Glen, You write: > > [EMAIL PROTECTED] on 01/02/2008 08:51 AM: > > Well, feedback loops begin and end too, and that displays an even > > greater 'inefficiency' for causation... Just plane old bloody gaps. > > The rub is that systems of loops originate for no efficient cause. > > That's why I turned the sci method around to warch them since it's > > clear we can't explain them. > > I disagree. First, to say that feedback loops begin and end > is an _assumption_ of a discrete ontology. I.e. feedback > loops may not have a beginning or an end, they may merely be bounded.
Yes, sure, that's an option of interpretation, but does it fit with the rest of what I was saying? I think there's an interpretation that fits the data of nature better than any other, so it's reached as a 'conclusion' not as an 'assumption'. For example, can you offer any example of physical growth (accumulative change) without a beginning and end? > Second, most of what people seem to point at when they use > the phrase "feedback loop" is an aggregation of phenomena > caused by an aggregate set of mechanisms. Hence, even if the > ontology is discrete or discretizable, we may not be able to > discuss them in the same language (attributes, properties, > predicates, operators, etc.) we use to discuss the phenomena > and mechanisms of which they're composed. And further, not > only may we need a different language, they may not even give > rise to the same categorization of actual behaviors. I.e. > the components can be very different from the composition. > To conflate the two is to commit the fallacy of composition/division. Complex systems are always poorly represented by our models, but does that restrict them, or just us? :-) It's completely normal to discover that in describing one physical thing you often need a combination of different languages of description. For example, you might describe something's chemistry, it's appearance and its various roles in its environment. They're all useful, especially together, though each is highly incomplete and they hardly connect at all in terms of the formalities of each mode of description. > > And third, we might posit that "feedback loop" is _merely_ an > ascription having nothing to do with the ontology and > _everything_ to do with our psychology. I.e. "feedback > loops" may not actually exist except as a convenient lexical > structure we use to describe the world. Well, certainly a term needs to be understood so that when one persons uses it another person can know what is being referred to. But isn't that a normal problem with language, not an inherent flaw in language? In this case I'm using 'feedback loop' in a way I thought would be understood, from your referring to the physical model of the 'chicken & egg' cycle. It wasn't that clear perhaps. I meant it to refer to the type of feedbacks we commonly find in nature, not a theoretical construct. Like the chicken & egg cycle, all cycles in natural systems seem to develop and decay by transient accumulative change processes. The name 'feedback' gets attached since they generally fit the model of exponential-like accumulative change. Can you think of any regular cycle that does not begin and end with accumulative processes on scales that make them untraceable? > > In the first case, we can't make the logical leap to say that > feedback loops have no efficient cause. In the second case, > the cause of the loops is _complex_... and we've had that > discussion recently. And in the third case, feedback loops > do have an efficient cause... _us_. [grin] I draw the conclusion that natural system feedbacks have no efficient cause since it's 'inefficient' to have causes separated from effects. With growth systems there are usually time lags between cause and effect, so any 'cause' is instrumentally disconnected from the process that follows it. Growth systems also usually have complex emergent properties with a complexity not evident in the original environment, and so outside cause fails to be 'efficient' for requisite variety too. In the case of a real physical growth system you'd be quite right to say that any feedback loop we can define has us as its efficient cause. A physical system's own feedback loops are indeed complex. For talking about them it seems you need words that take their meaning from what they refer to rather than be defined so they can't. That's an issue, of course. Then I think the best of all evidence is the myriad physical systems that hide their designs inside themselves. That's very 'inefficient' isn't it, to have things designed and operating according to principles that are universally invisible from outside? Isn't that typical for physical systems though? > > I'm not saying that any of these are true; but they are > certainly defensible positions... as defensible as the > assertion that the loops have no efficient cause. When you talk about 'defensible' but ambiguous positions I'm reminded of questions like whether trees falling in the woods make a sound if no one hears them. The interest in that question seems to rest entirely on the argument for either position being completely 'incontrovertible', i.e. defensible by being impossible to contradict. To me people seem interested in that because it turns on whether the universe is composed of information or things. If just information, then the unobserved falling tree makes no sound. If you approach the world as composed of things, then it does. Why anyone would even wonder about that might be that our mental pointers to physical things get mazes of self-references attached to them, so our thoughts can wander without end looking for what's real, and find nothing but themselves to connect to. When you strip the interpretations from the pointers, they can work again. I find it gives reality great substance, and having pointers reliably lead to where there are new things to discover very useful. Do you think Rosen is thinking at all about this issue? It sounds like he's looking at an equally central problem of explanation I think. Cheers, Phil > > - -- > glen e. p. ropella, 971-219-3846, http://tempusdictum.com > When buying and selling are controlled by legislation, the > first things to be bought and sold are legislators. -- P.J. O'Rourke > > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v1.4.6 (GNU/Linux) > Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org > > iD8DBQFHe+JXZeB+vOTnLkoRAhrvAKCNlM31w5lG4mLCJdQh6+8KxMAvggCdFb8r > v9bMfdkZPoDUCrJTxolKpWs= > =Ji8l > -----END PGP SIGNATURE----- > > ============================================================ > FRIAM Applied Complexity Group listserv > Meets Fridays 9a-11:30 at cafe at St. John's College > lectures, archives, unsubscribe, maps at http://www.friam.org > > ============================================================ FRIAM Applied Complexity Group listserv Meets Fridays 9a-11:30 at cafe at St. John's College lectures, archives, unsubscribe, maps at http://www.friam.org
