From: THEORM::STAPP 24-MAR-1997 15:01:45.68 To: @KLEINLIST.DIS CC: STAPP Subj: Re: Reply to Sloman Dear Pat, March 24, 1997 Thanks for the TWO replies. (Fate intervened to give me both of your replies.) The second one is very helpful: focusing our discussion on the key question of `principled distinction' within the framework of various logics is quite specific, and basic to the entire discussion. Settling it should enable us to settle all of these matters. This key question was posed already in your earlier letter (though not quite so emphatically and clearly), and I recognized it already there as the key, and was answering it in the reply I was composing. The most expeditious way to proceed is to begin with the beginning of that reply. I shall argue that the resolution of our dispute lies in this one central fact: "The correspondence between classical physical theory and empirical experiential evidence lies in the matching of theoretically specified spacetime forms with empirically experienced spacetime forms recognizable and communicable by observers" I begin with my response to your first reply---your letter of Mar20. You say: (#...# is my highlight) "Yes and no. I agree that the language of science is an extension and refinement of everyday language. However, I do not agree that everyday language is #entirely# concerned with, or can be translated into a language which is entirely concerned with, 'visual appearances' which are 'associated' with words, or that it does not involve understanding of a theory. Everyday language is absolutely thick with implicit theory: in fact, my own professional life has been devoted to uncovering some tiny parts of this theory. See Aaron's recent response which expresses this point better than I now have time to do here." I did not say #entirely#: The issue here is the nature of the entailment relationship between the physical world and our experiences of it. It was only that special part of ordinary everyday language that I tied to spacetime form and visual appearance. The problem is to disentangle this part from the rest. I absolutely agree that everyday language (and, in fact, almost every adult thought and experience) is absolutely thick with implicit theory. But the more pertinent question is how much (of modern biological) theory is needed to allow the boy gardener to recognize and report that he plucked a `leaf'from a `rose bush', and that the light reflected off of it landed on a spot lying between `8' and `9' on the dial, and to thereby introduce a fact pertaining to a biological system into the biological data base. According to the principles of classical physics, the fact that this spot of light lies between `8' and `9' is ENTAILED by the complete classical physical description of nature, which thus entails a particular biological datum. This example illustrates how #some# of the data upon which biology is based was created: although there is certainly a complex intertwining of the developing biological theory and the growing data base, #some# of that biological data base depends only on theories already woven into the web of our language before the development of modern biological theories. The boy gardener's ability to recognize this leaf as a `leaf', and to report what he did with it, and what he saw, certainly does depend on theory, but not, for example, on any theory about cell biology. Entailment requires definitions: that is your point. The word `leaf' is used in biology, but is not part of the language of physics: it is not found in physics text books. So how can there be entailment from PAF (classical particle and fields) physics to biology? That is your question. My answer is, and has been, that the physics formalism is tied into experiment by means of a suitably refined extension of ordinary everyday language. Without that language the formalism is not physics: it has no connection to experiments or empirical observations. That language has at its base the ordinary words that a child learns to use to describe the objects in his normal life, and to communicate about them to others. This language becomes refined by our theoretical ideas in a kind of step by step way. Certainly one is using the ideas of physics (e.g., that light normally moves in straight lines, to a good approximation; that rigid objects retain their shapes under transportation, to a good approximation) to interprete the experiments, and revisions of the `meanings of what we see' come about as the theory evolves. But this whole process is an evolutionary one, and at various stages one has a contemporary conception of how one's observations are tied into the PAF description. And certain basic facts pertaining to biological systems, such as the fact represented by the boy-gardener's report that he took a `leaf' from a `rose bush' and measured its length to be `definitely more that one inch and definitely less that 2 inches' can be regarded as being part of the classical physics description because `leaf' and `rose bush', in the context of his report, and the meaning of `measuring an object', is part of what is specified by the ordinary language (suitably refined) that classical physics uses to tie the formal mathematical apparatus involving PAFs into experiments and empirical observation. Because the PAF description is expressible in terms of spacetime forms, the primary linkage of PAF to properties of objects identified in the everyday language is to properties that are expressible in terms of recognizable (to trained observers) spacetime forms: the operative correspondences are matchings between the empirical and theoretical (PAF) spacetime forms. Because of this tie-in of the classical physics formalism to ordinary language one can say that the complete classical physical description ENTAILS the results of a certain measurement, even though the object that is being measured (e.g., the `leaf') is defined in this context only to a degree that allows the experimenter to recognize it, and be able to report what he has done to it, and what he has then seen. Using this data about this leaf, and a lot of similar data, the community of classical physical biologists can start to build biological theories that rest ultimately on this kind of data. Hence these theories become tied into an expanding web of scientific language and knowledge that can be regarded as having the PAF description as its base: all the physical facts about the physical and biological objects are presumed to reside in principle in the PAFs. This is the first premise of what I would call "classical physical science". But the other key premise, and this is what is critical here, is that the evidence about the physical system---be it lifeless, or biological, or whatever---is gathered in essentially the same way as it is gathered in physics, by the matching of theoretical and empirical spacetime forms: i.e., by matching spacetime forms specified by the theory to empirically observed spacetime forms. These forms might specify the locations of pointers of measuring devices in relation to numbered marks on dials, or graphical plots, or recogizable and describable shapes of some other kind. But when one pinpoints the place where the interface between the PAF desription and the experienced empirical evidence occurs, it is always this same sort of matching of some spacetime form specified by the theory to some empirical spacetime form recognizable and describable by an observer. This is the central and crucial point. It characterizes the nature of the linkage between classical physical theories and the empirical experience that tie them into the broader realms of human experience. This view is very much a "Bohrian" perspective: science is a human endeavour intertwined with evolving language and knowledge; not some static idealized ossified structure. [I hope I am saying enough about the evolution of science, and the intertwining of concepts, and the importance of concepts, to comfort you.] The central role of spacetime forms in the classical conceptualization is hardly a new thing: it certainly was stressed by Locke. >PH: >"Look, isnt it obvious to you that you are here begging the question? How do >you know they enter so differently into the web of knowledge? Suppose that >by the end of the next millenium we come to understand how mentality is >physically realised. Then it WILL be completely entailed. Or, it may be >impossible to do this. But right now, we can't argue one way or the other. >You havnt pointed to any PRINCIPLED DIFFERENCE between 'experiental >content' and any other aspect of reality, other than a firm conviction you >seem to have that such a difference must exist (on what I am increasingly >coming to think must essentially be religious grounds, since your belief is >so unshakeable and so unjustified.) This is the key question. Let me focus on it, and spell out my answer in still more detail. I have emphasized that a theory is not physical science unless it is related to our empirical experiences about the world: a connection of theoretical ideas to empirical experience is what characterizes physical science in general. Physical science is related to experiments on physical systems, and a key part of the theory, beyond the mathemastical formalism, is what migt be called `the theory of scientific physical observation'. On this point Bohr says: "... we must recognize above all that, even when phenomena transcend the scope of classical physical theories, the account of the experimental arrangements and the recordings of observations must be given in plain language suitably supplemented by technical physical terminology. This is a clear logical demand, since the very word `experiment' refers to a situation where we can tell others what we have done and what we have learned." (Atomic Physics and Human Knowledge, Wiley, NY, 1958, p.72) (See also my book, p.62) This remark applies equally well to all the physical theories: thermodynamics, chemistry, meteorology, neuroanatomy, neurobiology, ... In all of these fields the history is essentially as I have outlined it above: the experimental and theoretical developments lean on each other, and the language and techniques evolve together. But one thing remains stable, and that is the nature of the interface between between evolving theory and the pertinent empirical human experiences `of' the systems being studied, or `of' the measuring devices that are used to get quantitative data about these systems. That interface is always a matching of a spacetime form specified by the theory to a spacetime form recognizable and communicable by some observer. Of course, sometimes an observer uses color, or taste, or sound, or feel to the sense of touch. But any property thus observed can be more accurately, quantitatively, and reliably measured by a device that communicates the result by a displayed number. I think we can get to the core of this key question of the PRINCIPLED DIFFERENCE between physical and phenomenal facts by focusing on one single case. Consider the following three ways that the boy-gardener might report two different aspects of what he sees, depending on his degree of caution. 1A: The spot lies between `8' and `9'. 1B: The spot is green. 2A: The spot looks like it lies between `8' and `9'. 2B: The spot looks like it is green. 3A: It seems to me that the spot lies between `8' and `9'. 3B: It seems to me that the spot is green. At level 1 the assertions are about the spot, at level 3 they are about the observer's `seemings'. Symmetry suggests here that "lies between `8' and `9' " and "is green" are properies of the same kind: that there is NO PRINCIPLED DIFFERENCE between them: the two phrases enter symmetrically into all three pairs of statements. On the other hand, in the classical tradition as described by Locke there is supposed to be a huge difference: "lies between `8' and `9' " specifies a physical fact about the spot itself, whereas "is green" can specify a phenomenal fact that is really about somebody's experience about the spot. Care is required in discussing this point: the three pairs of statements seem to indicate a basic similarity between the "physical" and "phenomenal" facts. So what is the basis of the purported PRINCIPLED DIFFERENCE between them that Locke expounded upon. We are working here within the context of a classical physical theory of nature, and are considering, specifically, our empirical experiences "of" a physical system, namely the spot of light. The words "lies between `8' and `9'" specify experiences lying in each each of two realms, theory and experiment, and for reasons I have described at length in earlier communication, both experiences are of the same kind: they can `match' because both are built essentially from our visual experiences. In the case of "greenness" the property of the *spot itself* referred to by "is green" pertains to the frequency of motions of the PAFs in the neighborhood of the spot. But the corresponding empirical experience is not a direct experience `of' such motions of particles at the location of the spot: one does not directly experience, in the experience of greeness, the fact that the particles near the spot are vibratrating at some particular frequency. This is a PRINCIPLED DIFFERENCE between the two experiences. It seems clear that in the case of "is green" the experience is more directly a property of the brain of the observer than of the spot itself. Of course, the experience of the spot's lying "between `8` and `9' must likewise be a property of the brain of the observer. But in this case this experience is of the type that classical physics uses to build its theory of the world upon. This is a PRINCIPLED DIFFERENCE, given that we are considering the issue from a context in which the classical physical theory is taken to give a correct description of the elements out of which the physical world is made. In contrast, this would be no principled difference at all in the context where the world is accepted as being built out of an intrinsically unified reality in which the aspects of the brain that we now identify as physical and phenomenal are inextricably combined, as in quantum mechanics. So the point is that, although in both cases the empirical experience is (presumeably) an aspect of brain activity, in one case the experience is an experience of the kind that the presumed-to-be-correct theory of physical reality is built upon---and hence the experience can be regarded as an actual likeness of the physical reality that is the object that it purports to represent---whereas in the other case the experience is not an actual likeness of that object as it is represented in the assumed-to-be-correct physical theory. In the first case the experiencing, although it may actually be an aspect of brain activity, `sees through' the brain activity to expose properties that can---insofar as the classical theory is correct---be equated to actual physical properties the object itself. But in the second case the experiencing does not `see through' the brain activity to the object in this direct way. So in the first case one can, within the classical conceptualization, regard the experience as essentially a (to first order) direct experiencing of some physical facts about the object, whereas in the second case this is not possible, because the experience is not an experience of the kind that the classical theory of the object is built upon. Nor is the experiencing in this second case a direct experiencing (even to first order) of real physical facts about the brain, as those facts are represented in classical physical theory. So the point is that the classical physical theory has at the theoretical or formal level a fixed kind of structure based on spacetime form, and the connection of the physical sciences to our empirical experiences must tie into that kind of structure. This tie-in can be direct if the empirical experience itself has the same character as the concepts from which our classical conception was based, namely an experience of spacetime structure, but it is indirect otherwise, and requires some translation or bridging principle. We identify the `physical facts' as the facts represented in the physical description in terms of PAFs, and say that these facts ENTAIL the empirical experiences that appear to be direct experiences `of' these facts as they are reprsented in the physical theory. Experiences that are not direct experiencings of spacetime forms specified by the PAF description are regarded as experiences that are aspects of brain activity that do not `see through' the brain activity to expose directly (at least to first order) some spacetime reality of the object itself, as that the object is conceptualized in the classical physics conception. All experiential facts are phenomenal facts, by definition. But the ones that `see through' the brain activity to seemingly expose directly spacetime structures of the kind that the PAF description entails play a special role, essentially because these kinds of experiences, generalized, are the kind of experiences that classical physics has chosen to build its picture of the world upon. This choice is the basis of the PRINCIPLED DIFFERENCE between general phenomenal facts and physical facts. I turn now to your second reply, and the issue raised there of PRINCIPLED DISTINCTION within the framework of various logical schemes for giving meaning to "entailment". One point is apparent: I am taking physical theory to include a Bohrian theory of `experiments'. This has been my main point throughout. Thus certain aspects of a `leaf' (a B-concept in your terminology) are include in what is meant by classical physics. But these aspects are limited to what it needed to specify the actual experimental set-ups, and the recordings of the observations. These can be described in everyday language (which certainly has a lot of primitive-level theory built in) and includes the words used to describe everyday activities plus `technical physical terminology', to use Bohr's words. This language includes `leaf' limited to mean, for example, what the boy-gardener needs to know to carry out the experiment and make his report of its outcome. Although science can evolve in many ways, it remains physical, at least in my definition, only insofar as the interface between theory and experiment remains Bohrian, with the `physical' in `technical physical interpretation' meaning that the PRINCIPLED DISTINCTION between physical and phenomenal described above be retained In view of my embracing of a Bohrian theory of `experiments' as the centerpiece of both my argument, and of my conception of the meaning of classical physics, I would say that my line of argument falls into the class that Gregg Rosenberg was talking about. But I do not think I fall the trap that you see there, precisely because the Bohrian theory provides for a PRINCIPLED DISTINCTION between physical and phenomenal that can in principle be preserved under the development of the physical sciences, and in fact *has been preserved*, as far as I can see. I see no reason for physical science to suddenly abandon this posture, and think it reasonable to use the term "physical sciences" to characterize only that part of science in which this Bohrian conception of experiment and this distinction is maintained. I think it encompasses a substantial part of biology, and the other sciences that we normally consider to be the physical sciences. Of course, it does not include mere opinions or theories in general, and certainly not all of language and human thought: for that to be encompassed by the complete classical physical description one would need to append to the classical theory the bridging principles that specify, in general, the connection of the phenomenal facts to the physical facts (presumable of the brain). I believe that this explanation of the PRINCIPLED DISTINCTION between phenomenal and physical facts is both necessary and sufficient to answer the various questions raised in your first (MARCH 20) reply: all else hinges on this one point, and all else falls into place once it is understood. I need only mention, in connection with other points in your first reply, that when I say "generally believed" I am *distancing* myself from that opinion, not using the commoness of the belief as evidence for its correctness. Best regards, Henry ----------