third persons. My
>> view of the experiment is that you would lose a bit of consciousness,
>> that you can't slice consciousness arbitrarily finely in time.
>
> Could the question be settled by actual experiment, i.e. asking the
> subject if they noticed anything unusual?
>
>
> --
> Stathis Papaioannou

For this you would need an actual AI and also that everybody agreed on
the fact that this AI is conscious and not a zombie.

If you can settle that, then an interview should be counted as proof.
But I'm not sure you can prove the AI is conscious, nor with the same
argument I'm not sure I could prove to you that I am.

Regards,
Quentin

--
All those moments will be lost in time, like tears in rain.

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups "Everything List" group.
To post to this group, send email to everything-list@...
To unsubscribe from this group, send email to everything-list+unsubscribe@...
For more options, visit this group at http://groups.google.com/group/everything-list?hl=en
-~----------~----~----~----~------~----~------~--~---

[Stathis Papaioannou-2]

2009/4/29 Jesse Mazer <lasermazer@...>:

>
> Kelly wrote:
>>
>> Not if information exists platonically. So the question is, what does
>> it mean for a physical system to "represent" a certain piece of
>> information? With the correct "one-time pad", any desired information
>> can be extracted from any random block of data obtained by making any
>> desired measurement of any physical system.
>>
>> If I take a randomly generated one-time pad and XOR it with some real
>> block of data, the result will still be random. But somehow the
>> original information is there. You have the same problem with
>> computational processes, as pointed out by Putnam and Searle. The
>> molecular/atomic vibrations of the particles in my chair could be
>> interpreted, with the right mapping, as implementing any conceivable
>> computation.
>>
>> So unambiguously connecting information to the "physical" is not so
>> easy, I think.
>
> This is essentially the problem discussed by Chalmers in "Does a Rock
> Implement Every Finite-State Automaton"
> at http://consc.net/papers/rock.html , and I think it's also the idea behind
> Maudlin's Olympia thought experiment as well. But for anyone who wants to
> imagine some set of "psychophysical laws" connecting physical states to the
> measure of OMs I think there may be ways around it. For example, instead of
> associating an OM with the passive idea of "information", can't you
> associate with the causal structure instantiated by a computer program
> that's actually running, as opposed to something like a mere static printout
> of its states? Of course you'd need a precise mathematical definition of the
> "causal structure" of a set of causally-related physical events, but I don't
> see any reason why it should be impossible to come up with a good
> definition. I think Chalmers attempts one based on counterfactuals in that
> paper, though I'm not sure if I like that approach.

The atoms vibrating in a rock have a causal structure, insofar as an
atom moves when it is jiggled by its neighbours in perfect accordance
with the laws of physics. And in the possibility space of weird alien
computers it seems to me that there will always be a computer
isomorphic with the vibration of atoms in a given rock. This
requirement becomes even easier to satisfy if we allow a computation
to be broken up into short intervals on separate computers of
different design, with the final stream of consciousness requiring
nothing to bind it together other than the content of the individual
OM's.


--
Stathis Papaioannou

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups "Everything List" group.
To post to this group, send email to everything-list@...
To unsubscribe from this group, send email to everything-list+unsubscribe@...
For more options, visit this group at http://groups.google.com/group/everything-list?hl=en
-~----------~----~----~----~------~----~------~--~---

[Stathis Papaioannou-2]

2009/4/29 Quentin Anciaux <allcolor@...>:

>>> In Bruno's Washington/Moscow thought experiment that information isn't
>>> in your consciousness, although it's available via third persons. My
>>> view of the experiment is that you would lose a bit of consciousness,
>>> that you can't slice consciousness arbitrarily finely in time.
>>
>> Could the question be settled by actual experiment, i.e. asking the
>> subject if they noticed anything unusual?
>>
>>
>> --
>> Stathis Papaioannou
>
> For this you would need an actual AI and also that everybody agreed on
> the fact that this AI is conscious and not a zombie.
>
> If you can settle that, then an interview should be counted as proof.
> But I'm not sure you can prove the AI is conscious, nor with the same
> argument I'm not sure I could prove to you that I am.

Well, you could just ask the teleported human. If he says he feels
fine, didn't notice anything other than the scenery changing, would
that count for anything? I suppose you could argue that of course he
would say that since a gap in consciousness is by definition not
noticeable, but then you end up with a variant of the zombie argument:
he says everything feels OK, but in actual fact he experiences
nothing.


--
Stathis Papaioannou

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups "Everything List" group.
To post to this group, send email to everything-list@...
To unsubscribe from this group, send email to everything-list+unsubscribe@...
For more options, visit this group at http://groups.google.com/group/everything-list?hl=en
-~----------~----~----~----~------~----~------~--~---

[Jesse Mazer]

Some parts of this message have been removed. Learn more about Nabble's security policy.

> From: stathisp@...
> Date: Wed, 29 Apr 2009 23:24:35 +1000
> Subject: Re: Consciousness is information?
> To: everything-list@...
>
>
> 2009/4/29 Jesse Mazer <lasermazer@...>:
>>
>> Kelly wrote:
>>>
>>> Not if information exists platonically. So the question is, what does
>>> it mean for a physical system to "represent" a certain piece of
>>> information? With the correct "one-time pad", any desired information
>>> can be extracted from any random block of data obtained by making any
>>> desired measurement of any physical system.
>>>
>>> If I take a randomly generated one-time pad and XOR it with some real
>>> block of data, the result will still be random. But somehow the
>>> original information is there. You have the same problem with
>>> computational processes, as pointed out by Putnam and Searle. The
>>> molecular/atomic vibrations of the particles in my chair could be
>>> interpreted, with the right mapping, as implementing any conceivable
>>> computation.
>>>
>>> So unambiguously connecting information to the "physical" is not so
>>> easy, I think.
>>
>> This is essentially the problem discussed by Chalmers in "Does a Rock
>> Implement Every Finite-State Automaton"
>> at http://consc.net/papers/rock.html , and I think it's also the idea behind
>> Maudlin's Olympia thought experiment as well. But for anyone who wants to
>> imagine some set of "psychophysical laws" connecting physical states to the
>> measure of OMs I think there may be ways around it. For example, instead of
>> associating an OM with the passive idea of "information", can't you
>> associate with the causal structure instantiated by a computer program
>> that's actually running, as opposed to something like a mere static printout
>> of its states? Of course you'd need a precise mathematical definition of the
>> "causal structure" of a set of causally-related physical events, but I don't
>> see any reason why it should be impossible to come up with a good
>> definition. I think Chalmers attempts one based on counterfactuals in that
>> paper, though I'm not sure if I like that approach.
>
> The atoms vibrating in a rock have a causal structure, insofar as an
> atom moves when it is jiggled by its neighbours in perfect accordance
> with the laws of physics.

They do have *a* causal structure, but I don't see why we should expect to find a set of events in the rock whose causal structure is isomorphic to the causal structure of a computer running a detailed simulation of a human brain for some extended period of time.

>And in the possibility space of weird alien

> computers it seems to me that there will always be a computer
> isomorphic with the vibration of atoms in a given rock.

What do you mean by "weird alien computers"? If we had a way of defining the notion of "causal structure", I'm sure it would be true that in the space of all computer programs (running on any sort of computer) there would be programs whose causal structure was isomorphic of the causal structure of vibrations in a rock, but this might be quite distinct from the causal structure associated with the brains of sentient observers. If you take a panpsychist approach like Chalmers, it might be that all causal structures have *some* type of qualia associated with them, even the ones in a rock (just as we might suppose that even an insect or an amoeba is not totally void of inner experience), but the sort of self-aware conceptual thought that humans have would probably be limited to a small subset of all possible causal structures.

>This

> requirement becomes even easier to satisfy if we allow a computation
> to be broken up into short intervals on separate computers of
> different design, with the final stream of consciousness requiring
> nothing to bind it together other than the content of the individual
> OM's.

As long as the separate computers are each passing the results of their computation on to the next computer in the series, then we can talk about the causal structure instantiated by the whole series. And if they aren't, then according to the idea of associating OMs with causal structures, we might have to conclude that these computers are not really instantiating an OM of a complex humanlike observer even if by some outrageous coincidence the output of all these separate computers *looked* just like the output of a single computer running a simulation of the brain of a humanlike observer.

Jesse

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups "Everything List" group.
To post to this group, send email to everything-list@...
To unsubscribe from this group, send email to everything-list+unsubscribe@...
For more options, visit this group at http://groups.google.com/group/everything-list?hl=en
-~----------~----~----~----~------~----~------~--~---

[Quentin Anciaux-2]

2009/4/29 Stathis Papaioannou <stathisp@...>:

>
> 2009/4/29 Quentin Anciaux <allcolor@...>:
>
>>>> In Bruno's Washington/Moscow thought experiment that information isn't
>>>> in your consciousness, although it's available via third persons. My
>>>> view of the experiment is that you would lose a bit of consciousness,
>>>> that you can't slice consciousness arbitrarily finely in time.
>>>
>>> Could the question be settled by actual experiment, i.e. asking the
>>> subject if they noticed anything unusual?
>>>
>>>
>>> --
>>> Stathis Papaioannou
>>
>> For this you would need an actual AI and also that everybody agreed on
>> the fact that this AI is conscious and not a zombie.
>>
>> If you can settle that, then an interview should be counted as proof.
>> But I'm not sure you can prove the AI is conscious, nor with the same
>> argument I'm not sure I could prove to you that I am.
>
> Well, you could just ask the teleported human.

Yes... but that needs working teleportation ;) don't know if it is
easier to do than an AI :D

But as you say below, the zombie argument still stands. So beforehand
we should have an accepted "test" that tells if an entity is or not
conscious. And I've the feeling it is the most difficult part (besides
construction of a teleportation device or an AI).

> If he says he feels
> fine, didn't notice anything other than the scenery changing, would
> that count for anything? I suppose you could argue that of course he
> would say that since a gap in consciousness is by definition not
> noticeable, but then you end up with a variant of the zombie argument:
> he says everything feels OK, but in actual fact he experiences
> nothing.
>
>
> --
> Stathis Papaioannou
>
> >
>

--
All those moments will be lost in time, like tears in rain.

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups "Everything List" group.
To post to this group, send email to everything-list@...
To unsubscribe from this group, send email to everything-list+unsubscribe@...
For more options, visit this group at http://groups.google.com/group/everything-list?hl=en
-~----------~----~----~----~------~----~------~--~---

[Stathis Papaioannou-2]

2009/4/29 Jesse Mazer <lasermazer@...>:

>>And in the possibility space of weird alien
>> computers it seems to me that there will always be a computer
>> isomorphic with the vibration of atoms in a given rock.
>
> What do you mean by "weird alien computers"? If we had a way of defining the
> notion of "causal structure", I'm sure it would be true that in the space of
> all computer programs (running on any sort of computer) there would be
> programs whose causal structure was isomorphic of the causal structure of
> vibrations in a rock, but this might be quite distinct from the causal
> structure associated with the brains of sentient observers.

Two computers of different architecture running the same program will
go through, on the face of it, completely different physical activity.
Now consider every possible general purpose computer, or every
possible Turing complete machine, running a particular program. I
don't know how to show this rigorously, but it seems to me that the
physical activity in a rock will mirror the physical activity in at
least one these possible computers, and that this requirement will be
easier to satisfy the shorter the period of the computation under
consideration.

>>This
>> requirement becomes even easier to satisfy if we allow a computation
>> to be broken up into short intervals on separate computers of
>> different design, with the final stream of consciousness requiring
>> nothing to bind it together other than the content of the individual
>> OM's.
>
> As long as the separate computers are each passing the results of their
> computation on to the next computer in the series, then we can talk about
> the causal structure instantiated by the whole series. And if they aren't,
> then according to the idea of associating OMs with causal structures, we
> might have to conclude that these computers are not really instantiating an
> OM of a complex humanlike observer even if by some outrageous coincidence
> the output of all these separate computers *looked* just like the output of
> a single computer running a simulation of the brain of a humanlike observer.

I would have said that every computer can generate an OM in complete
causal isolation from every other computer, and the OM's still
associate to form a stream of consciousness simply by virtue of their
content. That seems to me perhaps the main utility of the idea of
OM's. But it appears you agree with Brent that this association won't
happen (or at least, there will be a gap at the seams) unless the
computers are causally connected.


--
Stathis Papaioannou

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups "Everything List" group.
To post to this group, send email to everything-list@...
To unsubscribe from this group, send email to everything-list+unsubscribe@...
For more options, visit this group at http://groups.google.com/group/everything-list?hl=en
-~----------~----~----~----~------~----~------~--~---

[Jason Resch-2]

On Wed, Apr 29, 2009 at 1:05 AM, russell standish <lists@...> wrote:

>
> What you are talking about is what I call the "Occam catastrophe" in
> my book. The resolution of the paradox has to be that the
> random/white-noise filled OMs are in fact unable to be observed. In
> order for the Anthropic Principle to hold in a idealist theory
> requires that the OM must contain a representation of the observer, ie
> observers must be self-aware. Amongst such OMs containing observers,
> ones that are the result of historically deep evolutionary processes
> are by far the most common. And evolution of those observer moments
> must also be constrained to be similar to those previously observed,
> eliminating white rabbits, due to "robustness" of the observer.
>
> Cheers
>

Hi Russell,

What you said reminded me of this article, which appeared in the Boston Globe:

http://www.boston.com/bostonglobe/ideas/graphics/011109_hacking_your_brain/

See the section on hallucinating with ping pong balls and a radio.  It
would seem the way the brain is organized it doesn't accept perception
of pure randomness (at least not for long, I have not yet tried the
experiment myself).  If it can't find patterns from the senses it
looks like it gives up and invents patterns of its own.

Jason

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups "Everything List" group.
To post to this group, send email to everything-list@...
To unsubscribe from this group, send email to everything-list+unsubscribe@...
For more options, visit this group at http://groups.google.com/group/everything-list?hl=en
-~----------~----~----~----~------~----~------~--~---

[Bruno Marchal]

On 29 Apr 2009, at 00:25, Jesse Mazer wrote:

Kelly wrote:

> 
> Not if information exists platonically. So the question is, what does
> it mean for a physical system to "represent" a certain piece of
> information? With the correct "one-time pad", any desired information
> can be extracted from any random block of data obtained by making any
> desired measurement of any physical system.
> 
> If I take a randomly generated one-time pad and XOR it with some real
> block of data, the result will still be random. But somehow the
> original information is there. You have the same problem with
> computational processes, as pointed out by Putnam and Searle. The
> molecular/atomic vibrations of the particles in my chair could be
> interpreted, with the right mapping, as implementing any conceivable
> computation.
> 
> So unambiguously connecting information to the "physical" is not so
> easy, I think.

This is essentially the problem discussed by Chalmers in "Does a Rock Implement Every Finite-State Automaton" at http://consc.net/papers/rock.html ,

Yes. And I don't buy that argument. I will not insist because you did it well in your last post. Also, if it was the case that rock implement sophisticated computations, it would just add some measure on some computations in the Universal Dovetailing. Also, a rock cannot be a computational object: it is a projection of an infinity of computations when we look at the rock at a level which would be below our common substitution level. Eventually we will met the quantum vacuum (assuming comp implies QM, as I think), and in some "parallel world" that vaccum will go through all accessible states, but this is part of so many variate histories that they interfere destructively and does not generate any classical history stable relatively to any observer coupled with the rock.

and I think it's also the idea behind Maudlin's Olympia thought experiment as well.

Maudlin's Olympia, or the Movie Graph Argument are completely different. Those are arguments showing that computationalism is incompatible with the physical supervenience thesis. They show that consciousness are not related to any physical activity at all. Together with UDA1-7, it shows that physics has to be reduced to a theory of consciousness based on a purely mathematical (even arithmetical) theory of computation, which exists by Church Thesis.

The movie graph argument was originally only a tool for explaining how difficult the mind-body problem is, once we assume mechanism.

But for anyone who wants to imagine some set of "psychophysical laws" connecting physical states to the measure of OMs I think there may be ways around it. For example, instead of associating an OM with the passive idea of "information", can't you associate with the causal structure instantiated by a computer program that's actually running, as opposed to something like a mere static printout of its states? Of course you'd need a precise mathematical definition of the "causal structure" of a set of causally-related physical events, but I don't see any reason why it should be impossible to come up with a good definition.

Actually this is a good idea if you define causality by the logical relation linking a universal machine and a computation. But from the first person perspective you will have to take into account all universal machine relating those states, that is an infinity of computational histories. This comes from the invariance of the 1-perspective for arbitrary long delays in the arithmetical universal dovetailer.

Bruno

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

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups "Everything List" group.
To post to this group, send email to everything-list@...
To unsubscribe from this group, send email to everything-list+unsubscribe@...
For more options, visit this group at http://groups.google.com/group/everything-list?hl=en
-~----------~----~----~----~------~----~------~--~---

[Johnathan Corgan]

On Wed, 2009-04-29 at 10:28 -0500, Jason Resch wrote:

> It
> would seem the way the brain is organized it doesn't accept perception
> of pure randomness (at least not for long, I have not yet tried the
> experiment myself).  If it can't find patterns from the senses it
> looks like it gives up and invents patterns of its own.

It is perhaps the other way around.  The portion(s) of the brain
responsible for qualia perception appear to operate as a complex,
dynamical system with a variety of chaotic attractors, and sensory
information only serves to "nudge" this system from one set of attractor
cycles to another.  In the absence of sensory input, these then operate
in open loop mode, and the person may experience all variety of
interesting qualia uncorrelated with the "real" world.

The overall mechanism of dissociative anaesthetic agents such as
Ketamine or nitrous oxide is poorly understood, but one notable property
they have is that in sub-clinical dosages they suppress sensory input
while retaining consciousness.  This results in similar, "open loop"
qualia.

Johnathan Corgan


--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups "Everything List" group.
To post to this group, send email to everything-list@...
To unsubscribe from this group, send email to everything-list+unsubscribe@...
For more options, visit this group at http://groups.google.com/group/everything-list?hl=en
-~----------~----~----~----~------~----~------~--~---

[Brent Meeker-2]

Stathis Papaioannou wrote:

> 2009/4/29 Brent Meeker <meekerdb@...>:
>
>  
>>> I'm not sure I understand. Are you saying that the information in most
>>> physical processes, but not consciousness, can be discrete? I would
>>> have said just the opposite: that even if it turns out that physics is
>>> continuous and time is real, it would still be possible to chop up
>>> consciousness into discrete parts (albeit of finite duration) and
>>> there would still be continuity.
>>>      
>> I could buy that if the finite duration was long enough that the content
>> of the conscious interval was sufficient to order the intervals.
>> Otherwise you'd need some extrinsic variable to order them (e.g physical
>> time, brain states).
>>    
>
> It seems to me that if the seconds of my life were according to an
> external clock being generated backwards or scrambled, I would have no
> way of knowing this, nor any way of knowing how fast the clock was
> running or if it was changing speed.

That assumes that one second can be cleanly (no causal or other
connection) sliced from the next second with no loss, which is what I doubt.

> So how would the external clock
> be able to impose an order on moments of quasi-consciousness below the
> critical minimal interval, when it has no subjective effect on
> supercritical intervals?
>
>  
>>> In fact, I can't imagine how
>>> consciousness could possibly be discontinuous if this was done, for
>>> where would the information that tells you you've been chopped up
>>> reside?
>>>      
>> In Bruno's Washington/Moscow thought experiment that information isn't
>> in your consciousness, although it's available via third persons. My
>> view of the experiment is that you would lose a bit of consciousness,
>> that you can't slice consciousness arbitrarily finely in time.
>>    
>
> Could the question be settled by actual experiment, i.e. asking the
> subject if they noticed anything unusual?
>  

Yes, I think it could - if we could do the experiment.  Certainly when
I've been unconscious, either from concussion or anesthesia I've noticed
something unusual.  :-)

Brent
>
>  


--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups "Everything List" group.
To post to this group, send email to everything-list@...
To unsubscribe from this group, send email to everything-list+unsubscribe@...
For more options, visit this group at http://groups.google.com/group/everything-list?hl=en
-~----------~----~----~----~------~----~------~--~---

[Bruno Marchal]

On 28 Apr 2009, at 22:14, Kelly wrote:

>
>
> On Apr 27, 12:23 pm, Bruno Marchal <marc...@...> wrote:
>>
>> So you have indeed the necessity to abandon comp to maintain your  
>> form
>> of immaterialist platonism, but then you lose the tool for  
>> questioning
>> nature. It almost look like choosing a theory because it does not  
>> even
>> address the question ?
>
> Okay, going back to basics.  It seems to me that there are two
> questions:
>
> A) The problem of explaining WHAT we perceive
> B) The problem of explaining THAT we perceive
>
> The first issue is addressed by the third-person process of physics,
> and of just generally trying make sense of what we perceive as we go
> through the daily grind of life.  Everybody has a grasp of this issue,
> because you're faced with it everyday as soon as you wake up in the
> morning, "what's going on here???".

Well, that is "grandmother physics". It works well locally, but is  
refuted by quantum mechanics, and actually is not tenable with just  
the assumption of computationalism. There is an hard problem of matter.



>
>
> The second issue is obviously the more subtle first-person problem of
> consciousness.

Computationalism makes necessary the reduction of the hard problem of  
matter to the less hard problem of mind. The problem of mind is less  
hard, with comp, because computer science and the provability logics  
can reduce the classical mind problem to the study of the correct  
discourse of the self-introspecting machine using mathematical logic  
to define the notion of self-referencial correctness. We get freely  
the many nuances between true, communicable, provable, knowable,  
inferable, observable, sensible, etc...

I think you have opted for platonist idealism at the start, so perhaps  
you are not motivated to run through the Universal Dovetailer  
Argument, whose main goal is to show that if we assume  
computationalism (the assumption we can use classical teleportation as  
a motion means, or that we are Turing-emulable) then we have to reduce  
the physical laws to number theory/computer science (and a theory of  
consciousness, which I take as a theory of machine's knowledge, with  
the usual modal axiom of consciousness. Consciousness is the "true  
believe in a reality". But with comp that reality is not the physical  
reality, it is more the belief in elementary arithmetic, and from the  
point of view of the machine it is a bet on self-consistency: there is  
a reality which "satisfy me", there is a model making "me" valid. I  
use Gödel Henking "completeness" theorem here.

You can take a look:
http://iridia.ulb.ac.be/~marchal/publications/SANE2004MARCHALAbstract.html

I have already explain ten times this in this mailing list. I think  
most grasp the six or seven steps, but some trouble remains for the  
8th step. I intent also to come back soon or later to the seventh step  
(which contains typical computer science difficulty) in my  
conversation with Kim.
Search UDA in the archive of this list. I have finished in March a new  
version of UDA (in 8 proposition) which has benefited from the  
conversation with the list, and I will put it, someday, on my web page.


>
>
> But, for A, the fact that we are able to come up with rational-seeming
> explanations for what we experience, and that there seems to us to be
> an orderly pattern to what we perceive, doesn't answer the deeper
> question of the ultimate nature of this external world that we are
> observing.

I agree. Aristotle's naturalist hypothesis is an excellent  
methodological simplification, but it departs from all fundamental  
questions asked by Plato. And once Aristotle simplification is taken  
as a granted "fact", or as authoritative axiom,  or worse: as obvious,  
then you can only be led to person elimination, in theory or in  
practice.




> Here we get into issues of scientific/structural realism.
> In other words, what do our scientific theories really mean?  (http://
> plato.stanford.edu/entries/structural-realism/)
>
> But I don't think we can assign any real meaning to what we observe
> until we have an acceptable understanding of the first person
> subjective experience by which we make our observations.


The mind-body problem is the problem of the relation between being and  
appearance. Between the always doubtful sharable objectivity and the  
always non communicable non doubtable direct true apprehension.  
Between third person povs and singular or plural first person pov.  
between Quanta and Qualia.
Physical reality is a first person plural type of things. QM confirms  
this through the multiplication of entangled population of observers.  
Physics concerns invariant patterns in all universal machine (self)-  
observation, and, (unless I am wrong 'course), the physical reality is  
the border of the intrinsic abyssal ignorance of all universal  
machines. My point is only that this entails verifiable/refutable  
facts, and up to now, QM confirms this. Comp easily refute any  
newtonian or classical physics.


>
>
> So the question of consciousness is more fundamental than the
> questions of physics.

Not at all. Those two questions are both fundamental and very  
fundamentally related. If you define consciousness by the true belief  
in a (non necessarily correct or probable) reality (we are conscious  
in dreams), then just computer science (and logic) can explain why and  
how consciousness differentiate in universal machine histories, and  
why stable and sharable dreams develop following long and deep (in  
Bennet sense) ultra-parallel computations (ultra-parallel =  
2^aleph_zero histories).



> We can come up with scientific theories to
> explain our observations, but since we don't know what an observation
> really is, this can only get us so far in really understanding what's
> going on with reality.  Until we have a foundation in place,
> everything built above is speculative.  To rely on physics as your
> foundation is "with more than Baron Münchhausen’s audacity, to pull
> oneself up into existence by the hair, out of the swamps of
> nothingness."

We totally agree on this, except that such a point is not obvious, at  
least for most people after 1500 years of abuse of the Aristotelian  
methological simplification. See the paper for a proof that once we  
assume computationalism (and thus keep the notion of consciousness at  
the start) physics  has to be entirely reduced to the last mystery:  
the mystery of numbers. This one can be shown insoluble by *any*  
machine. All self-referentially correct machine can understand that  
none of them can ever explain where the (natural) numbers comes from.  
This makes them again a natural start. Without assuming them, we never  
get them.


>
>
> But here we hit a problem because the process that we use to explain
> objective data doesn't work when applied to subjective experience.
> There is a discontinuity.  The third-person perceived reality vs.
> first-person experienced reality.  The latter apparently can't be
> explained in terms of the former.

All my point, Kelly, is that if you assume the computationalist  
hypothesis, (digital mechanism, or simply Mechanism) then the former  
has to be explained from the latter.


> But without an explanation for the
> latter, I don't see how any meaning can be attached to the former.

But the later is not so difficult ... once you take some time to study  
computer science and mathematical logic. And then the former reappear  
under a measure problem on computations.

I am amazed by the fact that we talk on comp since years here, and it  
seems many still don't realize comp is made possible by the discovery  
of the universal machine (by Babbage, Post, Turing, Church, Kleene,  
Markov ...). It is a bomb! A creative bomb, for a change. "Nature"  
invented it before 'course, again, and again, and again .... So does  
the numbers, trough their effective enumeration of their computable  
relations.


>
>
> And I think that is for this reason that I don't get hung up on the
> "white rabbit" problem.  Arguments based on the probability of finding
> yourself in this state or that state are fine if all other things are
> equal, and that's the only information you have to reason with.  But I
> don't think that we're in that situation.

Read the UDA and tell me where you have a problem, because this  
paragraph is far too ambiguous.


>
>
> So I start with the assumption of physicalism and then say that based
> on that assumption, a computer simulation should be conscious,

You mean by a physical computer?  Eventually this exist only in purely  
mathematical universal machine's dreams;



> and
> then from there I find reasons to think that consciousness doesn't
> depend on physicalism.

I agree. But this is the reason why we have to justify completely the  
physical appearance from computer science. No doubt information theory  
has a role there, but it is just a part of a very large and rich  
subject.



>  To me, the most likely alternate explanation
> seems to be that consciousness depends on information.


What do you mean by "information"?  Which theory are you referring  
too?  The term "information" is as tricky as the term "random" or  
"infinite".
With comp pure noise (iterated self-duplication) multiplies "freely".  
But the deep things arrive in the many probable relative informations  
on possible histories, in the limit the redundancy makes the deep  
difference.

> However, I am
> relying on some of my thought experiments that assumed physicalism as
> support for my conclusion of "informationalism".
>
> But I think the discontinuity between first and third person
> experience is another important clue, because I think that this break
> will be noticeable to all rational conscious entities in all possible
> worlds (even chaotic, irrational worlds).  They should all notice a
> difference in kind between what is observed (no matter how crazy it
> is), and the subjective experience of making the observation.
>
> Further, let's say that I am a rational observer in a world where
> changes to brain structure do not appear to cause changes to behavior
> or subjective experience.

? (I guess here you come back with the idea that comp is false, isn't  
it). Remember comp is incompatible with the very idea that there is a  
material thing somewhere (by UDA). Ontologically there is only numbers  
together with their additive and multiplicative structure. This  
determine the whole dreamy-reality inside views structure of the  
number theoretical "matrix". See my URL if interested.
It seems to be consistent with your ontological view, but with comp,  
computer science provides the math for the epistemological sides, and  
thanks to the incompleteness phenomenon (and others) all the nuances  
gives very different and incompatible, as such, inside views of  
Arithmetic. Most divided in two, the communicable and the non  
communicable.


>  Physicalism wouldn't have much appeal in
> this world.

In any world. Weak materialism, the idea that matter exists  
*primitively" is provably false in the comp theory, with a minimal use  
of Occam.
Matter is not the answer, matter is the question. Reversal.

> Rather, dualism would seem to have a clear edge as the
> default explanation.

Dualism does not work at all. It uses the identity theory. The UDA  
step 8 and 7 gives no choice in the matter (no pun intended).

But you are platonist we can agree with this.


> But it might be even easier to make the leap to
> platonism in such a world, as presumably Plato's "ideal forms" might
> be even more appealing.

Especially with Church thesis. Theoretical computer science provides  
an entire new realm. Theoretical computer science is a branch of math  
100% unrelated with physics, a priori. Of course many physicists  
follow Landauer and his idea that reality is based on quantum  
information, but my point is that even that move cannot work: quantum  
information has to be a first person plural reflect of classical  
digitalness. Bits and Qubits are related by a double arrow. This makes  
comp a testable theory. David Deutsch seems also to believe that  
physical computability supersedes in fundamentality the usual à-la-
Post-Church-Kleene-Turing... classical computability, but I provide a  
reason to believe that this form of quantum fundamentality is a  
consequence of the numbers fundamentality. In their extensionnal  
relation (like Fermat theorem) and intensional relation (like in  
Kleene or Godel's theorem).
See my paper on Plotinus. The universal machine seems to plagiate  
Plotinus!
We have a theology here. Correct machines are always humble and can  
*only* pray on their consistency, and pray and work on the  
satisfiability of their dreams.



> So in such a world you wouldn't get to
> platonism by way of thinking about computer simulations of brains
> (since brain activity isn't correlated with behavior),

If you mean "brain physical activity isn't correlated in an one-one  
way with subjectivity", then I am OK.


> but I think you
> would still get there.
>
> The question is, what kind of world NOT lead you to Platonism?  I
> think only a world that didn't have first person experience.

We agree on Platonism. But come on, the Pythagorean did already knows  
that numbers quick back. Since then we know that there are universal  
numbers which quick back universally. Comp and computer science are  
interesting for providing the shape (the math) of the uncomputable and  
insoluble which machines have to live with and sometimes name. And  
physics is solidified by its ultimate foundation in arithmetic. With  
comp there is a lot of work to do, that's sure.

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




--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups "Everything List" group.
To post to this group, send email to everything-list@...
To unsubscribe from this group, send email to everything-list+unsubscribe@...
For more options, visit this group at http://groups.google.com/group/everything-list?hl=en
-~----------~----~----~----~------~----~------~--~---

[Jesse Mazer]

Some parts of this message have been removed. Learn more about Nabble's security policy.

Bruno wrote:

On 29 Apr 2009, at 00:25, Jesse Mazer wrote:

and I think it's also the idea behind Maudlin's Olympia thought experiment as well.

>Maudlin's Olympia, or the Movie Graph Argument are completely different. Those are arguments showing that computationalism is incompatible with the physical supervenience thesis. They show that consciousness are not related to any physical activity at all. Together with UDA1-7, it shows that physics has to be reduced to a theory of consciousness based on a purely mathematical (even arithmetical) theory of computation, which exists by Church Thesis.

The movie graph argument was originally only a tool for explaining how difficult the mind-body problem is, once we assume mechanism.

OK, I hadn't been able to find Maudlin's paper online, but I finally located a pdf copy in a post from this list at http://www.mail-archive.com/everything-list@.../msg07657.html ...now that I read it I see the argument is distinct from Chalmers' "Does a Rock Implement Every Finite-State Automaton", although they are thematically similar in that they both deal with difficulties in defining what it means for a given physical system to "implement" a given computation. Chalmers' idea was that the idea of a rock implementing every possible computer program could be avoided if we defined an "implementation" in terms of counterfactuals, but Maudlin argues that this contradicts the "supervenience thesis" which says that "the presence or absence of inert, causally isolated objects cannot effect the presence or absence of phenomenal states associated with a system", since two systems may have different counterfactual structures merely by virtue of an inert subsystem in one which *would have* become active if the initial state of the system had been slightly different.

It seems to me that there might be ways of defining "causal structure" which don't depend on counterfactuals, though. One idea I had is that for any system which changes state in a lawlike way over time, all facts about events in the system's history can be represented as a collection of propositions, and then causal structure might be understood in terms of logical relations between propositions, given knowledge of the laws governing the system. As an example, if the system was a cellular automaton, one might have a collection of propositions like "cell 156 is colored black at time-step 36", and if you know the rules for how the cells are updated on each time-step, then knowing some subsets of propositions would allow you to deduce others (for example, if you have a set of propositions that tell you the states of all the cells surrounding cell 71 at time-step 106, in most cellular automata that would allow you to figure out the state of cell 71 at the subsequent time-step 107). If the laws of physics in our universe are deterministic than you should in principle be able to represent all facts about the state of the universe at all times as a giant (probably infinite) set of propositions as well, and given knowledge of the laws, knowing certain subsets of these propositions would allow you to deduce others.

"Causal structure" could then be defined in terms of what logical relations hold between the propositions, given knowledge of the laws governing the system. Perhaps in one system you might find a set of four propositions A, B, C, D such that if you know the system's laws, you can see that A&B imply C, and D implies A, but no other proposition or group of propositions in this set of four are sufficient to deduce any of the others in this set. Then in another system you might find a set of four propositions X, Y, Z and W such that W&Z imply Y, and X implies W, but those are the only deductions you can make from within this set. In this case you can say these two different sets of four propositions represent instantiations of the same causal structure, since if you map W to A, Z to B, Y to C, and D to X then you can see an isomorphism in the logical relations. That's obviously a very simple causal structure involving only 4 events, but one might define much more complex causal structures and then check if there was any subset of events in a system's history that matched that structure. And the propositions could be restricted to ones concerning events that actually did occur in the system's history, with no counterfactual propositions about what would have happened if the system's initial state had been different.

Thinking in this way, it's not obvious that Maudlin is right when he assumes that the original "Olympia" defined on p. 418-419 of the paper cannot be implementing a unique computation that gives rise to complex conscious experiences. It's true that the armature itself is not responding in any way to the states of successive troughs it passes over, but there is an aspect of the setup that might give the system a nontrivial causal structure, namely the fact that certain troughs may be connected to other by pipes to other troughs in the sequence, so that as the armature empties or fills one it is also emptying or filling the one it's connected to (this is done to emulate the idea of a Turing machine's read/write head returning to the same memory address multiple times, even though Olympia's armature just steadily progresses down the line of troughs in sequence--troughs connected by pipes are supposed to represent a single memory address). If we represented the Olympia system as a set of propositions about the state of each trough and the position of the armature at each time-step, then the fact that the armature's interaction with one trough changes the state of another trough the armature won't visit until a later step may be enough to give different programs markedly different causal structures, in spite of the fact that the armature itself is just dumbly moving from one trough to the next.

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups "Everything List" group.
To post to this group, send email to everything-list@...
To unsubscribe from this group, send email to everything-list+unsubscribe@...
For more options, visit this group at http://groups.google.com/group/everything-list?hl=en
-~----------~----~----~----~------~----~------~--~---

[Brent Meeker-2]

Stathis Papaioannou wrote:

> 2009/4/29 Quentin Anciaux <allcolor@...>:
>
>  
>>>> In Bruno's Washington/Moscow thought experiment that information isn't
>>>> in your consciousness, although it's available via third persons. My
>>>> view of the experiment is that you would lose a bit of consciousness,
>>>> that you can't slice consciousness arbitrarily finely in time.
>>>>        
>>> Could the question be settled by actual experiment, i.e. asking the
>>> subject if they noticed anything unusual?
>>>
>>>
>>> --
>>> Stathis Papaioannou
>>>      
>> For this you would need an actual AI and also that everybody agreed on
>> the fact that this AI is conscious and not a zombie.
>>
>> If you can settle that, then an interview should be counted as proof.
>> But I'm not sure you can prove the AI is conscious, nor with the same
>> argument I'm not sure I could prove to you that I am.
>>    
>
> Well, you could just ask the teleported human. If he says he feels
> fine, didn't notice anything other than the scenery changing, would
> that count for anything? I suppose you could argue that of course he
> would say that since a gap in consciousness is by definition not
> noticeable,

I see no contradiction in a "noticeable gap in consciousness".  Whether  
noticing such a gap depends on having some theory of the world or is
intrinsic seems to be the question.

Brent

> but then you end up with a variant of the zombie argument:
> he says everything feels OK, but in actual fact he experiences
> nothing.
>
>
>  


--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups "Everything List" group.
To post to this group, send email to everything-list@...
To unsubscribe from this group, send email to everything-list+unsubscribe@...
For more options, visit this group at http://groups.google.com/group/everything-list?hl=en
-~----------~----~----~----~------~----~------~--~---

[Bruno Marchal]

Maudlin's point is that the causal structure has no physical role, so if you maintain the association of consciousness with the causal, actually computational structure, you have to abandon the physical supervenience. Or you reintroduce some magic, like if neurons have some knowledge of the absence of some other neurons, to which they are not related, during some computations.

But read the movie graph which shows the same thing without going through the question of the counterfactuals. If you believe that consciousness supervene on the physical implementation, or even just one universal machine computation, then you will associate consciousness to a description of that computation. but the description, although containing the genuine information is just not a computation at all. It miss the logical relation between the steps, made possible by the universal machine. So you can keep on with mechanism only by associating consciousness with the logical, immaterial, relation between the states. from inside they are infinitely many such relations, and this means the physical has to supervene on the sum of those relations "as seen from inside". By Church thesis and self-reference logic, they have a non trivial, redundant, structure.

Bruno

On 29 Apr 2009, at 21:16, Jesse Mazer wrote:

Bruno wrote:

On 29 Apr 2009, at 00:25, Jesse Mazer wrote:

and I think it's also the idea behind Maudlin's Olympia thought experiment as well.

>Maudlin's Olympia, or the Movie Graph Argument are completely different. Those are arguments showing that computationalism is incompatible with the physical supervenience thesis. They show that consciousness are not related to any physical activity at all. Together with UDA1-7, it shows that physics has to be reduced to a theory of consciousness based on a purely mathematical (even arithmetical) theory of computation, which exists by Church Thesis.

The movie graph argument was originally only a tool for explaining how difficult the mind-body problem is, once we assume mechanism.

OK, I hadn't been able to find Maudlin's paper online, but I finally located a pdf copy in a post from this list at http://www.mail-archive.com/everything-list@.../msg07657.html ...now that I read it I see the argument is distinct from Chalmers' "Does a Rock Implement Every Finite-State Automaton", although they are thematically similar in that they both deal with difficulties in defining what it means for a given physical system to "implement" a given computation. Chalmers' idea was that the idea of a rock implementing every possible computer program could be avoided if we defined an "implementation" in terms of counterfactuals, but Maudlin argues that this contradicts the "supervenience thesis" which says that "the presence or absence of inert, causally isolated objects cannot effect the presence or absence of phenomenal states associated with a system", since two systems may have different counterfactual structures merely by virtue of an inert subsystem in one which *would have* become active if the initial state of the system had been slightly different.

It seems to me that there might be ways of defining "causal structure" which don't depend on counterfactuals, though. One idea I had is that for any system which changes state in a lawlike way over time, all facts about events in the system's history can be represented as a collection of propositions, and then causal structure might be understood in terms of logical relations between propositions, given knowledge of the laws governing the system. As an example, if the system was a cellular automaton, one might have a collection of propositions like "cell 156 is colored black at time-step 36", and if you know the rules for how the cells are updated on each time-step, then knowing some subsets of propositions would allow you to deduce others (for example, if you have a set of propositions that tell you the states of all the cells surrounding cell 71 at time-step 106, in most cellular automata that would allow you to figure out the state of cell 71 at the subsequent time-step 107). If the laws of physics in our universe are deterministic than you should in principle be able to represent all facts about the state of the universe at all times as a giant (probably infinite) set of propositions as well, and given knowledge of the laws, knowing certain subsets of these propositions would allow you to deduce others.

"Causal structure" could then be defined in terms of what logical relations hold between the propositions, given knowledge of the laws governing the system. Perhaps in one system you might find a set of four propositions A, B, C, D such that if you know the system's laws, you can see that A&B imply C, and D implies A, but no other proposition or group of propositions in this set of four are sufficient to deduce any of the others in this set. Then in another system you might find a set of four propositions X, Y, Z and W such that W&Z imply Y, and X implies W, but those are the only deductions you can make from within this set. In this case you can say these two different sets of four propositions represent instantiations of the same causal structure, since if you map W to A, Z to B, Y to C, and D to X then you can see an isomorphism in the logical relations. That's obviously a very simple causal structure involving only 4 events, but one might define much more complex causal structures and then check if there was any subset of events in a system's history that matched that structure. And the propositions could be restricted to ones concerning events that actually did occur in the system's history, with no counterfactual propositions about what would have happened if the system's initial state had been different.

Thinking in this way, it's not obvious that Maudlin is right when he assumes that the original "Olympia" defined on p. 418-419 of the paper cannot be implementing a unique computation that gives rise to complex conscious experiences. It's true that the armature itself is not responding in any way to the states of successive troughs it passes over, but there is an aspect of the setup that might give the system a nontrivial causal structure, namely the fact that certain troughs may be connected to other by pipes to other troughs in the sequence, so that as the armature empties or fills one it is also emptying or filling the one it's connected to (this is done to emulate the idea of a Turing machine's read/write head returning to the same memory address multiple times, even though Olympia's armature just steadily progresses down the line of troughs in sequence--troughs connected by pipes are supposed to represent a single memory address). If we represented the Olympia system as a set of propositions about the state of each trough and the position of the armature at each time-step, then the fact that the armature's interaction with one trough changes the state of another trough the armature won't visit until a later step may be enough to give different programs markedly different causal structures, in spite of the fact that the armature itself is just dumbly moving from one trough to the next.

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

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups "Everything List" group.
To post to this group, send email to everything-list@...
To unsubscribe from this group, send email to everything-list+unsubscribe@...
For more options, visit this group at http://groups.google.com/group/everything-list?hl=en
-~----------~----~----~----~------~----~------~--~---