On 06 Dec 2011, at 20:44, Stephen P. King wrote:
On 12/6/2011 1:42 PM, Bruno Marchal wrote:
On 06 Dec 2011, at 18:25, Quentin Anciaux wrote:
2011/12/6 meekerdb <meeke...@verizon.net>
On 12/6/2011 4:11 AM, Quentin Anciaux wrote:
The only thing that matter is digitalness... the
> fact that you run it on your pingpong ball computer doesn't
It does matter. If you run computations on pingpong ball computer
interact with the environment
This is relative to the environment. If you want to interact with
the "simulated" brain, you *must* run at the same level. That
does not preclude that the simulated brain can be run on any
level, only interaction with you require a specific level... your
A human being is not a closed system. So the "substitution level"
for Bruno's argument to go thru could include digital simulation
of a large part of the universe - or maybe all of it.
But if all the universe is needed, then computationalism is
certainly false and that would prevent any conscious AI and even
if the argument could still go through with the whole universe...
it seems really like plain old solipsism in that case.
Also, the argument is not about feasibility of capturing the
consciousness of a living person and puting it in a computer but
about the concept and the compatibility with materialism.
Yes, an environment is needed for consciousness, but I doubt that
to capture an existing consciousness (mind uploading) the level
would be more than neuronal or maybe atomic and hence the
environment needed could be feeded via input/output system without
it being explicitely included (weither the "real" one or a virtual
one) in the captured consciousness.
If we assume comp, and if the whole physical universe is needed for
the 'generalized brain', then, by comp, all the universe's states
have to be digitally accessible, and the UD will still access to
those states infinitely often. So the whole reasoning still go
through, even in the case of a concrete physical UD (step seven).
Empirically this is doubtful, though. If the quantum indeterminacy
relies on the first person indeterminacies, then we can bet that we
share the computational states of "our matter constitution" at, or
above, the quantum state of our bodies. Our level is probably above
the quantum level. This makes QM saving comp from solipsism, and is
coherent with Tegmark's argument that the brain does not exploit
quantum superpositions when handling our relevant mechanist
computational states (Sorry, Stephen). We most plausibly do share
deep dynamical histories. Beware the collision with Andromeda!
Yes, I am still reading this LIst. :-) Tegmark is not even wrong
but I do concede the point as it is not relevant to digital
substitution but I, like Craig, caution against thinking that using
classical theory to reason about consciousness is doomed from the
start. Your result, for me, proves that material monism fails
miserably as a T.O.E.
but so does ideal monism.
The irony is that they fail for the exact same reason, the problem
I don't follow you on this. We have discussed that before. Matter
(primitive matter) simply does not exist. It can be an ideal
(immaterial) appearance (by the reasoning). Matter can not be an
epiphenomenon. It is just a phenomenon, and not a primitive one.
But with material monism, matter has to exist primitively (by
definition) and consciousness has to be an epiphenomenon indeed.
The role of matter and consciousness is not symmetrical. Matter can be
an illusion, but consciousness cannot. In all case consciousness has
to be real, or eliminated (which makes no sense). And it t makes
logical sense to eliminate primitive matter, not consciousness. Only
material monism needs to use the notion of epiphenomenon, not
immaterial (number like) monism.
The main difference is that matter (or physics) single out (or try
to single out) one universal system, where comp explains that such
a universal physical system has to be justified from a measure, on
all computations, invariant for all universal machine points of
view, which includes the working of an infinity of universal systems.
The other difference is that by extracting physics from a
computational measure constrained by the logic of self-reference,
we get a natural distinction between qualia and quanta (even if
quanta appears as special case of qualia by the first person plural
nature of physical histories).
But this measure simply does not exist! The set of all
computable functions is of measure zero in the set of all functions.
What are you going to do about this fact?
The measure is on the computations (going through my actual state),
not on all functions. The set of all functions plays some role,
including an "random oracle background", but this is only due to the
fact that from a first person perspective machines cannot distinguish
an oracle from a more complex machine than herself.
We cannot simply postulate a measure
Indeed. We have to extract it entirely from the (math of the) self-
that is not contained by the requirements of our physical world.
UDA shows that once we assume we are digitalizable machine we cannot
use the assumption of the existence of a physical world, even if that
assumption were true.
The solution is to understand that our physical world is what
determines a local measure on the computations.
You can indeed redefine the physical worlds by that property, but you
have to justify its existence by the measure on all relative
computations, as seen from the first person perspective, or you have
to show a flaw in the reasoning.
A search for a "global" or universal measure is quixotic at best.
No, it makes sense thanks to Church thesis and computer science,
especially when you take the logical self-reference constraints into
account, as UDA illustrates the necessity.
You received this message because you are subscribed to the Google Groups
"Everything List" group.
To post to this group, send email to email@example.com.
To unsubscribe from this group, send email to
For more options, visit this group at