On 03/07/07, Bruno Marchal <[EMAIL PROTECTED]> wrote: > But it might then be > questioned how observer and physical narratives could somehow > 'converge' on a common or consistent environmental interface as a > result of any form of co-evolution. Such a notion would seem to imply > that equivalent selection effects could be operating on both > environments despite their orthogonal orientation. It is not > immediately apparent why this should be so.
As far as I understand, this is *the* problem. It seems insoluble untill you take into account explicitly the incompleteness phenomenon which put high constraint of what sound machines can believe, bet, observe, feel ... (that what the hypostases are all about). DN: On reflection, I have something else to ask about this. If I have understood, the comp project is to give an account, as you say, of what emerges from interviewing (computationally) sound machines about their beliefs, bets, observations, feelings etc. In other words, the maximal description of their mental and physical worlds that is possible under such constraints. What would still be indeterminate would be exactly how a specific machine's account could be correlated with a physical description of the machine itself: i.e. the perennial 'mind-body problem'. But do you believe that there can be a way of closing this question theoretically? Is comp agnostic to this correlation being 'functional' or 'physical'? Or will 'yes doctor' remain always a gamble in the sense of "you picks your theory and you takes your choice"? David > > > Le 02-juil.-07, à 18:12, David Nyman a écrit : > > > > > After very kindly concurring with bits of my recent posts, Bruno > > nonetheless quite reasonably questioned whether I followed his way of > > proceeding. Having read the UDA carefully, I would say that in a > > 'grandmotherly' way I do, although not remotely at his technical > > level. > > > Concerning the Universal Dovetailer Argument (UDA) a subtle > prerequisite is Church Thesis, if only to understand the relation with > arithmetical realism and also to understand the term "Universal" in the > UD. Coming back from Siena I am not sure "Church thesis" is well > understood today. > BTW I have discovered that the book edited by Martin Davis "The > undecidable" has been republished in 2004 by Dover. This is really the > comp basic bible. It contains the readable original paper by Godel, > Church, Kleene, Rosser, and, above all the incredible anticipation by > Post. Actually Post did even anticipate my thesis, that is the > platonistic reversal physics/number-computer-science. To be sure he > adds a footnote (footnote 118) saying he changed his mind ... I think I > will write a paper just on that footnote .... > > > > > > But I had been doing thought experiments of a somewhat similar > > nature literally for decades, based on questions like "why am I me and > > not you?" or "how do I know that me now is the same as me 5 minutes > > ago?" or "is the person who gets out of the transporter the same > > person as the one who began the journey?" For some time, faced > > largely with incomprehension or disinterest, and seeing hardly > > anything remotely like this referred to in print, I despaired of > > finding others who believed these questions were anything but > > irrelevant or crazy. But gradually these topics seemed to emerge into > > discussion from a variety of directions, and now I've found a > > community of similarly crazy people on the Internet. > > > Actually those typical comp or mechanist thought experiences > (reasonings) exist since humans use tools. Reference are in the biblios > of "conscience et mécanisme". Many Sc. fiction book go through such > experience, and the book "Mind's I" (ed. by Hofstadter and Dennett) > contains relevant thought (but miss my favorite sc. fi. book, the > "SIMULACRON III" by Daniel Galouye). > Mind'I eyes missed the first person comp indeterminacy and the > subsequent reversal. > > > > > > > The conclusion I had come to is broadly summarised in my recent > > posts. It seemed to me that the 'transporter' questions could only be > > resolved if I thought in terms of my being incorporated in some unique > > or 'global' pre-differentiated manner, which nonetheless > > multifariously self-localised by differentiation of structures that > > embodied distinct 'histories'. > > > I would say that, once we accept comp, the only problem which remains > is the "white rabbit problem", that is: the problem of isolating from > computer science the measure on the relative computational histories > capable of justifying the apparent normality of the observable laws. > > > > > This seemed somehow to entail the > > emergence of finitude from the not-finite, which seemed weirdly > > right. > > > ? (comp presupposes the natural numbers. Indeed they constitute the > absolutely unsolvable mysteries). > > > > > Anyway, it would be the histories that differed, not the > > 'self'. > > > This could depend on choice of vocabulary. If you define the self by > what is consistent, or better sound, and invariant in all comp > histories, you will get the arithmetical hypostases. > > > > > > The histories would break the symmetry of the self into > > differentiated sub-selves that would be 'I' with respect to their own > > private environments. > > > Hopefully. That is what is under the course of verification. Again, > accepting the positive integers makes such symmetry breaking easy to > understand. The real mystery (partially solved though) is in the > understanding of the physical initial apparent symmetry. > > > > > These environments, being participatory, could > > only be shared with other such sub-selves by signalling', and the sum > > total of shareable signals, re-embodied, would be the 'objective' or > > 'outside' physical description of the situation. > > Hopefully again. This would correspond to the first > person plural notion, as far as some part of the comp indeterminacy is > sharable (like the quantum reality seems to confirm). > > > > > But since these > > 'entities' could only be self-defined emergents of the original self- > > relativisation, everything was in fact 'outside-less' and continued to > > exist uniquely or monistically as a network of self-relation. > > > ... itself emerging from the additive/multiplicative number relations. > The self itself is what computer science and provability logic explains > the better. > > > > Depending on whether the participatory or 'objective' perspective was > > adopted, self-relation could apparently decompose into 'sense' or > > 'action' narratives, but such decomposition was in fact illusory, or > > perspective-dependent. Self-relation in fact remained singular or > > decomposable in nature > > This is fuzzy. I can agree but I have more than one interpretation. > It's hard not being more technical here. > > > > > > Having said this, I can now perhaps contextualise more clearly my > > concern about functionalism. Functionalism is the doctrine that > > consciousness is a function of the relationship between parts. > > Hmmm... I would say that this is the old, pre-comp, physicalist account > of functionalism. With comp, and with the fact that consciousness is a > typical first person notion, it is not clear if we can describe > consciousness by any local relation between part. Consciousness (but > also matter) is necessarily related to "all histories". Only the > possible third person discourse can be locally related to describable > relations between describable parts. > > > > > This > > entails that, discounting eliminativism, consciousness must be > > actualised by such relations, > > by an infinity of them. This should follow from the UDA. > > > > and that if such parts were to be > > considered 'ultimately' to be physical, > > > I think this has no meaning in the comp frame. The physical is what > glues the (lobian) personal experience. This includes an infinity of > "linear stories" capable of multiplying themselves by "tensor product" > of independent computational histories. But I have not yet isolate this > tensor structure, to be sure. If it does not exist, comp is false. > Cosmology could be related to the depth and other Church-Turing like > complexity notions. > > > > > then the relevant relations > > could only be physical relations. If this were so, the actual or > > realised relationships existent in a physical structure would be > > exhausted by its physical description, and the ascription of a super- > > added set of 'computational' relationships would merely be > > metaphorical and hence not real enough to be "I". Consequently, if > > physics is held to be fundamental to consciousness, and consciousness > > is an observer effect, then such observers must be fully describable > > by physical relationships, not functional ones, and the appropriate > > substitution level is physical duplication, to some level of > > tolerance. > > OK, but this is explicitly what cannot be done in the comp frame. A > good thing given that "physical" can hardly be defined by the product > of observation. > > > > > > By contrast, if the reality of parts and relationships is to be > > considered fundamentally numerical, > > The main first half of my work consists in showing that comp does not > give a choice in those matter. > > > > > then consciousness and physics > > could indeed be derived functionally or computationally from this kit > > of parts and their relations. From this perspective, the physical > > structure of the body and the observational structure of the mind > > could be held to emerge > > ... should emerge, ok. > > > > > respectively from 'action' and 'sense' > > decompositions of the fundamental self-relative nature of number- > > relations. > > OK. > > > > Nonetheless, if the observer decomposition continues to be > > regarded as 'functional' with respect to the physical one, > > This is unclear, I'm afraid. > > > > > they remain > > in some deep sense orthogonal - i.e. the 'functionalism' is that of > > 'imaginary parts' and 'imaginary relations' with respect to the > > physical description. > > ? > > > > It follows that there may be no final way of > > 'de-crypting' any unambiguous observer structure from the physical > > description alone. We would then be left unavoidably with an > > 'objectively' unknowable and unprovable imputation of consciousness to > > any physically-defined structure. So be it. > > ? > > > > > But it might then be > > questioned how observer and physical narratives could somehow > > 'converge' on a common or consistent environmental interface as a > > result of any form of co-evolution. Such a notion would seem to imply > > that equivalent selection effects could be operating on both > > environments despite their orthogonal orientation. It is not > > immediately apparent why this should be so. > > As far as I understand, this is *the* problem. It seems insoluble > untill you take into account explicitly the incompleteness phenomenon > which put high constraint of what sound machines can believe, bet, > observe, feel ... (that what the hypostases are all about). > > > > > > It may consequently offer some theoretical advantages to suppose that > > a single evolutionary path is followed as a consequence of one-to-one > > reciprocity at some level between sense and action components of a > > primitively self-relative narrative, and that consciousness is > > consequently not 'functional' with respect to the physical > > description, but in some real sense isomorphic or analogic. > > Hmmm ... there is a fixed point ... I am not sure this helps here .... > > > > This > > would not, ISTM, be to postulate any fundamental 'material' substrate, > > but rather to be saying something about the way a 'neutral' > > fundamentally self-relative process could 'decompose' into apparently > > reciprocal components as a consequence of the embodiment of > > observation. However, I can't at this point see whether such > > alternatives can be resolved purely theoretically, or whether they are > > fundamentally empirical issues. > > > > Grandma > > I think I have make progress in that direction. BTW you can consult my > last SIENA paper "A Purely Arithmetical Interpretation, yet empirically > falsifiable, of Plotinus' theory of Matter". Click on the title here: > > http://iridia.ulb.ac.be/~marchal/publications.html > > or click just on > > http://iridia.ulb.ac.be/~marchal/publications/CiE2007/SIENA.pdf > > Hope this can help, feel free to comment. (They are also some typo > error in some formula). > > Best regards, > > 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 -~----------~----~----~----~------~----~------~--~---