[Date Prev][Date Next][Thread Prev][Thread Next]   [Date Index] [Thread Index] [Author Index

On the nature of abstraction in science...



(... and/or the abstraction of nature in science!)
 
While we are discussing mathematical logic and the nature of systems which can be completely formalized (hence, are "computable"), I thought we might as well throw a little more bio-diesel on the fire...
 
From page 213 of "Anticipatory Systems", Robert Rosen wrote:
 
"Let us suppose we are given only a single meter, which may interact with the states of some natural system N. As far as the meter is concerned, the only quality of N visible to it is the one embodied in the observable it measures. As far as that meter is concerned, no other qualities of N exist; and indeed N behaves for that meter as if it possessed no other qualities. Instead of seeing the full range of interactive capabilities which N can manifest, then, the meter sees only one.
 
It is in this sense that we shall say that any observation procedure, applied to a natural system N, generates an abstraction. For instead of telling us about N, the observation procedure can by definition tell us only about a single quality of N; it must necessarily forget, or neglect, all the other qualities which N may manifest. It allows us to see only a projection of N along a single interactive dimension. This kind of forgetting, or neglect, of existent qualities is the essence of abstraction.
 
In popular parlance, "abstraction" has a pejorative connotation. The antonym of "abstract" is "concrete"; thus in the popular view (which is shared by many empirical scientists) abstraction is a property of theory, and one engaged in the direct observation of nature cannot be accused of performing abstractions. Above all, one who observes nature directly must always be solidly anchored in concrete reality...
 
We see, however, that the facts are quite otherwise. "
 
Robert Rosen often said "there is nothing more abstract than a measurement." (Raw data, without the referents, is meaningless.) He viewed theory as the way to actually make empirical, i.e.; direct observation/experimental science "less abstract". (The theoretical underpinnings are what suggest where to look, what to observe, what sorts of experiments to design in order to get useful data that pertains to the questions we have. In other words; information.) He believed that any science which proceeds from, or is based on, an unsound theoretical basis will be of questionable value: It cannot fail to be incomplete and/or inapplicable, at least part of the time, even in the best-case scenario. My father asserted that this is precisely the situation in science, currently. Science today is based on the notion that all systems in the universe (indeed the universe, itself) are "just like a machine." This assumption is further extrapolated into a presumption that the ways we can study machines and the things we learn from such a study are going to be both relevant and applicable to all of material reality. So, for example, when there are foundational questions to be answered about the epistemology of any given system, the way science studies the system is by taking it apart and studying all the subsystems, then taking the subsystems apart and studying all the components, and ultimately taking the components apart and studying all the raw ingredients.
 
This turns out to be useful when working with certain kinds of systems, but less and less useful with other kinds. Among the kinds of systems it works very well with are machines, which is not a surprise, of course. However, no amount of studying of the ingredients and components of a living system will ever be able to answer the question; "Why are living organisms alive?" My father spent his life's work finding out the answer to that question as well as uncovering the reasons why science currently has so much trouble learning fundamental things about biological systems. As it turns out, it's NOT that biological systems are "too strange and bizarre" compared to everything else in the universe-- on the contrary; Einstein was right that there is a common thread running through the fabric of all systems, all phenomena. He was also right that it has to do with the power that can be contributed to a system by a relation.
 
All systems have relations as ingredients, in a sense; as parts. But not all relations will have the same impact, even when the relation involves interaction between the same two components. The sun and the Earth have a set of distinct relations which are responsible for the nature of the temperature conditions we experience on our planet. If our planet were in a different set of relations-- say, similar to Mercury's-- the effects would be entirely different. The common thread that runs through all systems in the universe is based on this fact; that all causality is relational. That is; it is the interactions between "things"-- as specified by the particular relations involved-- which drives causality.
 
That conclusion was reached after a great deal of investigation into the history and the foundations of science, searching for the reasons why certain aspects of biological systems were proving to be unapproachable using the scientific tools which existed "on the shelf," as it were. His diagnosis was that several incomplete and/or frankly erroneous assumptions and conclusions were inadvertently built into the foundations of science and were never really scrutinized or re-analyzed adequately, since. The most damaging of them: the Cartesian "machine metaphor" as a conceptual model in science for all natural systems. About this particular foundational idea, my father wrote: "The machine metaphor isn't just a little bit wrong, it is entirely wrong, and must be discarded".
 
The changes to science which are required (in order for science to learn about, understand, and solve problems/make accurate predictions with regards to complex systems) are not just "new capability" that can be added onto what already exists. There must also be an excision of certain conceptualizations of natural systems which are currently mortared into the foundations themselves. I can attest to the fact that, in all of my research on the internet into various different theoretical descriptions of "complexity" (all of which differ markedly from the Rosennean one),  I have not seen any serious scrutiny of the foundational concepts that underlie all of science (and, therefore, become part of the definition for what most people view as being "scientific").
 
Furthermore, if there are not some foundational changes made, then any new "scientific" capability that we generate is likely to proceed from, and/or incorporate, at least a few vestigial aspects of the same flawed conceptualizations. The primary problems that science was concerned with in previous centuries, up to about the middle of the 20th Century, were nearly always much simpler than the issues we are facing today. Examples include: Aspects of our solar system's mechanics, industrial applications, mechanical engineering, architecture, civil engineering, chemistry... those were/are areas where the relational nature of causality can (with relative safety) be either ignored or treated entirely as a series of variables in notation. However, times are changing. As humanity is increasingly being called upon to manage resources and solve problems in the planetary biosphere-- and with human physiology, human social systems, and the human conscious mind, etc.-- we will be dealing almost entirely with complex systems.
 
Judith