Systems: Science, and Sciences
One of the most frustrating aspects of systems science research is the lack of clarity around the field’s basic boundaries and terminology.
In a recent meeting of the ISSS’s research towards general theories special integration group (SIG), I was trying to better understand group members’ perspectives about how to achieve progress towards our stated purpose.
The SIG’s official page reads:
“General Systems Theories (GSTs) are needed to unify the systems sciences under common frameworks of description, analysis, and explanation, and open new routes to systems-scientific innovation.
This SIG provides a venue for developing and discussing ideas, strategies, frameworks, opportunities, and challenges relevant to research towards developing and applying GSTs.”
I asked two of the group members — “what do you think is needed to unify the systems sciences, and what barriers have blocked unification so far?”
Rather than any clear answers, we got three more fundamental questions that need to be addressed.
What is meant by general and what is the proper level of generality?
Should we be focused on constructing and comparing theories, or practicing science using existing theories?
What exactly is meant by the systems sciences?
I’m going to provide my perspective on question number three, because we can’t unify a set of things (systems sciences) if we don’t know what they are.
The Systems Sciences
In his magnum opus, General System Theory, Ludwig von Bertalanffy argued that humanity needed a new general science of “wholeness” in order to identify universal principles that applied to each of the individual empirical sciences — physics, chemistry, biology, along with the various social sciences. This would facilitate greater integration in scientific education and progress towards the unity of science.
Just as probability theory produced universal methods applicable for all sciences concerned with random events — from genetics research to life insurance statistics — General System Theory would produce methods universally applicable for all sciences concerned with organized wholes.1
Unfortunately, Bertalanffy wasn’t very precise with his terminology when describing the emerging components of his proposed new scientific paradigm.
In one section, he describes a variety of fields including information theory, cybernetics, and game theory as “approaches.”2 He intentionally uses this loose term because the fields “are logically inhomogeneous, represent different conceptual models, mathematical techniques, general points of view, etc….” Yet, he also argues they are similar in being systems “theories.”3
Later he describes the same set of fields as “disciplines” which share a common concern for systems wholes and organizations. He again emphasizes that they differ and sometimes contradict each other “in their basic assumptions, mathematical techniques, and aims.” 4
In a series of lectures delivered at Clark University he once again describes these disciplines as “systems sciences” unified by their concern with issues generally neglected by science — issues related to organization, the interaction of many variables, and purposeful behavior, among others.5
Were the fields that Bertalanffy held up as prime examples of general systems research approaches, theories, disciplines, or full-fledged systems sciences?
This is a question modern systems scientists must answer for themselves. I think it’s both accurate and useful for our group’s purposes to describe cybernetics, game theory, and information theory along with newer disciplines including complexity and network science as systems sciences.
A Unified Systems Science
So, what might it look like to unify the systems sciences?
Thankfully, we have a useful definition of systems science, in the singular, to guide us. Bertalanffy outlined three core aspects of General System Theory — science, technology, and philosophy. Each aspect deals with the same content, but with differing intentions.
For him, systems science was the scientific exploration and theory of “systems” in each of the various sciences (e.g. physics, biology, psychology, social sciences), and the search for general theoretical principles which apply to all types (or subtypes) of systems.6
While the individual systems sciences have been successful in their quest to identify such principles which seem to hold across all of the traditional scientific disciplines, they each operate largely within their own siloes. Systems science, with its various sub-disciplines, has fallen into the very same pattern of overspecialization and fragmentation that General System Theory was intended to overcome.
Unifying the systems sciences would involve identifying common frameworks of description, analysis, and explanation that allow for direct comparison and integration of results found in each sub-discipline, from cybernetics and information theory to network and complexity science.
With a clear goal in mind, we can turn our attention towards evaluating different strategies for accomplishing it. Next week I’ll share my thoughts, informed by recent modern advancements in systems science, on what some of those might look like.
von Bertalanffy, Ludwig. General System Theory: Foundations, Development, Applications (pp. 37-38). George Braziller Inc.
Other fields on the list include: classical system theory, computerization and simulation, compartment theory, set theory, graph theory, net theory, theory of automata. decision theory, queuing theory.
von Bertalanffy, Ludwig. General System Theory: Foundations, Development, Applications (p. 19). George Braziller Inc.
von Bertalanffy, Ludwig. General System Theory: Foundations, Development, Applications (pp. 188-189). George Braziller Inc.
Bertalanffy, L. von (with Internet Archive). (1967). Robots, men, and minds; psychology in the modern world. New York, G. Braziller. http://archive.org/details/robotsmenmindsp00bert
von Bertalanffy, Ludwig. General System Theory: Foundations, Development, Applications (p. xxv). George Braziller Inc.