On 28 Apr 2009, at 22:14, Kelly wrote:
> > > On Apr 27, 12:23 pm, Bruno Marchal <[email protected]> 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 [email protected] To unsubscribe from this group, send email to [email protected] For more options, visit this group at http://groups.google.com/group/everything-list?hl=en -~----------~----~----~----~------~----~------~--~---
