abbott@aerospace.aero.org (Russell J. Abbott) (03/06/90)
Replying to Ken Presting who writes: | ... ||In the second place, the dynamical properties of a running computer are ||neatly determined by its program. The program is of course a static ||object .... || Stephen Smoliar writes: | ... [I]f there were a clean |relationship between the static properties of a program and the dynamic |properties of the device running that program, we wouldn't have all the |software problems we have, would we? ... |I would argue that the reason for this is |that we still lack good ways to describe and reason about the dynamic |properties of processes. The best we have been able to do, thus far, |is the abstract those processes into static objects. ... [B]ut the |power of this approach is only as good as the abstraction |we develop. Finding the right abstraction often remains the intractable |problem in software engineering. But we are often quite successful in building static objects (programs) that exhibit the dynamic properties that we desire, e.g., language processors/interpreters. So we do know how to produce emergent properties in a great many cases. In those cases we call it building levels of abstraction. On the other hand, we don't know how to program neural nets. (I'm distinguishing "training" from programming.) So I wonder whether there is a well characterizable difference between programmable levels of abstraction and "emergent" levels of abstraction. Or will we eventually be able to program any system that is capable of exhibiting emergence once we develop the right abstractions. One obvious difference between most explicitly programmed levels of abstraction and most emergent systems is the degree of parallelism. Is it likely that the abstractions we will have to develop to program highly parallel emergent systems will resemble in their discreteness those we now use to program traditional levels of abstractions. Or will they necessarily be more statistical and hence always resemble "training" or environmental molding more than programming. -- -- Russ abbott@itro3.aero.org
smoliar@vaxa.isi.edu (Stephen Smoliar) (03/07/90)
In article <67994@aerospace.AERO.ORG> abbott@aerospace.aero.org (Russell J. Abbott) writes: > >One obvious difference between most explicitly programmed levels of >abstraction and most emergent systems is the degree of parallelism. Is >it likely that the abstractions we will have to develop to program >highly parallel emergent systems will resemble in their discreteness >those we now use to program traditional levels of abstractions. Or will >they necessarily be more statistical and hence always resemble >"training" or environmental molding more than programming. I'm not sure if this question was intended to be rhetorical, but I'd like to invoke the old rabbincal trick of addressing it with another question. Let us consider the cellular automata of Conway's Life as a source of emergent properties. What sorts of abstractions would we need in order to describe the consequences of starting with the R-pentomino? (Obviously, the "raw" rules of the cellular automata do not tell us what we need, since they cannot tell one initial pattern from another. Also, I chose the R-pentomino because it eventually hits a stable state; but only after some time. Therefore, I do not think describing that final state by itself constitutes a satisfactory abstraction. By the way, when you finish this one, I have this performance by the Kronos Quartet in need of some abstraction . . . :-) ) ========================================================================= USPS: Stephen Smoliar USC Information Sciences Institute 4676 Admiralty Way Suite 1001 Marina del Rey, California 90292-6695 Internet: smoliar@vaxa.isi.edu "Only a schoolteacher innocent of how literature is made could have written such a line."--Gore Vidal
kp@uts.amdahl.com (Ken Presting) (03/18/90)
In article <14431@phoenix.Princeton.EDU> eliot@phoenix.Princeton.EDU (Eliot Handelman) writes: >In article <b1UF028Q90Pd01@amdahl.uts.amdahl.com> kp@amdahl.uts.amdahl.com (Ken Presting) writes: >;. . . Susanne Langer, ... from "Feeling and Form" (1952) >; >; "The tonal structures we call "music" bear a close logical >; similarity to the forms of human feeling ... forms of growth >; and of attenuation ... speed, arrest ... the greatness and >; brevity and eternal passing of everything vitally felt. Such is >; the pattern, or logical form, of sentience; and the pattern of >; music is that same form worked out in pure, measured sound and >; silence. Music is a tonal analogue of emotive life. >; Such formal analogy, or congruence of logical structures, is >; the prime requisite for the relation between a symbol and whatever >; it is to mean. The symbol and the object symbolized must have >; some common logical form." >; . . . >;Returning to the topic of abstraction and computation, I want to use the >;second paragraph of the quote from Langer. . . . > >;Consider a formal theory of arithemetic, such as Peano's. . . . >;. . . there is a homomorphism under deduction between >;arithmetic and set theory. Michael Resnick applies a similar idea to >;the philosophy of mathematics, Langer applies it to symbolism in art, > >I don' really understand what you're trying to get at, maybe you could >clarify? >--eliot handelman (Eliot expanded on this question via E-mail:) >. . . The idea that music, sed Langer, is the "logical form" of emotions >would mean, in your terms, that there is a homomorphism between music and >emotions. I'm curious to see if this is what you're saying, and why you >think this might be. It's important (at least for logical purity) to distinguish between the claim that music is the logical form of emotions (which neither Langer nor I believe) and the claim that music and emotions have the same logical form (which is closer to the position I have adopted from her work). The first problem is that no actual object or event literally has a logical structure. Only a set of assertions has a logical structure. Even a set of assertions may have a vague or ambiguous logical structure, if the assertions are informal, etc. It makes some sense to talk about the logical structure of music or emotions by extension - a true description of a musical performance has (literally) a logical structure, which may then be attributed to the performance. Likewise, a phenomenological description of an experience, emotional or otherwise, has a logical structure. So the homomorphism of logical structure exists between the descriptions, rather than between the phenomena. The analogous relation which obtains between the phenomena is *resemblance*. The concepts of logical structure and homomorphism serve to analyze resemblance. (Homomorphism is a suggestive analog for resemblance, but does not literally apply outside the context of formal algebraic structures.) The role of descriptions in this analysis introduces a number of serious problems. Langer does *not* take this approach, and avoids most of the problems. I am primarily interested in philosophy of science and metaphysics, so I prefer to highlight the problems of descriptive abstractions. This makes my analysis more cumbersome in application to the experience of music - the selection of a descriptive abstraction becomes a central theoretical issue, rather than a practical matter to be decided according to the interests of a composer, reviewer, or listener. (Btw, Langer seems to be more influenced by Clive Bell's concept of "Significant Form" than by Wagner directly) On the other hand, selection (or invention!) of a descriptive framework is naturally a central issue for physical sciences, AI, and psychology of perception. So I think the concept of logical structure and homomorphism organizes the problems in a useful way. In particular, phenomenology and physics each purport (or at least strive) to offer a general descriptive abstraction, adequate for the expression of any insight or hypothesis in their respective domains. The theory of computation makes a similar claim within its own domain. I propose that AI is the attempt to find an "alignment" (or, better, *implementation*) between these three domains, such that the truth of each claim is simultaneously shown to be true. I regard it as a very happy accident (but not a very surprising one) that Langer, whose thought is motivated largely by the aesthetics of music, provides the concept that just might do the trick. Ken Presting
eliot@phoenix.Princeton.EDU (Eliot Handelman) (03/18/90)
In article <2cSD02rC93BD01@amdahl.uts.amdahl.com> kp@amdahl.uts.amdahl.com (Ken Presting) writes:
;(Eliot expanded on this question via E-mail:)
;>. . . The idea that music, sed Langer, is the "logical form" of emotions
;>would mean, in your terms, that there is a homomorphism between music and
;>emotions. I'm curious to see if this is what you're saying, and why you
;>think this might be.
;
;It's important (at least for logical purity) to distinguish between
;the claim that music is the logical form of emotions (which neither
;Langer nor I believe) and the claim that music and emotions have the
;same logical form (which is closer to the position I have adopted from
;her work).
I seem to remember Langer having endorsed the former view. I can't
find my copy, but I have the following Wagner quote she used as an
introduction to the relevant argument:
"What music expresses is external, infinite and ideal; it does not
express passion, love or longing of such-and-such an individual on
such-and-such an occasion, but passion, love or longing in itself, and
this it presents in that unlimited variety of motivations, which is
the exclusive and particular characteristic of music, foreign and
inexpressible in any other language."
This, she writes, is "the most explicit rendering of [a] principle" that
she describes as "the most persistent, plausible and interesting
doctrine of meaning in music [...] "(219-220) Langer is trying to show,
among other things, what music actually does, and among the things it
appears to do is not merely to communicate emotion (a term that she's
trying to get away from) but even to induce it. The historical context
of her argument was the work of Wundt and his school, who thought the
proper way to subject music to scientific scrutiny was to figure out
which sounds were "pleasing" and which "less pleasing," treating music
as a succession of physically pleasing and less pleasing moments.Langer
opposed this impoverished account of musical experience in favor of
a transcendental view, motivated by the Wagner excerpt above,
that promoted the perception -- here one really should say "cognition" --
of form rather of sound. The conundrun here is how music, which, as she
says, is nonpropositional, is able to communicate anything at all, except
in the uninteresting sense of drawing upon associations (military music, etc).
Her answer is roughly that music is an acoustic rendering of the shapes of
emotions, and that the listener is drawn into the emotion by virtue of
being in the presence of the artifial emoting organism that is sound.
The point is very much that music IS, according to Langer, a "logical"
(meaning transcendental) form of the emotions.
;It makes some sense to talk about the logical structure of music or
;emotions by extension - a true description of a musical performance
;has (literally) a logical structure, which may then be attributed
;to the performance. Likewise, a phenomenological description of an
;experience, emotional or otherwise, has a logical structure.
(Again, replace "transcendental" for "logical." She's being idealistic,
not empiricist, which I assume is what you're promoting when you say
a "true" description.)
;So the homomorphism of logical structure exists between the descriptions,
;rather than between the phenomena. The analogous relation which obtains
;between the phenomena is *resemblance*. The concepts of logical
;structure and homomorphism serve to analyze resemblance.
Ken, I think you've just negated the whole value of introducing musical
approaches to the study of emergent processes, unless you're using some
terms here in ways that I don't understand. "Logical" means to me
(roughly following Langer's usage) "abstracted to its bare but vital
properties." You are using the term in the "what you see is what you get"
approach. Even if you mean "logical" in my sense, I still can't see in
what way a description of a performance has to do with a phenomenological
account of listening experience, unless you think of the listener as
a sort of box into which music is poured, and to open up the box to discover
the experience is to rediscover the music sitting inside him.
;The role of descriptions in this analysis introduces a number of serious
;problems. Langer does *not* take this approach, and avoids most of the
;problems.
That's your approach, I suspect. Let me try to explain Langer's idea in
a way that Langer may not have actively pursued, but wehich is, I think,
germane. You may or may not be familiar with the James-Lange theory of
emotional response. It says that an emotion is the perception of bodily
responses, avoidance reactions and the like. It was discounted in the 20's
I believe and in the past few years has been revived by experimenters who
have found high correlations between facial expressions and associated
emotions -- if you are forced to smile a lot you start feeling happy (tell
that to the Japanese, but nevertheless). Let's suppose that there's some
sort of bodily MOTION whose perception is the eMOTION -- acupuncturists
seem to believe thsi. Let's assume, futher, that the determining motion
can be provoked by external means -- specifically by vibratory means. Then
you have a theory that music physically evokes this motion, which then
becomes emotion.
Note that Langer is decidedly not saying this, sinmce music is for her
NOT sound, whereas in this theory music is ONLY sound. In any case the
idea that physiological response to sound may be part of music can no longer
be challenged -- effects of noise haved recieved some attention since
Langer. My point here is that if any of this theory is at all true, and
some small part of it is, I think, then tensions and movements are being
established in the listener that must correspond to physiological encodings of
emotional experience. Langer is suggesting something similar, though not
necessarily quite as mechanistically as in this particular theory.
I am primarily interested in philosophy of science and
;metaphysics, so I prefer to highlight the problems of descriptive
;abstractions. This makes my analysis more cumbersome in application
;to the experience of music - the selection of a descriptive abstraction
;becomes a central theoretical issue, rather than a practical matter to
;be decided according to the interests of a composer, reviewer, or
;listener.
Well, maybe the point here is that you can think of music as a descriptive
abstraction of more general experience. Not one that does anybody any good
except for trained musicians.
;On the other hand, selection (or invention!) of a descriptive framework
;is naturally a central issue for physical sciences, AI, and psychology
;of perception. So I think the concept of logical structure and
;homomorphism organizes the problems in a useful way. In particular,
;phenomenology and physics each purport (or at least strive) to offer a
;general descriptive abstraction, adequate for the expression of any
;insight or hypothesis in their respective domains. The theory of
;computation makes a similar claim within its own domain.
;I propose that AI is the attempt to find an "alignment" (or, better,
;*implementation*) between these three domains, such that the truth of
;each claim is simultaneously shown to be true. I regard it as a very
;happy accident (but not a very surprising one) that Langer, whose thought
;is motivated largely by the aesthetics of music, provides the concept
;that just might do the trick.
Of the general concept I have no doubt. How the concept is to be
worked out is a little bit trickier. smoliar@vaxa.isi.edu (Stephen Smoliar) (03/19/90)
In article <14648@phoenix.Princeton.EDU> eliot@phoenix.Princeton.EDU (Eliot Handelman) writes: >In article <2cSD02rC93BD01@amdahl.uts.amdahl.com> kp@amdahl.uts.amdahl.com >(Ken Presting) writes: >;I am primarily interested in philosophy of science and >;metaphysics, so I prefer to highlight the problems of descriptive >;abstractions. This makes my analysis more cumbersome in application >;to the experience of music - the selection of a descriptive abstraction >;becomes a central theoretical issue, rather than a practical matter to >;be decided according to the interests of a composer, reviewer, or >;listener. > >Well, maybe the point here is that you can think of music as a descriptive >abstraction of more general experience. Not one that does anybody any good >except for trained musicians. > I think this is one of the points Minsky was trying to develop in his "Music, Mind, and Meaning" article in the Fall 1981 issue of COMPUTER MUSIC JOURNAL. He has a paragraph which begins with the rather striking sentence, "Compare a sonata to a teacher." To a great extent this paragraph elaborates on how the PROCESS of that sonata is similar to the process a teacher engages in communicating with the pupils. He does not call the sonata a "descriptive abstraction" of that process, but that seems to be the direction in which he is going. Is this of any use outside the community of trained musicians. For Minsky, it provided some of the first steps in laying down his "society of mind" theory. In other words, ANY descriptive abstraction is only as good as an agent's ability to manipulate it. IF a sonata abstracts the process of teaching, what sorts of agents will be able to exploit that abstraction? In trying to answer that question, Minsky began to develop the agents which populate his society of mind. Thus, thinking about music as a descriptive abstraction of more general experience may point us in the direction of machines which may be able to reason about that more general experience. ========================================================================= USPS: Stephen Smoliar USC Information Sciences Institute 4676 Admiralty Way Suite 1001 Marina del Rey, California 90292-6695 Internet: smoliar@vaxa.isi.edu "Only a schoolteacher innocent of how literature is made could have written such a line."--Gore Vidal
kp@uts.amdahl.com (Ken Presting) (03/20/90)
In article <12434@venera.isi.edu> smoliar@vaxa.isi.edu (Stephen Smoliar) writes: >>>In article <b1UF028Q90Pd01@amdahl.uts.amdahl.com> kp@amdahl.uts.amdahl.com >>>(Ken Presting) writes: >> >>. . . Certainly the physicists have produced >>a very interesting analysis of time, which seems to me to be adequate for >>the description of natural processes. If you mean that physics is more >>dependent on phenomenology than most physicists often admit, then I would >>probably agree. However, the inscribed graduations on most measuring >>devices suggest to me that natural science is more dependent on *texts* >>than on perceptions. >> >I would be prepared to argue that some of those texts may run the risk of being >abstractions which were imposed in an effort to dodge critical questions of >phenomenology. This may be a worthwhile topic on which to focus. I would agree that the texts which science assigns to events throught the use of measuring devices are part of an imposed abstraction. However, I would deny that this is a "dodge", in any sense. On the contrary, I think this is an inescapable and salutary methodology. Natural sciences, on my view, cannot possibly make any direct use of phenomenological descriptions. Science is an essentially public and dialectical activity, whereas transcendental phenomenology is the opposite. > Where physical analysis has made its progress seems to be in >the question of the perception of the passage of time. . . . >. . . clocks which measure units of time, . . . (Nitpicking: Clocks measure Time no more than rulers measure Space. Both of these devices measure *intervals*, when they are used to make measurements at all) >. . . there is more story to tell. Where physics has not yet provided >adequate analysis, to the best of my knowledge, is in the relationship that >time plays in the GENERAL perception of phenomena. This is true in some sense - the analysis of time provided by relativity explicitly excludes many of its interesting phenomena from observability within a single reference frame (which is as close to subjectivity as physics is likely to get). This is not a problem. From natural science, we should not expect anything more than a description of the hardware in which our minds are implemented. This is actually quite a lot to expect, and it is remarkable that so much has been described. > This is where the issues >of experience which were nicely summarized by Ken being to rear their >collective heads. However, while Ken wanted to relate experience simply >to listening, I suspect he would agree that it is a factor in ANY form of >sensory perception (including, for better or worse, our perceptions of any >of the measuring devices provided us by physics). I do agree - can we drop time-dependent phenomena and concentrate on perception of static phenomena? Compared to the difficulties of explaining perceptual abstraction in "simple" object cases, the additional problems of individuating processes and events are mere increments. Individuating objects themselves already involves re- identification, motion of the objects, figure/ground discrimination, vanishing intersections, vigilance dependance ... ... ... Btw - meters are not just perceived, they are *read*. That is a much more difficult issue. Once the reading is accomplished, and the observation becomes a statement in a formal symbol system, the logic of data interpretation, hypothesis testing, and deduction applies in familiar ways. Reading includes all the problems of perception of course, but also the abstract background of logical (and other) rules, not to mention "understanding" and other impolite terms. > > . . . We began by debating the question of how we perceive the >behavior of an array of cellular automata, such as the R-pentomino problem. Well, actually, you asked for an abstraction for describing that behavior. We may have a substantive difference over the relation between description and perception. I think any such relation is a fantasy. Literally. I would say that there is a formal similarity between description and perception, but the psychological interaction between a description of an object and subsequent attempts to identify such an object are of the utmost complexity. Capital UT. Especially if the description is *read*. Listening involves, I would say, a transference onto the speaker. But reading is a transference onto a fantasy object. I'm very reticent to bring up Freud, who is not only controversial, but is an object of general ridicule, even beyond Descartes. It is no accident that Freud seminars are generally filled with literature students if not sponsored by literature departments. How many psychology depts. even have a course on Freud in their catalog? Reading is a lot easier to do than to think about. Freud's psychology deliberately attempts to understand the most complex phenomena of human behavior, and has little if anything to say about the simple elements of behavior. This is a grandiose strategy, virtually assured of abject failure. If there were any other system that had made the attempt, I'd gladly consider it, but till one comes along, Freudian concepts will have to do. (Wouldn't you rather discuss measurement? :-) > . . . a terminology of chords, but that does not imply >that they will tell you anything about what those chords actually SOUND >LIKE. I think the word "like" is important here because of the experience >connection: we hear in relation to what we have heard. > >On the other hand, if we put aside all the notational games which keep the >academic types busy, we discover that there are other attempts at description >which tend to dwell more on what we can express in natural language. I agree that natural languages provide a more direct association to experience and feeling than artificial notations. This is not to say, however, that the association between expressions in natural language and experiences is itself direct (except for the association between expressions and experiences of language use). >. . . There has been recent recognition that phenomenology plays a role >in what is being communicated by such means. . . . >. . . All I am arguing is that such an approach to music leads us in >a direction of getting a better handle on the description of dynamic processes. Steven, unless you intend to include "hyletic" phenomenology, I would disagree very strongly. Description of any type is necessarily non- dynamic - a text is always presented as extended in space, but independent of time. If musical time is not clock time, how much less so is time in a novel? Only in the natural sciences do we find examples of a text associated to an object or dynamic process, without the intervention of a conscious interpretation. Only measuring devices produce "dynamic texts". Perception, in my view, does not produce texts at all, in any sense. It may be helpful to compare my view on measurement to Kant' view of perception, especially the "transcendental schematism". Kant supposes that "schemata" bridge the domains of sensation and conception, the way an antibody recognizes an intruding organism and presents it to the immune system (the analogy is quite close). My view of measurement is simply Kant's view of perception, externalized. (cf _Critique of Pure Reason_) >To sum up: I think that trying to approach music strictly in terms of its >notation is a bit like Ken's suggestion that we understand cellular automata >in terms of the state definitions of the automata themselves. Yes, that tells >us about what is going on; but it does not tell us how we PERCEIVE what is >going on. Such perceptions are the basis for abstractions we form which we >then engage in our reasoning. Thus, if we have a better handle on the >abstractions which facilitate our describing the experience of listening >to a performance of music, those abstractions may also apply to our perception >of the behavior of the R-pentomino. I would suggest that we defer all discussion of the relation of perception to description, until we have decided on the logical status of the concept of emergence. I believe that we will have little success in any attempt to understand language until we have replaced "emergent" with "normative". Does your concept of "emergence" essentially depend on perception? Or could you define it in some other terms? Ken Presting
smoliar@vaxa.isi.edu (Stephen Smoliar) (03/21/90)
In article <cc7d02re933H01@amdahl.uts.amdahl.com> kp@amdahl.uts.amdahl.com (Ken Presting) writes: > >I would suggest that we defer all discussion of the relation of perception >to description, until we have decided on the logical status of the >concept of emergence. I believe that we will have little success in any >attempt to understand language until we have replaced "emergent" with >"normative". > >Does your concept of "emergence" essentially depend on perception? Or >could you define it in some other terms? > I'm not sure that I (or anyone else) can effectively DEFINE "emergence," which may mean that emergence, itself, is a normative property of dynamic systems. All attempts I have made to deal with emergence thus far have, indeed, depended on perception. By this I mean that an "emergent property" is some property which is observed during the course of the system's behavior without necessarily being directly traced to any of the system's components. This is what Minsky is trying to get at in THE SOCIETY OF MIND. You have this society of agents which, acting together, yield behavior which we would call "intelligent;" but that does not mean that you can isolate which agent is "responsible" for that "intelligence." A glider in LIFE is the same way. We can only talk about a glider as something we can observe; so I think it DOES make sense to say that, whatever emergence may be, it depends on perception. ========================================================================= USPS: Stephen Smoliar USC Information Sciences Institute 4676 Admiralty Way Suite 1001 Marina del Rey, California 90292-6695 Internet: smoliar@vaxa.isi.edu "Only a schoolteacher innocent of how literature is made could have written such a line."--Gore Vidal
kp@uts.amdahl.com (Ken Presting) (03/22/90)
In article <14648@phoenix.Princeton.EDU> eliot@phoenix.Princeton.EDU (Eliot Handelman) writes: >In article <2cSD02rC93BD01@amdahl.uts.amdahl.com> kp@amdahl.uts.amdahl.com (Ken Presting) writes: >;So the homomorphism of logical structure exists between the descriptions, >;rather than between the phenomena. The analogous relation which obtains >;between the phenomena is *resemblance*. The concepts of logical >;structure and homomorphism serve to analyze resemblance. > >Ken, I think you've just negated the whole value of introducing musical >approaches to the study of emergent processes, That's half of what I'm trying to do ... > . . . I still can't see in >what way a description of a performance has to do with a phenomenological >account of listening experience . . . This is the other half of what I'm trying to do! My position wrt descriptions and perceptions is that we can't understand how hearing a description or reading a review will tell us about the experience of listening to music, until we have solved the problem of understanding literature and poetry. This means we should not attempt to understand perception in terms of its relation to descriptions, except by means of the formal similarity between the semantics of a description and the operation of a transducer. I want to reduce emergence to what I think is its appropriate status - a logical concept which is suggestive but misleading. > . . . "Logical" means to me >(roughly following Langer's usage) "abstracted to its bare but vital >properties." You are using the term in the "what you see is what you get" >approach. I'm using logic in the "How to succeed in arguments without really trying" approach :-). Logic is a total recursive algorithm for accepting the set of valid formal arguments, and absolutely NOT the transcendental categorization of the understanding. Sort of "Kant times Wittgenstein divided by Aristotle", if that makes any sense. I think logic has the epistemic role that Kant gives it, but not the psychological role. On the other hand, I'd say that logic is part of the human way of life, both as a normative abstraction for dialogs, and as a descriptive abstraction for abstractions (ie formal systems). On the psychology of logic, note that you are at this moment *reading* an argument, ie, a literary production offerred in the hope of eliciting logical criticism. Persuation is a sin against the Categorical Imperative, as is loose reading. (I generally avoid the former, and I appreciate Eliot's avoidance of the latter) The most important aspect of logic for present purposes is the immense distance between the *practice* of logic and the *process* of perception. Ken Presting
smoliar@vaxa.isi.edu (Stephen Smoliar) (03/23/90)
In article <a9Mo024=94yf01@amdahl.uts.amdahl.com> kp@amdahl.uts.amdahl.com (Ken Presting) writes: > >The most important aspect of logic for present purposes is the immense >distance between the *practice* of logic and the *process* of perception. > So we DO have some agreement! The whole reason I wanted to get into this discussion in the first place was because of my interest in perception AS A PROCESS. Trying to focus attention on the perception of dynamic processes may have been a mistake, but I'm not willing to concede on that point yet. Nevertheless, it is the process side of the picture which interests me; and it seems that you have now admitted that the practice of logic is not appropriate for this concern. In a similar vein, I would argue that solving "the problem of understanding literature and poetry" is also not appropriate, since I am prepared to argue that what we "understand" about any artistic experience depends on what we PERCEIVE. Now, do you want to talk about the process of perception; or was it your intention to push it off into the background? ========================================================================= USPS: Stephen Smoliar USC Information Sciences Institute 4676 Admiralty Way Suite 1001 Marina del Rey, California 90292-6695 Internet: smoliar@vaxa.isi.edu "Only a schoolteacher innocent of how literature is made could have written such a line."--Gore Vidal
kp@uts.amdahl.com (Ken Presting) (03/23/90)
In article <12502@venera.UUCP> smoliar@vaxa.isi.edu.UUCP (Stephen Smoliar) writes: >In article <cc7d02re933H01@amdahl.uts.amdahl.com> kp@amdahl.uts.amdahl.com (Ken >Presting) writes: >> >>I would suggest that we defer all discussion of the relation of perception >>to description, until we have decided on the logical status of the >>concept of emergence. I believe that we will have little success in any >>attempt to understand language until we have replaced "emergent" with >>"normative". >> >>Does your concept of "emergence" essentially depend on perception? Or >>could you define it in some other terms? >> >I'm not sure that I (or anyone else) can effectively DEFINE "emergence," which If this is true, it will be difficult to use the concept of emergence in an argument, or as a design objective. Nagel's _The Structure of Science_ includes an extended discussion of emergence, as does the _Encyclopedia of Philosophy_, article "Emergent Evolutionism". The usual definition is relational, ie: 1) P is emergent wrt Q,R,S... if Px & Qx & Rx & Sx ... holds, and it is not the case that Qx & Rx & Sx & ... entails Px. This definition reduces to a straightforward notion of "logical independence" unless some additional claim is made about Q,R,S... such as: 2) x=y if and only if Qx<=>Qy and Rx<=>Ry and Sx<=>Sy ... ie, Q,R,S,... form a "complete individual description". Note that a property can be emergent under this definition only in a formal system which is *incomplete*. For example, if arithmetic is given the usual Goedel interpretation into its own sentences, and P is extensionally equivalent to truth, and Q,R,S... are recursive syntactic predicates, then (1) and (2) can both hold. Therefore truth is an ... whoops ... Truth is a *normative* property - it only looks emergent. This definition of emergence is effective for MY purposes. :-) >. . . >may mean that emergence, itself, is a normative property of dynamic systems. > Bite your tongue! er, well, then bite your fingers. I don't want to let one of my favorite concepts get tarred with a vague brush. Truth, Beauty and Goodness are normative concepts - do you want to put (yechh) emergence (ptooie) in that class? Some normative properties are in fact not usefully definable, but so are a large number of descriptive concepts, notoriously "game" and "chair". The practical vocabulary of everyday speech needs little in the way of definitions to advance its purposes, and suffers from most attempts to supply precise analyses. When normative concepts are part of everyday speech, they generally resist definition. As long as a predicate is restricted to use in communication, this is only a small problem. But when arguments or designs are stated with undefined terms, confusion or failure is likely to result. "Truth" is a significant exception to the undefinability of normative properties which have everyday utility. Tarski's theory of truth is widely applicable, although some work does continue among logicians who feel that Tarski's approach may not sove every problem related to truth. In Tarski's theory ("The Semantic Conception of Truth", in a collection by Leonard Linsky whose title escapes me), truth is defined in a metalanguage. A metalanguage is a descriptive abstraction whose terms refer to sentences and formulas of another descriptive abstraction, called the "object langauge". For example, in the metalanguage one might make assertions about the sentences in the object language, such as "It is true that 'Snow is white'". In this example, the object language includes the terms "Snow" and "white", whereas the term "true" is NOT a part of the object language. (I am omitting many fascinating details) Note that the truth of "Snow is white" depends on the *interpretation* given to the words. If "Snow" is taken to denote slush as well as new, drifted flakes, then "Snow is white" will be false. Normative properties always have this characteristic of dependence on an interpretation. Thus, normativity is a strong candidate for inclusion in a theory of perception. Discriminating and identifying and object within a scene seems to require interpreting a region of high contrast as a boundary, for example. Perception also functions as a sort of bridge between structured systems - the outside environment and the representational concepts of the mind. Concepts such as truth are descriptive in the context of a metalanguage, but are normative when applied "across" formal systems, ie, from the metalanguage to the object language. > . . . A glider in LIFE is >the same way. We can only talk about a glider as something we can observe; I don't think this is true. Stability of cellular structures, as well as stable translational motion of structures, is not difficult to formally define. The definition of emergence I gave above does not explicitly mention relational properties or dynamic properties, but that is easily remedied, and the conclusion stands. The closest I have been able to come to a useful conception of emergence is in relation to chaotic systems. In a chaotic system, it may be impossible to know the "complete description" of the initial state. But this does not show that subsequent states are emergent, it only shows that they may be surprising. To adopt emergence in this form, and claim that consciousness or intelligence is emergent is to admit that AI cannot succeed on the basis of a deliberate design. The concept of emergence in a weakend form such as this (where the appearance of emergence is attributed to limitations of the starting description) does have a genuine use, I think. Similarly, when a system evolves probabilistically, the appearance of emergence may be attributed to "imperfect knowledge" of the rules for inferring future states from past states. I wouldn't go so far as to say that emergence never occurs, ie, that unexpected things happen. I just want to avoid any claim that certain properties can be exhibited ONLY by means of emergence. >so I think it DOES make sense to say that, whatever emergence may be, it >depends on perception. Stephen, why don't we invert this and substitute, viz, "whatever perception may be, it depends on a process which is formally similar to interpretation"? Ken Presting