Theory vs Experimental Science

[Judith Rosen <***>]

My oldest daughter, who is in her third year of university (Environmental Sciences) at Guelph, spent much of her Thanksgiving break creating a poster about Robert Rosen and his work for an upcoming science fair. Naturally, I was helping her with this project, answering her questions and giving her what I think are the most important aspects of her Grandfather's work to highlight. I've posted stuff about all of these to the discussion list, over the past couple years: What Rosennean Complexity is, what it means in terms of science, what the current machine-metaphor-based scientific paradigm is and why it misses the most salient aspects pertaining to complex system epistemology, what must be done within science to allow more intelligent problem solving in human interaction with complex systems, the entailments of relational causality, and the difference between entailment and causality, etc. In the process of working on this poster, I had another of these little insights that have absolutely no value outside of me unless I pass them on. So I'm passing it on...

One of the big dualities my father often wrote about was the duality between "hard" (quantitative) and "soft" (qualitative) science. He likened this duality to a much larger duality which exists between Experimental science and Theoretical science. Experimental science regards itself as "hard" science and regards Theoretical science as "soft" science. My insight has to do, in part, with recognizing that Experimental science is often only dealing with and/or looking to explain a particular aspect of causality, whereas Theoretical science (as Robert Rosen employed it) is involved with entailment, i.e., using causality to learn about the underlying entailment patterns and/or exploring what we know about the entailment (via causality) of a system in order to figure out what else must be  entailed within that system, etc. I think this duality can be traced to the fact that the machine metaphor, as a model for all natural systems, is tainting the view: In a machine, one can limit one's self to causality and one will generally get most of the entailments along with that. Furthermore, those answers can be arrived at via reductionist science. When dealing with a complex system, though, this is not the case at all-- there is far more entailment within the system than what can be discovered via reductionist science. The differences between the two types of system are generated via relational means and therefore pertain to matters of organization.

So, in a machine, the relational aspect is muted and entailment seems to be a synonym of causality. This "reality" then is applied to all systems in the universe, including organisms, ecosystems, atoms, social systems, and the nature of the consciously aware human mind. However, it is more than obvious in biological systems that any single incidence of causality or any single species example of causality will not explain the general behavior of all living organisms-- their causality is just too different. For example: if we accept that the functional capabilities of metabolism and repair are the hallmarks of all living systems, regardless of their material realization of these capabilities, then we are interested in understanding the entailments of metabolism, not just "how mammals do it" or "how plants do it" or whatever.  The same is true of the capability of repair. The same is true of the capability of reproduction... Clearly; just because an octopus reproduces in a certain way doesn't mean that's the only way reproduction in organisms can happen. Just because we cannot naturally re-grow a limb if one is torn off doesn't mean all organisms would be similarly stuck with a stump in that event. With biological systems as our observation field, there are limitless examples of these differences.

I then had the further insight that many Theoretical scientists from all fields of science aren't using natural systems as their anchor; instead they deal entirely within the realm of current theory in order to explore the unknown entailments of those which are already part of the model (of some natural system). Astrophysics and much of physics, it seems to me, are in this group. So, they are dealing with entailment all right, but it's entirely the inferential entailment structure within the model (whatever that model may be-- Quantum Physics, Newtonian Mechanics, Relativity, etc.) They are exploring what else will be entailed by those entailment structures, generating predictions which are then applied to the natural world. When the systems the predictions pertain to are relatively simple, those predictions may seem accurate (or at least, accurate enough) and it reinforces the notion that the inferential entailment structure of the model of reality actually commutes entirely with reality itself. And yet, predictions from one of these models don't always commute with predictions from another model, OR the predictions from one model aren't generated via the other model, even when both models purport to portray the entailments of the same natural system. Isn't this precisely what bothered Einstein about his own two theoretical models? He knew it meant something was wrong somewhere, and yet certain predictions made by each set had been "proven" to commute with the natural system.

As I've said before; In the past, I would have talked about these ideas with my Dad but, of course, he's not here anymore. So, for whatever they're worth, I'll pass them along and hope they might prove useful to someone, somewhere, somehow. At the very least, perhaps they can serve as a little food for thought?

Cheers,
Judith

Web address: http://www.rosen-enterprises.com
BioTheory: An electronic journal of general science based on the Relational (Rosennean) Complexity Paradigm