Hi David,
On 1/15/2012 9:47 AM, David Nyman wrote:
On 14 January 2012 18:56, Stephen P. King<stephe...@charter.net> wrote:
But the Turing Test is a bit of an oxymoron because it
is impossible to prove the existence of something that is solely 1p. There
is no 3p of consciousness.
I agree, and in a sense this implies the futility of all attempts to
argue from 3p to 1p. But there may be other ways to get there. For
example, I've always tended towards the view that Bruno often calls a
"universal mind" (cf. Schopenhauer, Schrödinger, Hoyle, Dyson,
et.al.). Think of this as a universal 1p. The argument from this
point of departure begins in such uniquely present conscious
instances, whose internal logic implies the possibility of other,
mutually exclusive, such instances. As a first approximation, this
internal logic might imply that the present instance is a selection
from a uniquely "personal" serialisation of such instances (i.e. the
RSSA). However the same logic is consistent with all possible such
instances, the implied personal serialisation now playing a secondary
role to some transcendental, "impersonal" selection (i.e. the ASSA).
This insight offers an escape route from solipsism.
I like your thinking here but need to point out a few things. It
seems to me, and this is purely a conjecture of mine, that the 1p's are
limited to being representable by Boolean algebras (or equivalently (?)
lists of questions with yes or no answers). I think of this in terms of
all facts that can be determined by measurements from a point of view
tied to a place in space-time (using the idea that space-time is a
"container"), a center of the universe if you will. (This idea comes
from the work of Prof. Hitoshi Kitada
<http://arxiv.org/find/gr-qc/1/au:+Kitada_H/0/1/0/all/0/1>.) How can a
large number of these be woven together into a consistent narrative? My
first attempt was to think of 3p as the intersection of many 1p but this
does not work out so well as there are concurrency issues to be dealt
with...
I have found that they might be able to be uniquely woven together
if the strict determinism of classical physics where to hold for all 1p
at all places and epochs (as it sets up separable and unique systems of
trajectories - worldtubes - for objects), but this is not the case as we
see from the evidence of QM. The structure of the logic of QM systems
(orthocomplete lattices) does not allow for unique decomposition into an
ordered set of Boolean algebras therefore one cannot use QM to construct
a universal mind that is isomorphic to ours. The escape from this would
be to consider finite collections of Boolean algebras representing a
plurality of 1p's having a "common world", a "consensus" reality of
sorts, and consider how to map such back into the Orthocomplete lattice
of the QM system that encompasses all of our common world. Our "gods
that view all of Reality from on high" can and should be relegated to
the category of useful but ultimately incorrect explanations.
If my conjecture is correct then it does not allow for an escape
from solipsism, but I do not think that that is such a bad deal as I see
solipsism as the natural implication that flows from the "privacy" of
the 1p. OTOH, our ability to reason coherently (given enough effort as
it is not a passive activity) allows us to jump past the isolation and
even alienation of the 1p to justifiably believe in the reality of other
minds. I like the way that Bruno addresses this with the Bp&p and
"betting that p is true" ideas.
Can one apply a view like this to the problem in hand? 3p is the
label applied to our theoretical proxies for the regularities of 1p
phenomena. These regularities are so compelling that for most purposes
we treat them perfectly naturally as realities independent of the 1p
context in which they manifest.
Yes, this is why, IMHO, people like Stephen Hawking think of physics
as "the mind of god". I do see the attractiveness of this idea but have
discovered that there are many reasons why it is incoherent. For one
thing there are theorems in network theory that show that arbitrarily
large networks cannot have a single global synchronization (unless the
speed of light is infinite and exact bisimulation between the nodes is
possible). What we actually seem to have in our physical world is a
speed of light that is finite but behaves locally as if it where
infinite as it defines the maximal lengths between events.. but I digress.
My questions here go back to this idea of "realities independent of
the 1p context in which they manifest". This "independence" seems to be
the same kind of "independence" that I am wrestling with in my debate
with Bruno. Is it truly independence in the sense of separability, in
the sense that a coherent notion of Reality can exist completely
isolated from the 1p's? I don't see how! But this takes me back to the
tar pit of solipsism... Why not use solipsism constructively? This is
what Andrew Soltau is attempting to do...
We situate them in an ever more
general explanatory framework, in terms of which we hope to trap even
the 1p localisation to which all such explanation is ultimately
referred. But frustratingly, attempts to achieve this by the direct
3p route seem always to rely in the end on some sort of unsatisfactory
bait-and-switch.
I agree!
Nonetheless we cannot deny that there are subsets of
the 3p schema which correlate strongly with the implied serialisation
of 1p moments: those subsets we accept as our local physical
embodiments. Consequently, it seems reasonable to postulate the
tightest of inter-relations, short of identity, between these two
domains, at least locally.
Yes, this is consistent with my conjecture. We can recover a
"local" notion of Reality that we can "bet" (per Bruno) is "Real" and be
amazingly successful in our predictions of behaviors, but at the price
of havig to give up "certainty" and having to use statistics in our
physics. I suspect that this is how Bruno comes to the conclusion that
we can derive QM in the model within his result.
Returning to the original point of departure with the inference of a
tight local correlation between some appropriate 3p physical
embodiment and the presently selected 1p instance, it might seem a
reasonable experiment to reverse the logic.
Yes, but there is not a bijective map between 3p and 1p unless we
assume that infinite information is available to do the reconstruction. :-(
If we could but identify
the relevant species of 3p embodiment - given the anti-solipsism
argument derivable from a strictly 1p point of departure - we could
reasonably infer its correlation with an instance of consciousness
mutually exclusive of the present one, entangled with its own coherent
personal serialisation (or more baldly, another person).
I like your thinking here, I like it a lot! But it doesn't work
since the identification of the "relevant species of 3p embodiment" is
computationally intractable. An analogous situation involves trying to
compute the smooth diffeomorphism between an arbitrary pair of
3-manifolds. Penrose mentioned this in one of his books as being proven
by Markov to be NP-Complete.
But how to identify the "relevant species"? Ordinarily, we do not
hesitate to ascribe this status to other human embodiments, because it
seems reasonable to suppose that if our own 3p constitution is of the
relevant species, so is theirs.
Yes, as this would be a form of universality...
But as we have no widely-accepted
definitive account of what this entails specifically, we must rely
essentially on the behavioural manifestations of intelligence.
I agree and so we must accept that we are condemned to isolation
within our 1p's, but this is not a reason to slide into nihilism (ala
Nietzsche and Sartre), we can see the computational intractability as an
answer. Instead of trying to formulate models of some kind of
"pre-ordained harmony" (ala Leibniz), we could think of the computations
as ongoing and eternal process (ala Bergson).
Accordingly, we have little option but to ascribe the "definitively
conscious" 1p-3p correlation to any embodiment that displays
sufficiently intelligent behaviour, by some agreed criterion such as
the TT.
Yes, this is where Bruno's ideas work wonderfully together with
others. We bet that Bp&p is true and go from there. We make conjectures,
test them, improve them, etc.
The critical exception to the foregoing is that we would clearly wish
to withdraw this ascription where there is demonstrable evidence of
fraud or pretence.
Absolutely agreed! Fraud detection seems to be hardwired into our
brains, co-existing with our ability to lie; the result of an arms race
of sorts between homo habilis.
Hence the vanishing point for controversy may well
be FAPP when the pretence of intelligence has become practically
indistinguishable, by any available criterion, from its actuality.
I agree. If we cannot detect a difference between X and Y we might
as well pretend that X = Y for most cases. One strategy that seems
useful is*Fallibilism <http://en.wikipedia.org/wiki/Fallibilism>*...
Onward!
Stephen
David
On 1/14/2012 1:15 PM, Evgenii Rudnyi wrote:
On 14.01.2012 18:12 John Clark said the following:
On Fri, Jan 13, 2012 meekerdb<meeke...@verizon.net> wrote:
There is no way consciousness can have a direct Darwinian
advantage
so it must be a byproduct of something that does have that virtue,
and the obvious candidate is intelligence.\
That's not so clear since we don't know exactly what is the
relation of consciousness to intelligence. For a social animal
having an internal model of ones self and being able to model the
thought processes of others has obvious reproductive advantage.
To do any one of the things you suggest would require intelligence,
and indeed there is some evidence that in general social animals tend
to have a larger brain than similar species that are not social. But
at any rate we both seem to agree that Evolution can only see
behavior, so consciousness must be a byproduct of some sort of
complex behavior. Thus the Turing Test must be valid not only for
intelligence but for consciousness too.
How would you generalize the Turing Test for consciousness?
Evgenii
John K Clark
Hi,
Perhaps we can generalize the Turing test by insisting on questions that
would require for their answer computational resources in excess of that
would be available to a computer + power suply in a small room. Think of the
Berkenstein bound.... But the Turing Test is a bit of an oxymoron because it
is impossible to prove the existence of something that is solely 1p. There
is no 3p of consciousness. I recall Leibniz' discussion of this...
Onward!
Stephen
--
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@googlegroups.com.
To unsubscribe from this group, send email to
everything-list+unsubscr...@googlegroups.com.
For more options, visit this group at
http://groups.google.com/group/everything-list?hl=en.
--
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@googlegroups.com.
To unsubscribe from this group, send email to
everything-list+unsubscr...@googlegroups.com.
For more options, visit this group at
http://groups.google.com/group/everything-list?hl=en.
