On 29 Apr 2017 8:00 p.m., "Brent Meeker" <[email protected]> wrote:
On 4/29/2017 12:16 AM, David Nyman wrote: On 29 Apr 2017 2:33 a.m., "Bruce Kellett" <[email protected]> wrote: On 29/04/2017 1:18 am, Bruno Marchal wrote: > On 27 Apr 2017, at 23:22, Brent Meeker wrote: > > The absurdity, if I've understood this, is that idea of physical >> substitution leads to a conclusion that nothing physical is needed. >> > > The absurdity is that the first person experience in physics has to be a > sum on all computations (to be short). > This might cause some problems for the SAN04 argument. The "Yes, doctor" assumption says that if my brain is replaced by a functionally correct digital device (at the appropriate substitution level), then I would not be aware of any experiential change. But if my first person experience is the sum on all computations that pass through my conscious state, then no digital computer could ever be a "sum on an infinity of computations", so my conscious state could not be reproduced in this way. In fact, as has been said, the sum on all computations is not Turing emulable. Thus my conscious state is not Turing emulable, and the "Yes, doctor" scenario fails -- we would have to say "No" to the doctor. The second problem with the idea that the first person experience has to be the sum on all computations, is that this renders duplication of persons impossible. If you duplicate the computation(s) that make up a first person experience, you have simply added some more computations to that experience and the sum over *all* computations is unchanged. Thus there is still only one first person experience, and the attempted duplication fails. So, if the "Yes, doctor" assumption, and the subsequent duplication scenarios, lead to the conclusion that the first person experience is a sum on all computations, the argument is self-contradictory: the conclusion contradicts the input assumptions and the argument is incoherent. Well, you've certainly found a way to make it appear incoherent. But here's another way to look at the matter that may help. The multiplicity of machines implied by the UD are in principle objectively distinguishable so in that sense there is no "sum". However let's consider the situation vis a vis identical machine states, A Turing machine consists of a finite state machine and a semi-infinite tape. When you write "identical machine states" I assume you mean to include the tape. But if the machine states were ever identical then they would remain so - and as abstract TMs they would be the same machine. Or are you referring to some subset of states of the TM which are instantiating "you". Yes, I said exactly that in the continuation of the sentence. Forgive me, but it might help if you read through the whole of my remarks, or at least a whole sentence, before replying. You're right about identical machines though. One must assume enough of a difference for there to be a divergence at some point in the continuations, however slight. any of which might comprise, by assumption, my present subjective state. Now there is no possibility of an objective mapping from the subjective state to a single machine in the multiplicity. As Bruno puts it, if I am indeed a machine, I can't know which one. Because you're not any one in particular? That would be expected if you'rea (quasi) classical being emerging from a quantum physics. But I don't think that's what Bruno meant. I think he means that an algorithm computing "you" can't know which algorithm it is, even if it is a single (not sum) algorithm. Yes, that's what I meant. I think when Bruno says machine he means a discrete computation or algorithm (i.e. machine + tape). Sorry if not clear. In that sense my present subjective state could be considered a "sum" over those identical machine states. So if my brain were then to be replaced by a functionally identical prosthesis there would be no change in that "sum" and so my experience, again by assumption, would be unchanged. I think unchanged experience is too strong a requirement. Your blood has K40 atoms whose decay may very well, on occasion, make a difference in experience or even decision and changing a neuron into some silicon based electronics would certainly change your susceptibility to local EM fields. To say "yes" to the doctor is just supposed to express confidence you'd still be conscious. Well, conscious and in a continuation somewhat consistent with your reasonable prior expectation. So, yes, mutatis mutandis. But I'm still unclear on what "identical machine states" refers to. The UD is running every possible program - not just Nyman1.0. So are you just talking about identical states of a prosthetic brain the doctor installed? No, I'm talking about quasi-identical machine (plus tape) states within UD*. Obviously these must include prosthetic brains at an adequate level of substitution. By assumption, I can't know which of these corresponds at any juncture to my current state. ISTM that this is consistent with a reasonable assumption about the relation between neurocognition and perception. Dementia and related conditions seem to be telling us that our "sense of self" is bound to some finite set of local conditions, to be deliberately vague. Assuming comp, those conditions presumably relate to some finite computational state, though one that is present within UD* (in substantially fungible terms) with extreme redundancy. As an aside, as you must know by now, I consider Hoyle's heuristic a simplifying way of thinking about this, in that it allows one to think quasi-solipsistically about one's present state, taking the absence of direct evidence of other "simultaneous" conscious states semi-literally. IOW, bodies within one's subjective perspective can be considered simply as substantive appearances whose behaviour can in principle be understood physically without invoking any notion of simultaneity of consciousness. After all, consciousness (nor computation, for that matter) cannot be directly observed. Nevertheless one appreciates equally that these selfsame bodies represent other points of view that will be encountered subjectively "in due course" within Hoyle's logical serialisation. David Brent Ditto for duplication of persons. The key point here is to distinguish between the third-person (computational) and first person (perceptual) situations. Of course, when one looks beyond any given present state, divergence of subsequent machine state continuations would then in principle render the associated personal histories both objectively and subjectively distinguishable. Hope this helps. David Bruce So even if a physical universe exists, it *cannot* have any influence on my > prediction. Physics lose *all* its prediction power. Computationalism saves > physics, we should say, but makes it more modest when wandering on > metaphysics. > -- 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 post to this group, send email to [email protected]. 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 [email protected]. To post to this group, send email to [email protected]. 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 [email protected]. To post to this group, send email to [email protected]. 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 [email protected]. To post to this group, send email to [email protected]. Visit this group at https://groups.google.com/group/everything-list. For more options, visit https://groups.google.com/d/optout.
