> On 26 Jun 2019, at 14:37, John Clark <[email protected]> wrote: > > On Tue, Jun 25, 2019 at 3:31 AM Bruno Marchal <[email protected] > <mailto:[email protected]>> wrote: >> >>> No problem with “reasonable”. My point is that mechanism, nor my >> >>> consciousness in two seconds, or the consciousness of another people, is >> >>> not something provable, > >> True but irrelevant if you're trying to decide what to do with your own > >> life. > > > True, but irrelevant when studying fundamental question. > > There are 2 fundamental questions and we already know the answers to both. > > 1) Question: Is Mechanism as defined by you true? > Answer: Yes.
? Hmm… we cannot know that, but it is almost trivial. It is really like self-consistency: for all self-consistent machine it is true, but none can prove it from its basic beliefs. We can logically conceive that it is wrong, even if it could look ad hoc. And we cannot derive Mechanism from less. > > 2) Question: Can there ever be a proof of Mechanism? > Answer: No. ? Again that depends. “Provable” is always relative to some theory we are willing to accept (which does not necessarily means that we are sure of it, or even that we believe in it: that consideration are personal). > > After that I don't know what more you can say about it. By the fact that the universal machine cannot distinguish between being emulated by any other universal numbers, eventually mechanism reduce the science of the observable by the universal machine, by a statistic on the first person experience. That is confirmed by the fact that restricted on the sigma_1 sentences, the variants of G provides the necessary logic to get a statistics, and then it is already quantum like. >> >>> That assumes digital mechanism. >> >> >>No! The only thing assumed is that if you have observed X then you have >> >>observed X. > > How would that tautology imply or justify Mechanism? > > Bruno, stop playing dumb. You are the one assuming a tautology. > We have both directly experienced consciousness despite the atoms in our > brains being replaced many many times, Assuming that atoms exists in some ontological way. I don’t do that assumption. I am agnostic at the start, then shows that such notions does not make sense with Digital Mechanism, which helps in the quantum mechanical interpretation debate, because the physicist notion of atom is everything but clear. I agree we exêrmeint consciousness, and that we very plausibly share a large part of the physical observable reality, but that does not make Mechanism entirely rationally justifiable. It is a belief in some reincarnation, and it is problematic because it makes us reincarnate in infinitely many computations (in Arithmetic). That can be tested experimentally, and then, indeed we have good sign of where the quantum weirdness might come from, as we get directly the “many-histories” and (less directly) its quantum logic. > so by your own definition of the word Mechanism is certainly true even if we > can't produce it from the set of axioms that we happen to be currently using. So we agree. > > >> If the defender of non-mechanism is conscious then he must conclude that > >> the continuum inside one hydrogen atom is identical to the continuum > >> inside another hydrogen atom because he has remained conscious despite an > >> astronomical number of replacement atoms. > > > But the non-mechanist invoke even an infinite here. Astronomical data will > > not impress him: > > An astronomically large number times an infinite number is infinite, and the > non-mechanist believes atoms contain some sort of mystical analog process > involving infinite digits; and yet the non-mechanist also knows for a fact > that all that swapping in and out those infinite strings of digits has had > precisely ZERO effect on his consciousness. That is what we can tested. > So a student of consciousness has precisely ZERO reason to be interested in > the continuum and those infinite digits even if they exist. Hmm… He learned about the continuum in his course of set theory and analysis. You force me to be the advocate of the devil. But that is what logician are. One counter-example defeat a theory, and when studying the consequence of mechanism, we have to be very cautious and precise on what is assumed, and what is derived from the assumption. > > >> Meaning needs contrast so if everything has a continuum inside it then it > >> would be safe to simplify things and just forget about the continuum. > > > That is reasonable, and echoes an argument by discernibility used by > > Turing. That’s good, but irrelevant in this discussion. > > How on earth is that irrelevant? If a complicated thing has no effect on the > phenomena you're researching then forget about it and spend your time working > on things that might have an effect on it. It has no relevance because you make the digital truncation. But a non mechanist might tell you that whatever truncation you do, even at a very fine grained level, you become a zombie if a decimal is not correct. He will argue that consciousness needs all the decimals. And even with mechanism that is not entirely impossible, but that would means that the first person needs some “real” oracle, fir which there is no evidences. So I agree that the student has no reason to invoke the continuum, but from a logical point of view, that is consistent, and even consistent with mechanism (which is not a lot, as "being wrong” is consistent with arithmetic already. > > > The goal is not to defend mechanism and find good argument for its truth, > > Let's recap what we know or strongly suspect is true about Mechanism as you > have defined it: > > 1) I know for certain it's true. No. Perhaps for uninteresting reason in practice, but we can’t know that, nor can I am sure I will still be alive in two seconds. We can suspect Mechanism to be true, OK. But we cannot claim that science has proven it. > > 2) If you're conscious then you also know for certain it's true. OK. > > 3) We both know for certain a proof of Mechanism can not be derived from the > axioms currently used. Assuming at some meta-level that our currently axioms are consistent, but that makes us very close to inconsistency. I dare to do that only on elementary arithmetic (Robisno Arithmetic). The mathematical logicians got a moment of panic when Nelson claimed to have proven that Peano Arithmetic is inconsistent, and they did study Nelson’s proof (and they found the mistake, to their relieve. > > 4) Although falling short of a proof very good arguments in favor of > Mechanism can be made. Absolutely. My own intuition comes from molecular biology. My first programs I discover was the Operon Lactose in Escherichia Coli. Note that Darwin used quasi explicitly Mechanism (and predicted the existence of a code, like Mendel almost confirmed indirectly, and molecular genetics confirmed). > > 5) I can see no reason why the truth of Mechanism should not be added as a > axiom and if you know of such a reason you have yet to state it. You can added as a sort of meta-axiom once you decide to practice it. It is just that the digital doctor cannot claim that it has been scientifically proved that Mechanism is true, which is a bit of a trivia because in science we never really prove things about reality. We infer theories from experience, and then try to test them. Ou laws involved infinities and universals, but as locally finite creature we can only guess. The French poet Fontenelle said once “de mémoire de rose je n’ai jamais vu mourrir un jardiniere” (from my memory of rose I have never seen a Gardiner who died). > > >> People may wish to be unique but the universe is not required to conform > >> to human aspirations . > > > Right. Yet, people wanting to be unique will be motivated to abandon > > Mechanism, and will try non computationalist theory of mind. > > I agree, people abandon logic and engage in magical thinking and believe if > they want something to be true strongly enough then it is true. OK. In fact it is the consistent on on mechanism which makes so much people willing to believe in what they want. It might play a role in the development of human, though, in helping to keep some optimism, or perhaps that optimism could be taken as an evidence truth and good are related and motivate us to search it. > > >> To hell with hypothesis and to hell with proof, direct experience outranks > >> them all. > > >But that cannot be invoked as an argument. > > So what? We both know for a fact that Mechanism as you have defined it is > true, It cannot be a fact, but it can be the simplest working theory today, not eliminating consciousness, nor the observable. But to progress in the mind-body problem, we have to progress in the extraction of physics from the mind of the universal machine, and we have to pursue its comparison with what we observe. > so what use could either of us have for an argument or a proof? Even if I had > a proof (which I never could have) it would not make me one bit more certain > of the truth of Mechanism than I already am. We can refute it. That is the whole point of making “metaphysics” into science again, a bit like with Bell’s inequality. If S4Grz1 & Co departs too much from the observation, then we have a clue that Mechanism could be wrong. Something else would be at play in the consciousness phenomenon. We can’t logically exclude this today, and I show how that can be tested. It shows a direction where we can progress. > > >> And nothing is more direct than consciousness. > > > Yes, we agree on this. But a theory has to be third person sharable. > > I have no need for a proof or a argument in favour of the truth of Mechanism > and can see no reason why a third party would need such a thing either. My point is only that Mechanism is refutable. It is enough to compare the physics shared by all universal numbers/machine from their first person (plural) perspective, and what we observe. > > > The debate was on the possible or not rational justification of mechanism, > > not about its truth or falsity > > When we both already know it's true I don't understand why a rational third > party would want to hear my justification of Mechanism, I certainly don't > want to hear his. It is needed to understand that Mechanism is refutable. It illustrates that we can keep the scientific attitude in philosophy, metaphysics or theology. It helps to shale a little bit the certainty of the materialist. Many people still believe that Mechanism and materialism go hand in hand, but the point is that they are incompatible, and we can test which one is the more plausible. Thanks to QM, mechanism is winning here. QM-with-no-collapse confirms the easiest quantum-like fact that the present is a sort of sum on all fiction, with superposed states and virtual particles. > > >> The non-computationalist knows the atoms in his brain have been replaced > >> many times, > > > The non-computationalist might not care about its atoms, and believes that > > his soul has a bit of <whatever-you-want>. > > If the non-computationalist is such a fool that he doesn't know that a change > in the arrangement of atoms in his brain changes his consciousness and a > change in his consciousness changes the arrangement of atoms in his brain > then no rational argument will convince him of anything and I'm wasting my > time talking to him. OK. Me neither. But my point is purely logical. You don’t need to believe in curved geometrical space, but you need to be able to conceive it to prove that the parallel postulate is not provable from the other postulate. To show that some proposition P is not provable in a theory, you need only to be able to conceive a model (a “reality”) which satisfy the axioms of the theory, and satisfy ~ P. The model can be as bizarre as you want. I might illustrate this by showing why Robinso Arithmetic cannot prove 0 + x = x, despite ut can prove, for all number n: 0 + n = n. > > > Or he could invoke the fact that the atoms are replaced in a continuous > > way, etc. > > Then, because his consciousness remains unaffected, He will tell ask why you claim this. His point is that his consciousness will be affected. > the continuous stream of continuum rich matter flowing out of the brain must > be identical with the continuous stream of continuum rich matter flowing into > the brain, and so we can simplify things and cancel out the continuum. Even > if the continuum exists, and I have my doubts, it plays no part in > consciousness. Some continuum is at play, even with mechanism, by the first person indeterminacy. But that is out of the scope here. > > >> he believed the substitute atoms do not contain a sufficient number of > >> digits for consciousness ( that is exactly what makes a > >> non-computationalist a non-computationalist) then he would have to think > >> that he does not think. > > > He has also the option to think that after having a brain transplant he > > will be dead, > > Every non-computationalist in the world has already undergone many brain > transplants. Not verifiability digital one. Unless you assume already mechanism. > So every non-computationalist in the world thinks he is dead. So every > non-computationalist in the world is insane. ... > > > and the copy is an impostor > > That's a little better, now every non-computationalist in the world just > thinks he's a fraud. At least he comes back to sanity! > > > No theories at all are provable. A theory is always the set of proposition > > that we assume. All theories are hypothetical. > > That's nice, but Mechanism is not a theory it is a observation of a direct > experience. Hmm, I don’t think so. It is a theory inferred from the current knowledge of molecular biology, and quantum mechanics, but as a philosophical assumption, we cannot asserts it as a scientifically demonstrated fact, like we agreed above. > > > If science reject what is not provable then [...] > > I don't need a proof and I don't need a theory and I don't even need science > if I have direct experience, and in this case I do. You don’t need a proof. That’s OK. What remain is called faith, and as long as you don’t impose this to someone else, there is no problem. > > >> If you had a proof that I felt no pain I would know immediately that you > >> either made a logical error when you formed the proof or you started from > >> a bad set of axioms. > > > Absolutely. But this is not relevant for the simple fact that Mechanism is > > possibly false, > > It's very relevant because both pain and consciousness are direct > experiences.There is no proof of Mechanism and there never will be but there > is no way it could be false, I know this from direct experience and if you're > conscious you know it too. How could I know that? I know only my consciousness here and now. I don’t know I will stay alive in the next seconds. FAPP those nuances are not interesting, but to understand that mechanism reduce the mind-body problem into a derivation of physics from the internal statistics in arithmetic on all computations, all the little details are important to be made precise. > > > and thus has to be assumed if we use it. > > If I live on a desert island and am going to use it for my own purposes I > don't need to assume anything if I want to use mechanism because my certainty > of it's truth was obtained from direct experience, and that is vastly > stronger than if I just had a proof it is true. Proofs are wimpy compared > with direct experience. No problem. The problem is when people could impose that truth to others. That would be scientism, not science. > > > > It belongs to the class of true but non provable truth, > > Yes! An axiom is suposed to be a self evidently true statement that can not > be derived, Well, actually we can prove the axiom easily, with the notion of “proof” used by logician. Just by mentioning “axiom”. In the theory with the only axiom “mechanism is true”, we can prove (trivially) that mechanism is true. Here the nuance between proof, proof and truth, proof and consistency, is what makes possible to make such nuances precise, as it get more and more counter-intuitive, as expected. > and that perfectly describes mechanism; so let's just add it to the list of > existing axioms. For personal use only. > > > If you could prove the inconsistency of all non-computationalist theories, > > then you would prove Mechanism, > > I don't know why you keep talking about proof as if it's the ultimate roadway > to truth. On the contrary. Every things proceeds from the fact that []p -> p is NOT provable by the machine, and I have to distinguish cautiously between proof and truth just to get the notion of proof and truth ([]p & p). I am the one that insist that mechanism is not provable, but trivially true, of course, in the mechanist theory. That will help to understand how theology (the Solovay modal logic G* and variant) has to extend science (the Solovay Logic G*). > If you had an error free proof that X=Y that proof would not convince me that > X is indeed equal to Y if I knew from direct experience that those 2 things > were not in fact equal. Such a proof would however tell me one thing, you > must be using a bad set of axioms because a proof is only as good as the > axioms it's built on. OK. But when interviewing the (Löbian) universal machine, we must keep all those nuances into account. We can’t know that mechanism is rationally justifiable, if only because it uses church’s Turing thesis, which is also easily argued to be not rationally justifiable/provable, yet quite a good guess if you want my *opinion*. Then when you know how to use Diagonalosation, it helps to understand how unbelievable Church-turing thesis is. It is not well known (but seen by Emil Post, Stephen Kleene and Judson Web) but incompleteness is a quasi-direct consequence of the Church-turing thesis. > > >> If he asserts "John Clark is not conscious" I know from the most > >> authoritative source there is, direct experience, that he's wrong. The > >> fact that I lack a proof to convince him that he is in error does not > >> change the fact that he has indeed made an error. > > > The discussion is on provability. You keep distracting us from where we > > started. > > Distracting from what? You have, at various times, agreed with me that > Mechanism is true and Mechanism has no proof and never will, so I don't see > what more there is to discuss about it. As a scientist, I do not discuss about he truth of my axiom/hypothesis. I derive consequences, and show them testable, and indeed, thanks to QM, rather well tested up to now. > > > We are not arguing truth or falsity of mechanism. Just its non rational > > justifiability, or provability. > > >>Then we're not arguing. Mechanism, as you've defined the word, is certainly > >>true and certainly can not be proven. > > > So we agree. > > I guess so. Sort of. > > > The pint has never been on the truth of mechanism, but on its absence of > > entire rational justification. > > I have a very rational reason for believing mechanism is true, it's just that > there is no way my reason can be communicated; fortunately that's not a big > problem because you have your own reason for believing mechanism is true so > you don't really need my communication . Actually I don’t abroad the question of the truth of mechanism at all. The whole point is that, thanks to the Church-Turing thesis, we can use it to formulate the min body problem. Indeed to reduce it to a statistics on all computations, and then the “measure one” is given by some variant of Gödel’s provability predicate ([]p), and we can test this by comparing this with experimental physics. It just fits well thanks to quantum mechanics without collapse. My meta-key point is that we can do philosophy/metaphysics/theology with the scientific method, which consists precisely in keeping for oneself our feeling of truth and wrong, but proposing theories and means of eventuating them. To be crisp and sharp: the human science and philosophy/theology is bs since we mixed it with state and temporal decision. We still lack the modesty, and that explains why we still use bombs and insults so often in such domain. > > >> Yes it's not provable but that's not important if you have something even > >> better than proof, and you do, direct experience. > > > So we agree. > > I guess so. Sort of. Good. Once you get the step 3, and the fact that the notion of computation is purely arithmetical, you will get the whole picture. I think. But now we could also ask this question. What if we build artificial brains and teleportation device, and eventually we all end up into (physical computers), but still pursuing the experimental testing, and discover that the Turing-observable (S4Grz1, Z1*, X1*) do depart from the observation of nature? Mechanism is discovered to be wrong, yet everybody was “sure” to survive *integrally*. The answer is that we did not survive integrally, and that we are suffering from some agnosia, like people becoming blind, but never noticing. Here, what we would lose can be subtle aspect of consciousness we are not much aware of, and that illustrates again other way computationalism might be conceived to be wrong, despite the experimenters claiming that they survive. Bruno > > John K Clark > > > -- > You received this message because you are subscribed to the Google Groups > "Everything List" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected] > <mailto:[email protected]>. > To view this discussion on the web visit > https://groups.google.com/d/msgid/everything-list/CAJPayv0tjDjMH4z%2BtgKPYGHbPah2fdE%2B--nnvj2rkp4iQdsNSA%40mail.gmail.com > > <https://groups.google.com/d/msgid/everything-list/CAJPayv0tjDjMH4z%2BtgKPYGHbPah2fdE%2B--nnvj2rkp4iQdsNSA%40mail.gmail.com?utm_medium=email&utm_source=footer>. -- You received this message because you are subscribed to the Google Groups "Everything List" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/everything-list/2FA35505-2E10-4A8A-8E27-F2E29B36543B%40ulb.ac.be.
