Posted Tue Aug 13 06:04:04 BST 1996 Newsgroups: sci.psychology.consciousness References: Subject: Re: Orch OR and the self Keith Sutherland writes: > Date: Mon, 12 Aug 1996 11:52:35 +0100 > Why is it that consciousness studies is so different from every other > field? Simple -- you can't see consciousness, you can't touch it, you > can't measure it, Many things scientists talk about have started like that and ended up otherwise. Genes for example? I suppose consciousness is special in that some people wish to protect it from any such change, and stipulate that by definition it should be immune. [KS] > ...so we ignore it and start talking about AI, space- > time, microtubules, Shannonian information and thalamocortical > oscillations. At least Sloman, Dennett etc. are consistent and say that > this is because there's nothing there to study in the first place. I can't speak for Dennett, but I have never said there's NOTHING there to study. On the contrary, I've consistently claimed that there are lots of things to study all referred to more or less vaguely by words like "conscious", "aware", "experience", "freedom", etc. All I've disputed is the existence of some *unique* thing which is either present or absent as a whole. If what we call "consciousness" is a loosely defined cluster of more or less closely related things then we can study what's in the cluster and how different subsets of the cluster can occur in different animals, in new born infants, in people with brain damage, in people from different cultures, and maybe in machines of the future. Thus, I've not disputed that there is a difference between being asleep and being awake, or a difference between dreaming and dreamless sleep, or a difference between being aware of your toothache and not being aware of it (e.g. because you are temporarily distracted by a terifying threat), or a difference between being aware of what you are doing and doing it automatically without awareness. Similarly there's a difference between the perceived shape of the table-top (rectangular) and the (non-rectangular) shape I can attend to which changes with viewpoint when I look at the table. These are all data to be explained by a good theory. Similarly, in previous postings I described a collection of data to do with differences between what we can and cannot (normally) control in our own thinking and other mental processes. These are all real phenomena. Responding to Stuart Hameroff Keith writes: > Like most people, Stuart has the cart well and truly up-front of the > horse. In any other area of scholarship you start with the data and > attempt to build a theory to fit, That's how psychologists and social scientists are taught what science is (or how they used to be taught in the UK: I have not checked recently). But it's not really how ALL the great scientists have worked (e.g. Newton, Einstein? I am not sure about Darwin). This does not mean they've ignored the data: merely that they've been provoked by a subset of data and dissatisfaction with previous theories into searching for a new theory. That, in turn may lead to a search for new data (e.g. the search, in 1919, for evidence of light bending in the vicinity of the sun). Of course, data are important. But getting the description of the data right is also important. One of the features of data relevant to discussions of consciousness is the general imprecision, hand-waving, ungrammatical nonsense, and theory-based prejudice that pervades the description of the data as soon as we put on our philosophical or scientific hats. So I am with you and I think we have to be FAR more careful in describing the data that we think we are trying to explain. In my own case, I have tried meticulously (e.g. in the paragraph above starting "Thus, I've not disputed..", and in my previous messages about control) to start only with descriptions that would naturally be regarded as correct by normal English speakers (I apologise for not knowing enough about any other language). The trouble is that too often people on both sides of almost every disagreement about consciousness quickly move from patient analysis of the detail to grand sweeping generalisations. They then quickly lose contact with reality. By the way I don't feel *committed* to any claim about computers as we now know them being capable of being programmed so as to exhibit all the phenomena. In 1978 I thought that was obvious. Since then I've learnt that there are several essentially different kinds of mechanisms all of which might, or might not, turn out to be needed, either in principle or in practice (e.g. because of physical constraints). It's an open question. All the arguments so far put forward to suggest that mechanisms of type M1 are inadequate or mechanisms of type M2 are essential (for various examples of types M1 and M2, symbolic comptuations, neural nets, quantum mechanisms), strike me as invalid, either because they are based on inadequate characterisations of what is possible for such mechanisms, or because they are based on inadequate characterisations of what needs to be explained. Or both. (Everything I've read about Orch OR or other new versions of Quantum theory, seems to fall into the latter category. I.e. they may be important new contributions to our understanding of physics or neuromechanisms but the alleged links with data regarding consciousness seem to be spurious. But I can't claim to have understood it all. One argument for the necessity of quantum mechanisms that does sound prima-facie convincing to me is that if you need to be able to store vast amounts of data in a relatively small space with very rapid changes possible using small energy consumption, where the data representations are not subject to serious degradation caused by thermal buffeting, then maybe quantum mechanisms are required because classical mechanisms (e.g. spring-loaded flip flops) could not be sufficiently stable without being too large, too slow, or too energy hungry. But proving this formally would require more mathematics than I can muster. This is a non-spurious link with human consciousness because one aspect of the relevant data concerns the information richness both of the content of experience and of the background knowledge involved in such experience, and the rapidity with which the content can change, e.g. as you look first in one direction then another.) Arguments purporting to show that NO form of mechanism can suffice seem to me usually to involve either undiscussable premises (e.g. theological or ethical commitments) or conceptual confusions that are very hard to shift. > But those of us who believe that consciousness is (in some sense) real > should start with the data. Yup! But beware of temptations to invent spurious data. This is commonplace in philosophical discussions about the mind, especially among people with a strong DESIRE for the truth to lie in a certain direction. With some of my students it often turns out helpful to suggest that they contemplate various theories and ask of each: Would I *care* if that turned out true? If so why? It's a useful way of uncovering unacknowledged blinkers. E.g. some people fear for their dignity or humanity if certain theories are true. An honest example is Joseph Weizenbaum, Computer Power and Human Reason: From Judgement to Calculation W.H.Freeman 1976 I find it very strange. Aaron ===