On 19 May 2017 00:55, "Russell Standish" <li...@hpcoders.com.au> wrote:
On Wed, May 17, 2017 at 08:42:22AM +0200, Bruno Marchal wrote:> > > On 16 May 2017, at 10:20, Russell Standish wrote: > > >On Tue, May 16, 2017 at 09:47:14AM +0200, Bruno Marchal wrote: > >> > >>On 16 May 2017, at 04:44, Russell Standish wrote: > >> > >>>On Mon, May 15, 2017 at 11:41:04AM -0700, Brent Meeker wrote: > >>>> >>> > >>>For some reason, people seem to believe that if a consciousness > >>>supervenes on A, then it must also supervene on the combined system > >>>A+B, for an arbitrary system B. > >> > >>We have discuss this before. I don't understand. If C supervenes on > >>A, it means only that a change in C necessitates a change in A. So > >>if C supervenes on A, it has to supervene on A + B, because a change > >>in C will necessitate a change in A + B. > > > >You are making an unwarranted assumption that A and B are > >independent. Just because A changes, does not entail that A+B > >changes. In > >these counter examples, B covaries with A such that A+B does not > >change when C changes. > > I don't succeed in imagining a simple example, still less a rock or > the UD*, where a change in "B" is capable of changing the > supervenience on "A", other than a change in the first person > indeterminacy (but this is taking into account by the abandon of the > physical supervenience: we know already that the first person > experience supervenes on all computations going through our state, > so that the physical does no more exist in fine, that is: the > reversal: it is the physical which, at that state, supervene on > infinitely many computations). > > If this is what you meant, then I am OK. It means eventually that > the comp-supervenience is incompatible with the physical > supervenience. Showing this was the goal of the MGA. Consciousness > supervenes on infinitely many computations (structured by > self-reference), and the physical has to emerge from the measure of > probability (or credibility) on all computations (again, structured > by the modal points of view associated to the machine by > incompleteness). > The argument also works with the UD. No physical supervenience need be involved. In this case, lets say a conscious person a at a particular point in time supervenes on program A, and another distinct person a' at some distinct point in time supervenes on program A'. The UD consists of A+à = A'+Ã', where à denotes all programs other than A (unicode doesn't seem to offer a way of adding an overbar to a letter, so I'll use ̃ instead.). The conscious states a and a' differ, yet the UD A+à is identical to the UD A'+Ã', so we _must_ conclude that consciousness does not supervene on the dovetailer. I offer this as a counterexample to your alleged proof. The B in this case covaries from à to Ã'. I don't follow you (or David) that this necessarily implies physical supervenience is incompatible with computational supervenience, particularly when your work only implies inconsistency with physical ontology (aka primitive physics). My point was somewhat different. I was saying rather that any coherent notion that a particular state of consciousness supervenes uniquely on a specific physical object * qua computatio * is effectively poison ed by ambiguity with respect to how that object can be variously parsed as implementing computation. But one could still say that consciousness supervenes on the observed physical transitions per se without a further entailment to computation. Indeed this is in effect a direct implication of the comp theory itself. The claim is that conscious states "in fact" supervene on computation, in the sense both of covariance and fundamental explanatory relation. But at the same time those states *must appear* to supervene, in a brutely covariant but no longer computational sense, on observed physical transitions. So it would seem then that the implicit "theory-of-mind of observable physics" will always appear in the form of a straightforward identity relation. In this sense the two theories are not strictly incompatible, but they rely on two different explanatory relations. Your point about the dovetailer I think bears on a separate ambiguity . Of course particular conscious states cannot be understood as supervening on the dovetailer as a whole, for the reason you specify . The trace of the UD is by definition unchanging so there is no possibility of covariance. ISTM then that the ambiguity here is that a principle for singling out particular programs and their associated conscious states has not been made explicit. When first one then another particular set of such correlations is successively "selected" from the trace as a whole then there is indeed a covariance. When the selection changes, there is deemed to be a corresponding change in the "states" of both consciousness and computation. In Bruno's work IIUC the initial selection is simply implicit - i.e. we simply start from a particular state and progress by computational relations to the various continuations. One of the reasons I like Hoyle's heuristic is that it makes the serialisation of "selections" explicit and it is consequently easier to see what is supposed to be changing. Anyway, I think in sum I must agree that computational and physical supervenience (i.e. covariance) are not incompatible, but rather that they entail two different modes of explanation. It also seems to be the case that these modes of explanation must themselves be compatible for comp to be viable. David Cheers -- ---------------------------------------------------------------------------- Dr Russell Standish Phone 0425 253119 (mobile) Principal, High Performance Coders Visiting Senior Research Fellow hpco...@hpcoders.com.au Economics, Kingston University http://www.hpcoders.com.au ---------------------------------------------------------------------------- -- 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 everything-list+unsubscr...@googlegroups.com. To post to this group, send email to everything-list@googlegroups.com. Visit this group at https://groups.google.com/group/everything-list. For more options, visit https://groups.google.com/d/optout. -- 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 everything-list+unsubscr...@googlegroups.com. To post to this group, send email to everything-list@googlegroups.com. Visit this group at https://groups.google.com/group/everything-list. For more options, visit https://groups.google.com/d/optout.
