From hpstapp@lbl.gov Wed Apr 9 17:33:52 2008 Date: Wed, 9 Apr 2008 17:33:51 -0700 (PDT) From: Henry P. Stapp To: Marcos Giusti Subject: Re: Lucas-Penrose thesis. On Wed, 9 Apr 2008, Marcos Giusti wrote: > Dear Dr. Stapp, > > As I was thinking of building an ontology consistent with the idea of interaction between mind and brain, I was led to consider the primary elements of the reality like aspects of a not antropomorphical consciousness. But I found a difficulty here: if consciousness is not an human attribute, an automaton can have consciousness. However there's the Lucas-Penrose thesis that states that an automaton can not have consciousness (minds cannot be modelled as machines due to Godel's Incompleteness Theorem) How can I refute it? Or is this a false problem? > > Cordially, > > M. Giusti > > The issue here revolves around the fact that a "computer" is supposed to be a system that processes data by means of algorithms, which are processes involving discrete well-defined-step-followed-by-a-next-well-defined-step. Penrose argues, on the basis of Godel's theorem, that human thinking, and in particular the conclusions arrived at by mathematicians are sometimes non-algorithmic; the intuitive grasping of the "truth" of a certain argument seems to involve something beyond a purely algorithmic process. But it is not clear that Penrose's Godel-based argument is actually valid (See the final entry on my web site). And even if it were correct, there is the qustion of whether the judgements of mathematians really do demand non-algorithmic processes. I suspect that if we had purely algorithmic computers with the complexity of a top human brain, then it could make judgements about the validity of mathematical proofs that are as good as those of a top mathematician. On the other hand, the choices that we make that are associated with our process-1 physical actions are not explained by currently known physical laws. So it is, as far as I can see, an open question whether the ultimate causal roots of these choices lie completely in the physical realm, but act via as-yet-undiscovered laws, or whether these choices can have causal roots that cannot be traced to the quantum-physically described aspects of reality. If our conscious choices are not ultimately causally rooted completely in the physical, then the question is: what sort of quantum mechanically described physical structures are susceptible to influences from outside the physical realm. The first question is whether physical systems that have the requisite structure simply sometimes accidently exist; or whether the systems that have the property of being influencible by non-physical aspects of reality have this influencibility property by virtue of the fact that they were constructed by a process that itself has non=physical causal roots? This latter possibility may be attractive to you---namely that systems that enjoy the property of being influencible by non-physical causes are systems that have been constructed by a process that is itself influenced by causes not traceable to quantum-physically described aspects of our science-based conception of reality.