On 1 March 2014 01:40, Bruno Marchal <[email protected]> wrote: >> If you start with the assumption that the physics relevant to brain >> function is not computable then computationalism is false: it would be >> impossible to make a machine that behaves like a human, either zombie >> or conscious. > > > I agree with you, the physics *relevant* to brain function has to be > computable, for comp to be true. But the point is that below the > substitution level, the physical details are not relevant. Then by the FPI, > they must be undetermined, and this on an infinite non computable domain, > and so, our "computable brain" must rely on a non computable physics, or a > non necessarily computable physics, with some non computable aspect. This is > what comp predicts, and of course this is confirmed by QM. Again, > eventually, QM might to much computable for comp to be true. That is what > remain to be seen. > > >> What I mean by functionalism is that the way the brain processes >> information, its I/O behaviour, is what generates mind. This implies >> multiple realisability of mental states, insofar as the same >> information processing could be done by another machine. If the >> machine is a digital computer then functionalism reduces to >> computationalism. If the brain utilises non-computable physics then >> you won't be able to reproduce its function (and the mind thus >> generated) with a digital computer, so computationalism is false. >> However, that does not necessarily mean that functionalism is false, >> since you may be able to implement the appropriate brain function >> through some other means. For example, if it turns out that a digital >> implementation of the brain fails because real numbers and not >> approximations are necessary, it may still be possible to implement a >> brain using analogue devices. > > > OK, but that functionalism seems to me trivially true. How could such > functionalism be refuted, if you can invoke arbitrary functions? > (Also, "functionalism" is used for a stringer (less general) version of > computationalism, by Putnam, so this use of functionalism is non standard > and can be confusing. > Last remark, I am not sure that the notion of information processing can > make sense in a non digital framework. In both quantum and classical > information theory, information is digital (words like bits and qubits come > from there).
I think functionalism is true, but it's not obviously true, at least to most people. It could be that the observable behaviour of the brain is reproduced perfectly but the resulting creature has no consciousness or a different conscious. That would be the case if consciousness were substrate-dependent. It could also be that the behaviour cannot be reproduced by a computer because the substitution level requires non-computable physics (true randomness, real numbers, non-computable functions), but it could be reproduced by a non-computational device. So there are these possibilities with brain replacement: (a) the behaviour is not reproduced and neither is the consciousness; (b) The behaviour is reproduced but the consciousness is not reproduced; (c) The behaviour is reproduced and so is the consciousness; (d) The behaviour is not reproduced but the consciousness is >> What can be proved is that if consciousness is due to the brain then >> replicating brain function in some other substrate will also replicate >> its consciousness. > > > OK. What I meant is that we cannot prove that consciousness is due to the > brain. Yes, a dualist, for example, could consistently deny fuctionalism, but someone who believes that consciousness is due to the brain could not. -- 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.
