From stapp@thsrv.lbl.gov Thu Feb 8 17:12:39 2001 Date: Thu, 8 Feb 2001 11:40:26 -0800 (PST) From: stapp@thsrv.lbl.gov Reply-To: hpstapp@lbl.gov To: David Chalmers , Guido Bacciagaluppi , Philip Clayton Subject: Re: Answers (about Searle) ---------- Forwarded message ---------- Date: Thu, 8 Feb 2001 11:35:50 -0800 (PST) From: stapp@thsrv.lbl.gov Reply-To: hpstapp@lbl.gov To: Stanley Klein Cc: hpstapp@lbl.gov, A.sloman@bham.ac.uk, bdj10@cam.ac.uk, brucero@cats.ucsc.edu, hameroff@u.arizona.edu, jmschwar@ucla.edu, klaskey@gmu.edu, phayes@ai.uwf.edu Subject: Re: Answers On Wed, 7 Feb 2001, Stanley Klein wrote: > >[Stan previous] > >In the vN/W ontology, why is there the need for that extra postulate > >associating collapses with feels. > > > >Henry, wouldn't the vN/W ontology have worked just fine without the feels? > > > >[Henry] > >The Orthodox and Copenhagen quantum theories bring in the experiences of > >the experimenter/observers: they are basically about these experiences. > >There IS no existing theory of brain activity that is compatible with > >quantum theory and with the empirical structure of human experience > >aside from vN/W QT, and this theory ties brain activity to > >experiences, i.e., to feels. > > [Stan] > I think we are getting three items mixed up. > > 1) the metaphysical connection of mind and matter. I have always > agreed with Henry that the duality of QT solves many metaphysical > aspects of the Cartesian and Kantian mind/brain dualities. I am interested primarily in dynamical issues within science, not metaphysical ones. Let me explain my answer in more detail. You asked: > >Henry, wouldn't the vN/W ontology have worked just fine without the feels? My answer, spelled out in more detail, was this: The Copenhagen interpretation is basically about "feels", in the broad sense that all human experiences are, in this broad sense, feelings. And science is really, in the final analysis, about relationships between human experiences. So in order to maintain the connection to science von Neumann had to construct a theory that maintained the relationship between "feels" that the Copenhagen interpretation predicted. Hence it is an absolutely essential part of the vN/W interpretation, in my opinion, that once vN made the move of including all atoms and ions, etc. in the quantum system, and hence to include the brains of human beings, he had to tie brain activity to mind. This is a scientific requirement. So my answer is NO, the vN/W ontology would NOT have worked just fine without "feels", because leaving out "feels" (i.e., consciousness/experience) would have cut the connection to science. > > 2) the issue of whether mind acts on brain through mechanisms other > than the neural circuits of consciousness. I do not think there is any question about that: Of course mind acts on brains through the the NCC. That is a basic feature of vN/W theory. But I believe that classical physical theory has a real problem in this connection, unless one goes to identity theory, because if mind is not the SAME as features of the brain, then within CPT mind is epiphenomenal: it makes not one whit of difference whether mind is present or not. My reasons for rejecting the identity theory have been explained at length on this forum. > Standard QT does not say > anything about that as far as I know. The "most standard" QT is Copenhagen QT, which leave brains out. But the core idea of the von Neuman account of the connection between mind and brain is directly via a projection operator P that acts on the brain to effectively actualize the neural correlates of the conscious experience, and eliminate neural activities that are inconsistent with the conscious experience. > A convincing quantitative > calculation of whether the Zeno effect is strong enough to be > efficacious hasn't yet been given as far as I have seen. The neural > circuits seem plenty strong to account for all the behavioral data > I've ever heard of. But what are the laws that govern the neural circuits, when the person is exerting conscious mental effort? Obviously no strictly CPT calculation is possible in the whole of the brain that is involved in these cases. So nothing but unsupported prejudice backs the claim that CPT is adequate, when basic physical principles prove conclusively that CPT CANNOT be adequate. ( I am referring to the uncertainties induced by the Heisenberg uncertainty principle in the determination of whether or not the contents of a vesicle of neurotransmitter will be released.) No CPT computation can be accurate enough to determine the neural activity in cases involving deep moral dilemmas: the quantum noise would be too great, even if the classical noise were not. And certainly nothing even approaching such a confirmation of CPT exists today! As far as empirical evidence is concerned the question of how important MACROSCOPIC quantum are is unresolved, but general principles say that quantum principles should be used unless the adequacy of classical physics is established. > > 3) the extra postulate which was the topic of my question. Let me > frame my question more precisely since this is an important point: > wouldn't the ontology von Neumann presents in his book (unitary > evolution (Process 2) followed by collapse (Process 1)) have worked > just fine without the feels. I had presumed your answer was yes since > last week I thought you agreed that the presence of feels had to be > added as an extra postulate. As I explained above, the connection to "feels" forms the essential link between theory and experience, and hence is crucial to the connection to science. However, let me strongly reinforce my position that although "emergence" is a trivial concept within CPT, because everything is determined bottom-up, and all other descriptions are just alternative ways of describing the consequences of the contact interactions between microelements, "emergence" can be highly nontrivial within QT, because of the causal gaps, which allow emergent constructs/entities to act as wholes, and exert influences on the conglomerates of which they are formed. These influences are not specified or determined by the laws governing the micro-elements. QT provides naturally for this possibility. It provides the mathematical machinery for describing how *nonlocal* process can enter into the physical dynamics in ways that *truly augment* the dynamics in a way compatible with basic science, and in particular with relativistic local quantum field theory. The conscious aspect of our human beingness is a high-level manifestation of this exploitation by nature of this generally available process. But our cinsciousness cannot be the only manifestation, because it surely evolved from simpler applications. So one must see human consciousness as just one special case. In this way "feels", as we know them, are not a necessary part of the general ontological theory, but human knowings are an essential part of that aspect of the general ontology that connects it to human science. So my answer to your question is the "feels", and in fact human knowings, are an essential aspect of QT, as a part of science, but "feels" as we know them, are not an essential part of QT as the basis of an ontological conception of nature. > This postulate bears some similarity to > the epiphenomenal postulate that the consciousness circuits are > associated with feels. (As I said in point 1 above this postulate is > more natural in QT than in CT, but the epiphenomenal postulate could > be added to CT just as well as you can add the postulate to QT). > > No one doubts that feels are associated with brain activity. Within CPT the feels, if they are not *identical* to "consciousness circuits" are necessarily epiphenomenal, and practically everyone agrees that epiphenomenal consciousness is a non-starter. So it makes little sense to consider epiphenomenal consciousness in a QT context: the virtue of QT in this connection is precisely that it allows consciousness to be efficacious without being identical to a classically conceived brain. > >[Stan previous] > >While I am writing, let me ask again a question that I asked before: > >If Mind operates by QT, then we need to ask how Mind decides on which > >question (projection operator, P) to ask (von Neumann process 1): > > > >S -> PSP + (1-P)S(1-P) > > > >This is half of the collapse. I'm not asking how the cross terms got > >eliminated. I'm asking how Mind decided on the P direction. Do we > >need to add new particles and fields to the Lagrangian to specify > >how Mind determines P, or does it operate using the same Lagrangian > >that specifies the quarks and gluons (or superstrings or whatever > >the future Lagrangian of physical world will be)? > > No! The Lagrangian governs the unitarity evolution, the continuous development-in-time governed by the Schroedinger equation. And the quarks etc. specify the physical foundation (the mathematical degrees of freedom) of the objective aspect of reality. The projection-operator-controlled instantaneous jumps are, at the present stage of theory development, a second and different part of the dynamics, not controlled by the Lagrangian. I have given one conjecture about how this second process works: each emergent nonlocal system/entity defines a set {P_i} of projection operators that acts on the set of local degrees of freedom that characterize that system, and the P_i with the largest value of Tr P_i S(t) P_i = Tr P_i S(t) is singled out for "evaluation". This "evaluation" is made in accordance with an Evaluation Operator E defined over the degrees of freedom of the system": E(P_i, t)= Tr E P_i S(t) P_i/ Tr P_i S(t) P_i. The existence of such an evaluation function is the prerequisite for a system to be able to enter into the dynamics. This evaluation function assesses the entire subjective physical system. There must also be a "choice", or "decision" function that acts on the P_i corresponding to the E(P_i,t) with maximum value, and determines whether of not to put to nature the question of whether this P_i will be actualized. This decision can be based on a "quality" of P_i S(t) P_i, which needs to be characterized. The different ways of filling in the details must be conceptualized and tested, much as Newton and Maxwell and Einstein and Heisenberg and Schroedinger did for the physical aspect of the dynamics. Dennett's idea of the "intentional stance", in which one postulates that the process is "maximally rational" (or some such phrase--- I do not have his text before me) would be a good starting point, since the process of Natural Selection would tend to favor the systems that make decisions on the basis of the rationally best evaluative process. I see the theoretical working out of the details, and the empirical testing of various proposals, and hopefully the extensive confirmation of the predictions of one such theoretical scheme, as the fundamental scientific task of the century we have just entered---it took about two millenia to get from Aristotle to Newton. The contribution of von Neumann was to show how the mathematical foundation of this psycho-physical program is provided by a rational analysis of our basic physical theory, QT. > > > >[Kathy, asking a similar question] > >The first question is what P's and question-asking rates are > >possible? After that we ask how Mind selects. > > > >Henry, are you hypothesizing some heretofore unknown force? > > > >[Henry] > >Orthodox theory has not been developed to the point > >of specifying how P is selected. or how the time of > >application is fixed. So we are on our own at this point! > > > >vN/W QT provides a natural framework for linking mind to > >matter in a way that is completely compatible with > >the known laws, but it does not specify the nature of the > >needed added laws/rules, or the machinery that implements > >these laws/rules. > > > >[Stan] > Thanks, I think this is one of the clearest statements of what you > have in mind. You are saying that QT does not specify the nature of > the needed added laws/rules. By that statement it seems you are > saying that there are needed added laws/rules that govern how Mind > chooses the projection operator. So now I come back to my original > question: > > Do you think those needed added laws/rules would be additional terms > in the future Lagrangian (in addition to the terms governing quarks, > gluons or whatever the future Lagrangian of the physical world will > be)? No! > > Or do you think the operation of Mind will require an entirely > different type of physics not specifiable by a Lagrangian? > Yes! Nontrivially emergent macrosystems systems with evaluative decision-making capacities are needed. But it will take careful analysis and experimental work to empirically distinguish between the various possibilities that are compatible with the general vN/W framework. > Or do you think the needed added laws/rules will forever be outside > the realm of physics? I think physics and psychology will become melded: the theory of the mind-brain interaction must include both fields, and neuroscience as well. Cosmology may also come in, because quantum process is closely entwined with the evolution of the universe. But the basic dynamical issues here are scientific ones, although their ramifications extend beyond science. Henry