On Tuesday, January 28, 2014 8:37:04 AM UTC-5, stathisp wrote: > > On 27 January 2014 16:07, Craig Weinberg <[email protected] <javascript:>> > wrote: > > >> Do you think Barack Obama is conscious? If you do, then in whatever > sense > >> you understand that, can the Chinese Room also be conscious? Or do you > think > >> that is impossible? > > > > > > Yes, I think that Barack Obama is conscious, because he is different > from a > > building or machine. Buildings and machines cannot be conscious, just as > > pictures of people drinking pictures of water do no experience relief > from > > thirst. > > If Barack Obama revealed that he was a machine, would that change your > view of whether machines could be conscious? >
If nobody ever survived a translation into machine simulation, would that change your mind of whether consciousness can be simulated? > > The Chinese Room is not important. You are missing the whole point. > > Consciousness is beyond reason and cannot be discovered through evidence > or > > argument, but sensory experience alone. > > So the Chinese Room is conscious not through evidence or argument, but > through sensory experience alone. > It would be if the Chinese Room had sensory experience. Our experience of the Chinese Room doesn't matter. > > >> The claim is that the consciousness of the room stands in relation to > the > >> physical room as the consciousness of a person stands in relation to > the > >> physical person. > > > > > > There is no 'physical person', there is a public facing body. A person > is > > not a body. On one level of an animal's body there are organs which > cannot > > survive independently of the body as a whole, but on another level all > of > > those organs are composed of living cells which have more autonomy. > > Understanding this theme of coexisting but contrasting levels of > description > > suggests that a room need not be comparable to the body of a living > > organism. Since the room is not something which naturally evolves of its > own > > motives and sense, we need not assume that the level at which it appears > to > > us as a room or machine is in fact the relevant level of description > when > > considering its autonomy and coherence. In my view, the machine > expresses > > only the lowest levels of immediate thermodynamic sensitivity according > to > > the substance which is actually reacting, and the most distant levels of > > theoretical design, but with nothing in between. We do not have to > pretend > > that there is no way to guess whether a doll or a cadaver might be > > conscious. With an adequate model of qualitative nesting and its > relation to > > quantitative scale, we can be freed from sophism and pathetic fallacy. > > An observer might say that the Chinese Room or the AI in "Her" or > Barack Obama "naturally evolves of its own motives and sense" after > the point of creation. > Those are fictional examples. I don't deny that many people find it plausible that machines can evolve their own motives, but I think that I understand why they are mistaken. > > >> It could not become John Wayne physically, and it could not become John > >> Wayne mentally if the actual matter in John Wayne is required to > reproduce > >> John Wayne's mind, but you have not proved that the latter is the case. > > > > It has nothing to do with matter. There can only ever be one John Wayne. > A > > person is like a composite snapshot of a unique human lifetime, and the > > nesting of that lifetime within a unique cultural zeitgeist. It's all > made > > of the expression of experience through time. The matter is just the > story > > told to us by the experiences of eyeballs and fingertips, microscopes, > etc. > > What if it were revealed that John Wayne's body while he was asleep on > the night of his 40th birthday was annihilated and replaced by a copy? > Would you still say there can only be one John Wayne? What if you were annihilated at age 5, but you were replaced by a copy? Would you still say that the body using your name was you, even though you are not using it? > Why couldn't we > make a John Wayne Mk3 long after his death, who would stand in > relation to John Wayne Mk2 as John Wayne Mk2 stood in relation to John > Wayne Mk1? > For the same reason that we can't build a model of Paris out of clay and have it actually become Paris. It is because the publicly measurable end of what we are is not sufficient to describe who we have been and who we are becoming. > > >> That's what Searle claims, which is why he makes the Room pass a Turing > >> test in Chinese and then purports to prove (invalidly, according to > what > >> you've said) that despite passing the test it isn't conscious. > > > > > > The question of whether or not a Turing test is possible is beyond the > scope > > of the the Chinese Room. The Room assumes, for the sake of argument, > that > > Computationalist assumptions are true, and that a Turing type test would > be > > useful, and that anything which could pass such a test would have to be > > conscious. Searle rightly identifies the futility of looking for outward > > appearances to reveal the quality of interior awareness. He successfully > > demonstrates that blind syntactic approaches to producing symbols of > > consciousness could indeed match any blind semantic approach of > expecting > > consciousness. > > And he purports to do that by showing that despite external > appearances the Chinese Room cannot be conscious because the > components are not conscious, which you agree is not a valid argument. > It's not because the components are not conscious, it is because the description of the Room or 'system' as a whole is consistent with the absence of consciousness. This is the double standard of functionalism which makes the hard problem hard. Functionalism asserts on the one hand that all functions of consciousness can be produced mechanically, but then the effect of consciousness could not provide any additional functionality to the mechanism. The hard problem exposes the hypocrisy of 'We don't need consciousness to explain everything" when consciousness itself is what needs to be explained. The machine's definition of itself does not include consciousness. Building machines based on that definition will therefore not include consciousness. > > My hypotheses go further into the ontology of awareness, so > > that we are not limited to the blindness of measurable communication in > our > > empathy, and that our senses extend beyond their own accounts of each > other. > > Our intuitive capacities can be more fallible than empirical views can > > measure, but they can also be more veridical than information based > methods > > can ever dream of. Intuition, serendipity, and imagination are required > to > > generate the perpetual denationalization of creators ahead of the > created. > > This doesn't mean that some people cannot be fooled all of the time or > that > > all of the people can't be fooled some of the time, only that all of the > > people cannot be fooled all of the time. > > You still haven't come up with any reason better than a vague > prejudice why, for example, the AI in the movie "Her" could not be > conscious. > Because she doesn't need to be conscious. She could just as easily be programmed as a chameleon, which analyzes the profile of each user and builds an ad hoc identity to reflect some Bayesian extraction of their preferences. Would such a chameleon have to be all of the different people that it pretends to be? Craig > > > -- > Stathis Papaioannou > -- 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/groups/opt_out.
