hi John, zibbsey here, we left things that I'd get back to you. if ok with you I've some points / questions
On Saturday, November 29, 2014 5:57:38 AM UTC, John Clark wrote: > > > On Fri, Nov 28, 2014 at 6:01 PM, <[email protected] <javascript:>> wrote: > > I was talking about your root idea that Evolution cannot detect >> consciousness >> > > It can't and neither can we. > > > (because we can't, I think you said) >> > > The reason isn't because of us, it's just that neither we nor Evolution > nor anything else can detect consciousness other than our own, we can only > detect actions. > You are trying to perform a generalization, of two very different conceptions above, adding a third below in the form of Turing test. Generalizations of that kind are extremely hard to accomplish. The *Human Context*: Humans are yet to make dramatic progress defining consciousness, locating it in the brain, and so on. Humans are restricted Actions in what they have the potential to detect . Humans have abstract theory, for example. A large component of what abstract theory detects, are INVARIANT features. Which aren't Actions. The *Natural Selection* Context.. NS is an Abstract.. Not due to some preference, but fundamentally What was attracting selection, at what MAGNITUDE WITH WHAT LONG TERM IMPACT, whether it reaches unity in the population, or indeed whether organism gets eaten and the selection event simply did not happen. All of this does not begin to manifest into trait characteristics until generations,, millennia even, after the first signs of that through manifestation is attracting natural selection first appear. NS is fundamentally Abstract. That's the Invariance that m akes possible everything else to be exchangeable. What you want to do with the concept - your goal - drives natural selection from one structure to the next very different structure. e.g. if it's about fundamentals the structure might be 'replicators'; -if its about what trait was attracting selection at what magnitude, and how far trait went...whether it became unity in the population, then NS for that context is defined by two temporal points; cannot be tied to just one alone. This is because the characteristics being queried do not begin to manifest for generations , millennia even, after the first signs of that trait attracting selection first appear. AND ONLY manifest at all if there is sufficiently unbroken selection from that first appearance right through to that manifestation. AND ONLY if that interval of selection exhibits exponential effects. I could go on: Only if the source of the exponential effects correspond to the laws of population genetics. The point is John NS is an Abstract. A term such as "NS detects...." must be restricted to the Brittle Metaphor the phrase gets coined to furnish. Put differently, "Evolution detect...." is a slight variation of say "Evolution prefers..." or "Evolution creates...." or "kills..." They are strictly metaphors John. There no real overlay - at al - that is non-trivial with the ways "Humans detect...." has a range of contexts. We "detect" in literal, if abstract ways tied in closely with our temporal and macroscopic - constrained reality. Humans detect things via physical. Very often Actions as you say (not only Actions). Evolution has essentially zero non-trivial evolutionary dependence on Actions. For natural selection to 'detect' an 'action' the action would have to become INVARIANT an abstract space that was also Time INVARIANT, in that instance caused by there being no "t" term in the equation. Instead it becomes about generations. The Action consistently manifests given the same conditions to some resolution. The same conditions giving rise to the same Action consistently re-manifests in successive generations. In reality, it would go to deeper resolutions, like only the sub-region of the Action, that the most instances of recurrence fully overlay through all Actions, will being to approach the closest context in which evolution 'detects' ....'Actions'. *The Turing Test Context* The Turin Test is something wholly different again. The reason you are logically flawed here, is that Alan Turing specified - explicitly - an indirect detention of consciousness via everything another consciousness is able to throw at a notional black box, without I/O exhibiting traits of full intelligence. The formulation of the Turing Test makes no statement at all about the relationship of intelligence to consciousness. T There is no commitment implied or stated to a fundamental hard linkage of intelligence and consciousness. This is because Turing Tests for Consciousness given a blackbox I/O hallmarking intelligence.. NOT BY measuring the Intelligence. If that was the case the measure becomes highly standardizable. Which equates to being suited for AUTOMATION in which condition the there is no requirement for the consciousness entity stipulated in the Test. So it vanishes, which causes the test to vanish. Alan Turing clearly, observably, does not go in this direction. What he tests for is the presence of CONSCIOUSNESS THROUGH AN INTERFACE NECESSARILY EXHIBITING INTELIGENCE That intelligence is defined such that consciousness cannot be ruled out on technical grounds. Nor rule IN on technical grounds. Turning argues instead that a conscious life form - a species - has must embody the potential to come up with a formula that maximizes the Consciousness-Loading and minimizes the INTELLIGENCE by leveraging increasingly simple yet punitive terms, by techiques including things like reflecton ad introspection, see enrichment of the consciousness element. For example, you do appreciate I hope ad understand that Turing at all times defines intelligence in terms of some whole. thing, the same way that human intelligence is a whole thing. Turn NEVideER diverges from the holistic standard to the components . He doesn't. And you do, and then go much further. Through lo Purpose specific intelligence such as DEEP BLUE (and still you carry on on, apparently all the way to generic computation which involved Chess. Turning did not suggest that, in fact indicated the e opposite. Turing's Test would have the consciouis being evaluate Deep Blue All ways and any, save Chess. > What I showed in was that natural selection will detect any kind of >> difference between the same traits in two individuals, >> > > Only if those different traits produce different actions. If a intelligent > but non-conscious animal behaves differently than a intelligent and > conscious animal then Evolution can detect that and so can the Turing Test > No john. because the Turing Test specifies a specific route to detection. You'd need a different test for detecting the difference in that animals behaviour. > And Evolution will favor whichever behavior is smarter, and if I'm correct > and you can't have intelligence without consciousness then that would make > Evolution's choice easy. > It's obviously not my gift to say.But based on the fragilities and misconceptions and errors reasoning I should think it's reasonable to assume you are not necessary even wrong :o) Just kidding > > > > Give me a argument that Evolution can see consciousness and I'll either > give you a counterargument or concede and thank you for correcting my > error, but so far all I've heard is that consciousness makes a animal > behave differently, something I already knew MUST be true or Evolution > would have never produced it. And if it effects behavior then the Turing > Test must work for consciousness too because lack of consciousness implies > lack of intelligence and that implies lack of intelligent actions. > > 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]. To post to this group, send email to [email protected]. Visit this group at http://groups.google.com/group/everything-list. For more options, visit https://groups.google.com/d/optout.
