On 30 Apr 2009, at 15:49, Stathis Papaioannou wrote:
>
> 2009/4/30 Bruno Marchal <marchal.domain.name.hidden>:
>
>>> Putnam and Searle use the Rock argument to suggest
>>> that computationalism is false: they consider it absurd that any
>>> conscious computation supervenes on any physical activity (or
>>> equivalently no physical activity, since at one extreme the Rock
>>> argument allows that any computation is implemented by the null
>>> state).
>>
>> ?
>
> If the vibration of atoms in a rock can be mapped onto any
> computation, then there is a one to many relationship between a
> physical state and a computation.
But why and how should the vibration of atoms in a rock be mappable
onto any computations?
Accepting QM I can see one quantum computation: the simulation of the
rock.
Computation are global things. It is not the union of a lot of tiny
computations, it is the union of those tiny computations + the
universal state which unite them. (and then from inside the problem is
that there are an infinity of them).
> That is, you can't say that the rock
> implements one computation but not another.
I don't think it implements any computations. I could accept some tiny
apparition of tiny pieces of of tiny automata, but nothing big or
sophisticated. Some very special crystals perhaps, no doubt, but those
are, then, computer.
> So the rock is a massively
> parallel computer implementing every computation.
No, a finite rock goes in cycle and does not makes any long
computation, still less the deep one. It does at most some
computational noise. Computation, like brain are relatively rare in
the universe. If Hubble detect a computation somewhere, iy will be
taken as an argument for the presence of intelligent beings there. But
it hasn't.
The genome of a bacteria implements very simple form of computations.
There are IF ... THEN... ELSE, loops, and conditionnal (by regulatory
gene). But it took billions of year to "nature" to make them appear.
Just show me the computation of factorial(24) in a rock. No one has
shown that.
> Furthermore, any
> subset (in time and space) of the rock is a massively parallel
> computer implementing any computation.
This is probably true for something like the border of the Mandelbrot
set. But there is no concrete mandelbrot set in nature.
> At the limit, a minimal subset
> of the rock, such as a quark existing for one Planck interval,
> implements every computation.
Hard for me to think you are serious here. A case can be made that the
quantum vacuum is Turing universal, but this makes him doing
"sophisticated" computation relatively to us, only ... in the quantum
white rabbit universes.
> And why not go one step further and say
> that nothingness implements every computation?
If you stay with a physical realm, you will get only physical
nothingness, which even in classical physics is not nothingness.
> So you arrive at the
> conclusion, computation exists independently of physical activity.
That would please me, but I don't see at all the logic.
> Few
> people seem satisfied with this conclusion, so they try to argue
> either that computationalism is false
A lot of people try to argue that computationalism is false, and
usually the argument can be shown directly non valid; Searles for
example mixes levels of description (as Dennett and Hofstadter show
very well in Mind'sI).
Other have better argument, like Maudlin, but this shows only that
comp is not compatible with linking computation with the running of
one universal machine, or worse with the only physical one. It is more
interesting because it shows the real difficulty of the mind-body
problem once we take comp seriously.
But remember Jacques Mallah. He shows that there is an implementation
problem (with physicalism). Along those line a physicalist could
affirm that even a running computer does not run a "mathematical
computation". Unfortunately for Mallah, such a problem dissolves when
you understand that the physical world is not a primitive reality, but
something which emerge from the logical relations among numbers.
Indeed, through the "eyes" of the universal machines/numbers.
Arithmetical reality or alike are the only realms where computations
exists and are well defined.
> or else that computationalism is
> true and dependent on physical activity
Which is false by UDA+MGA.
> and therefore that the
> argument is invalid.
That is weird.
I think that you believe that a rock implements computations, because
you believe a computation can be decomposed in tiny computations, but
this is not true, you need much more. You need a universal machine
which links and complexify the states in a precise way.
Some alive beings do some computations (like some flowers compute tiny
part of the Fibonacci function). But again, this is sophisticated and
took time to appear. Waves do analog computations, hardly universal
digital one, or only when put in some very special condition.
Interesting and rich computations are relatively rare and exceptional
until they self-multiplied, like amoebas.
Nor do I believe the filmed movie graph do any computation, it "read"
a description of one, but does not link them logically in real time.
Today, genetical systems, brains, and computer (human or engineered)
do "concrete" computations.
The mathematical Universal Dovetailer, the splashed universal Turing
Machine, the rational Mandelbrot set, or any creative sets in the
sense of Emil Post, does all computations. Really all, with Church
thesis. This is a theorem in math. The rock? Show me just the 30 first
steps of a computation of square-root(2).
Robinson Arithmetic, Peano Arithmetic, Zermelo Fraenkel Set theory and
many theories compute, notably, all computations, through the
enumeration of the proof of all the True Sigma_1 arithmetical
sentences, but they do much more, they reason and prove more complex
propositions about them (Lobian Machine).
But again, even if rocks implements computations, this changes nothing
to the reversal reasoning. IF rocks implements all computations, it
means the Universal measure is a tiny epsilon more complex to compute,
given that the measure is put on all possible computations. It means
we have to take into account all digital rock's state accessed by the
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.domain.name.hidden
To unsubscribe from this group, send email to everything-list+unsubscribe.domain.name.hidden
For more options, visit this group at
http://groups.google.com/group/everything-list?hl=en
-~----------~----~----~----~------~----~------~--~---
Received on Thu Apr 30 2009 - 17:35:50 PDT