The Critique of "Systems"

by Hugh M. Lewis

In the course of my studies I have encountered the critique of systems-based approaches on several points. This critique was especially well elaborated at SUNY-Binghamton, home and host to World Systems Theory and other applied Systems-based approaches in development. My critique focuses on two main points: 1. the eidectic, abstract structure of "Systems" approaches that involve a fallacy of reification or misplaced concretization; and 2. the "information explosion" of the applied elaboration of general systems when carried to local levels of specificity. 

I would add to these two points the conclusion that there is a tendency in the use of "Systems" approaches to over-simplification of problem sets and the realities represented; that reality tends toward a level of complication of problem sets that has been far beyond the capacity of most system theories and approaches to sufficiently represent in symbolic form; and that general laws and rules of relationship that apply at one level or area of a natural systems framework do not apply to other areas. 

In other words there is a form of general relativity of systems laws that are specific to certain levels and contexts of systems, but not to others. This derives from the fact that all "systems" by definition are delimited in time and place and hence unique in their details, and all kinds of systems are similarly different. It is largely for this kind of reason that I have found the adoption of the term "meta-systems" appropriate as a means for stepping outside the kind of dilemma this entails. This would seem to mitigate against the idea that we can formulate a general, comprehensive theory of systems based upon a single abstract model that would be suitable for all possible kinds of systems we encounter in reality, though it is possible and possibly useful to have a general frame of reference (a meta-system) by which different systems and kinds of systems can be systematically grouped, compared and related to one another in a manner that makes greater sense of the ordering of reality.

The main point of the fallacy of misplaced concretization its that it is part of the overextension of reference of "Systems" and the tendency to reify our theoretical construction of "systems" as if they actually exist somewhere in the real world other than in the form of the ideas immanent in our own brains. Merely putting quotation marks around the upper-case word "SYSTEMS" serves to highlight and thereby bring notice to the implicit reification of such ideas, when just off-hand talking about systems in general serves to hide and smuggle in this idea as something more real and tangible in the world than it really, truly is.

Of course, it should be kept in mind that any general concept can be reified by overextension of reference and underdevelopment of the details and lack of specification. I would only claim that a concept like "systems" tends to be more easy to reify and forget the abstract, constructive nature of the concept, ultimately arbitrary, than other ideas that have more natural frames of reference. In this sense, a concept like "systems" is very much like the concept of "culture" that Anthropologists have used and abused for well over a hundred years now, without necessarily coming to agreement on a single, universally shared definition of culture in as few words as possible. The fallacy of reification is nothing intrinsic to the concept or theory of general systems, but is something that the use and abuse of such a concept brings with it, as with any concept that refers to patterns that are without direct reference.

The second charge, of the operational break down of such approaches in application and complication of details, is a more legitimate and more insuperable problem to consider. The challenge has been and remains to devise a methodological system of application that basically solves the "information explosion" that is attendant to the attempt to deal with large, complex problems with unknown solution sets. A large part of this problem is common to all methodological approaches in the scientific fields, because usually solution sets to large or complex problems are generally unknown, and methodological applications are usually experimental in intention and effort. Part of the problem also stems commonly from not having a clear theoretical understanding of the problem involved, as usually a correct theoretical understanding will yield results that permit simplification of the problem set. At the same time, applications that are intention primarily upon learning new things generally have a limited focus that serves to constrain this process of over-complication, while it is important to distinguish form these experimental efforts towards understanding with attempts at actual solution of real-world problems that involve some kind of engineering design or schema and the imposition of some kind of articulatory framework on the ground.

It is in this last class of applications, in the devising of alternative systems in solution to certain problem sets, that the greatest difficulties can be seen to arise in the extension of systems-based approaches, that are by definition top-down to begin with. This is actually a problem characteristic to all forms of application--the challenge is only amplified with systems-based approaches that make any pretense towards comprehensiveness of solution sets. 

To a great extent, computing and the information revolution has reduced the proportions of this problem, and has offered forms of technology which allow us to deal with the information explosion attendant to problem solution in a much easier and better method than previously. This was evident in the first development of computing devices in the solution of the German enigma war codes during WWII, and continues to hold true today. We now have the advent of supercomputing architectures and systems, capable of processing massive amounts of information at very high rates, and supercomputing has thereby allowed us to reliably and accurately model complex systems, in considerable detail, far better than we were ever previously capable of doing.

The only stop-gap solution to the general problem of over-complication is to narrow and delimit the focus of effort in solution sets to particular achievable goals with the hope and expectation that the goals thus achieved serve larger strategic ends in a decisive way. It is needless to comment at this juncture that a great deal of human effort in our collective history has been wasted on the achievement of goals with little or no long term consequence. It is also fair to say that many projects and efforts have been undertaking without overall coordination of effort and resources, with the result being a reduplication of effort on one hand, and a needless squandering of resources on the other.

Again we find what we would expect to be a deadly blow to systems-based approaches to be in fact a problem shared by all applied approaches, and hence from the standpoint of system-specific issues, to be more the nature of  a problem of the management of effort and resources in an efficient way, of planning and goal-seeking, more than anything else.

It must be remarked at this juncture as well, that in experimental approaches aimed at gaining new knowledge, strategic decisiveness also plays a key role--there is no point in amassing myriad details of information if this information is not to be used for addressing any particular questions one may have in mind. It is not the goal of systems-based approaches to describe in nitty-gritty detail every event of everything that occurs. The success of systems based approaches rests on being able to make generalizations that are applicable to many different instances of reality, without needing to resort to the unpacking of all the instances--it allows us to reliably summarize reality without thereby having to elaborate this reality on a finer level.

This last point is not to defend systems based approaches or thinking from all manner of criticism--these kinds of criticisms are legitimate and they are born of experience. It is rather to offer a way out of such problems that such approaches typically face in their worldly articulation. It becomes essentially the systems solution to the general problem of complexity and the related problem of chaos that is associated with all underdetermined systems, and all real systems are underdetermined in their outcomes. It is possible to arrive at such solutions in a rather formulaic manner, with the proviso that there are always exceptions to the rules that can be dealt with by other heuristic problem-solving means. It becomes therefore as much a strategy and a methodology of attack of unknown or semi-known problem sets as it is of anything else--call it a tool-kit of various methods and approaches that can be used ad hoc in an encounter style approach.

I would conclude this brief critique of an exclusive "Systems" orientation by claiming that any approach to knowledge, any way of knowing and framing the world, can become paradigmatically and ideologically closed if we allow it to. If everything is a system and everything is part of the same system of reality, then this "system" really tells us a lot about nothing in particular. We may easily substitute words like "structure," organization, sense of order, as synonymous with systems and lose little in translation. When we are highlighting the "systems" aspect of everything and ignoring the detailed differences, we are merely pushing the differences of and between things onto the concept of systems, thereby displacing dealing with such differences in a more direct manner. There is an inherent virtue in maintaining a diversity of ideas and wide range of alternative frames of reference, that even go beyond systems based approaches. As a comprehensive frame of reference a systems-based approach to general and specific knowledge should be capable of transcending the kinds of dialectical and epistemological philosophical dilemmas that tend to underlie the presuppositions we make about the world and thus to undercut our view of the world in basic ways.

To summarize, systems approaches offer a methodological means for dealing with huge uncertainties in large areas of problem sets that are mostly unknown or only partly known. As such they provide us, if properly used, and not overused, a powerful vehicle for tackling problem sets with unknown solutions, and for dealing effectively and in a methodical way with high levels of uncertainty that are attached to problem sets of all kinds. When we know that with whatever problem set we encounter, there is a simplifying solution that is rooted to the "system" of organization that underlies the problem set at hand, we have already have of the solution in hand, and we have the means for deriving the other half of the solution in hand as well. It is wise to take the critique of systems to heart in an objective way, and it is the epitome of wisdom to turn these kinds of criticism to the advantage of the application of systems-based approaches to real world problem sets.

General Systems Essays, Vol. I