Put your new question or comments in the comment to this entry.
Then it will go to Len as an RSS feed and we can put it up as an entry and start the stream.
If this is all new terminology, don’t let it scare you away.
Les Vogel wrote about the Yale course on fractals:
This site is a really good resource, but very technical. I based my paper on it, but in a sense was disappointed, because I wasn’t able to tease out many of our systems properties.
My response:
This gives a good opportunity to compare treatment styles. The Yale course is the up-to-date conventional treatment mostly from the standpoint of its mathematical characteristics and some applications. But we are asking a different question. As a property of systems-in-general how does our knowledge of fractality contribute to our understanding of how systems-in-general work. So the systems properties we seek may be different “takes” on the phenomena. But they are common to all such systems-level processes. That is what we seek. So any treatment within a specialty like fractals will certainly emphasize different stuff. It is our job to find the transdisciplinary sense within that specialty treatment which is sometimes quite hard as Les points out here. Still we can re-evaluate and change our systems properties at any time given consensus on a change.
That is why I started the last session with the statement that in most “manfestations” of systems the fractality is a function of large number iterations of the combination of a relatively determined process (which guides the aggregation, erosion, development etc.) and a relatively random process. It is the outcome of the iteration with variety (due to the surrounding environmental complexity) that creates the randomness which was misunderstood in my opinion by most of the group. As for the several comments that random is not truly random (which of course is true) it doesn’t matter much because it is close enough to random to add the element of chaos to the underlying developmental process.
Lynn Rasmussen:
I have a question. When a system operates for a short time in a manner seen normally at more complex higher levels and it takes more
energy and more pathways–what can I call that? Does it have a name, a process name? Is it a process?
Help! I have to title my little category.
SUGGESTIONS:
The terms or phrases I have used for this are variously
“hardwiring of what was formerly ad hoc”
“constrained states from originally unconstrained states” and simply
“subsumption” (one of the ID Features/Functions from the hierarchy/herterarchy/heteropoiesis complex)
I am not sure that I am focusing on the same phenomenon as you describe because I would describe it in somewhat different terms. Others have recognized it before although I could not tell you where I have encountered it specifically in the literature. Just a memory. I would describe it this way …. sort of ontologically the reverse of what you describe …..
Lower levels of a hierarchy that were earlier independent become constrained in their behaviors and require less maintenance energy once they are incorporated in or become part of new, emergent levels (by either the subspecialization or emergent processes). Described in this way, it is a phenomenon that has been repeated over and over again across the entire 40 or so “origin” levels I have described in the “Unbroken Sequence of Origins” streaming video in previous quarters. Other examples may include, but not be limited to, the evolution of the physical brain. According to Minsky (see Society of Mind) and others some of the parts of the brain involved independently to efficiently do various tasks and only later were incorporated into (connected with) the larger brain. As in the heartbeat, breathing and various physical regulatory procedures that once were all an organism could do until they were gradually incorporated (hardwired) into the cerebellum. Learning also may involve something similar. When we encounter new information (neural nets; memes) that we want to incorporate into our understanding and memory, we must give the new ideas much attention and energy (impulse activity; blood flow) until they become more comfortably embedded in existing neural nets. Once there, much less conscious attention (energy) is required. Consciousness itself may be an “exaggerated function” example of this. We spend great energy at the outset on experiences that gradually become incorporated into our subconscious requiring less energy. This has many important things to say about meditation and personal integration.
In most of these cases, the incorporation into larger wholes is thought to be accompanied by savings in energy and maintenance but actually I am not aware of any really solid data that proves this to be true. Until then it is merely a conjecture and a dangerous one to assume at that. Nevertheless, others assume it to be true. So subsumption is a consequence and function of hierarchies, but as such also of clustered networks. I hope this helps.
Next session I want to describe some of the consequences that our growing recognition of networks has on the fundamental human and scientific concept of causality. The scientific method and reductionist approaches to knowledge rest firmly on the assumption (explicit or implicit) of linear causality. But the growing list of case studies in science where networks are the dominant phenomenon calls this assumption into question or at least modifies it immensely. I would like to explain to you some of the early recognitions of this in conventional natural sciences. I will use as my example one from my home field of human genetics. I will describe to you the concepts of “penetrance” and “variable expressivity”, especially in terms of human diseases such as cleft palate and the resulting effects on our notions of causality. I might then extend this to the necessary adjustments we have to make when designing experiments and interpreting results in the conventional natural sciences much less in the growing systems integrated sciences like systems biology, earth systems science, systems neurosciences, systems immunology etc.
Of course systems types have spent a half a century telling reductionists that they were grossly oversimplifying the phenomena they studied and ignoring the “environment” affecting the phenomenon, without much impact. Reductionism was too successful both for understanding theoretically and applying practically. But now their own reductionist research is demonstrating this conundrum. And systems network processes are the prime example.
My casual reading of Nature and Science lately keeps coming up with examples that indicate that our emphasis on SIS (Systems Integrated Sciences) and SSP (System of Systems Processes) is both timely and significant. I have been perusing these interdisciplinary journals for decades but never have I seen articles like the following which are now appearing with greater and greater frequency. I cited Bar Yam’s article on using chaos theory and boundary conditions and clustering to better predict social conflicts in my last message. Another recent Nature has the following:
Scanlon et. al. (2007) “Positive feedbacks promote power-law clustering of Kalahari vegetation” Nature 449:209 (Sept 13).
Now most of us do not read detailed specifics about desertification of South African ecology to be sure. But that title implies they use at least three different systems processes we study (feedback, obviously; power-laws; and hierarchies, clustering) and show influences between them. If this is not a Linkage Proposition in “exaggerated function”, I don’t know what is. It even uses a cellular automata model like SFI studies. So one could study such an article for revelations about each of the systems processes taken alone and together. Then, when sufficiently abstracted from the desertification phenomenon and even ecology, info bits identified become “conjectures” to look for in other complex systems. This work is out of Princeton’s conventional science and engineering departments. Incidentally, the next two articles are also on clustering, allometry, and scaling laws involving “flows” and patterns in ecology.
So could positive feedbacks in motif configurations “promote” power-law clustering in human organizations? Great hypotheses, conjectures, and insights come from comparing systems. Thus, Comparative Systems Analysis (CSA).
Note that Nature and Science are only interdisciplinary in that they include articles in most all of the conventional natural and some social sciences. They are not really INTER nor TRANS because the articles do not compare across disciplines, yet. For example, this article illustrates one type of influence across systems processes, but only in one sub-discipline of biology.