Systems Thinking Part 1: Philosophy

Students in my faculty of environmental studies are often presented with romantic ideas of interconnectedness, interdisciplinary approaches to their work and working within a worldview that counters traditional analytical and reductionist science. Unfortunately, instead of rigorously looking at what is wrong with traditional scientific approaches in their work, fixing it and coming up with a new approach they simply write science off as colonial or anti-feminist or whatever other buzzword they would like to apply. Then they pick a new, and often random, approach to their work. Rarely is the importance of solidifying the philosophy behind these ideas and mindfully focusing on a methodology to approach this philosophy stressed.

It became apparent to me in my last year of undergraduate that both of these activities are necessary for rigorous social science, activism and real world interventions to be approached successfully. Systems philosophy and methodology, now the focus of my master’s degree, incorporate and discuss these three romantic ideas in a way that makes them substantially more pragmatic and meaningful in discussions. A combination of all three – epistemology, methodology and practice are required for an approach that is holistic itself. Gerald Midgley argues that pure philosophy can ignore real world issues, pure methodology will never be tested for usefulness and pure intervention without the backing of the other two will lack foundation (2000). I hope to convince you of this in a three part series while providing you with an understanding of what “systems thinking” is. In part one I will look at the philosophy behind systems theory.

Is There a Problem with Traditional Science?

Most discussions of systems thinking begin with a discussion of why the limitations of traditional science are problematic…so I suppose that’s a good place for us to start, and I realize this may be controversial in this space. However, understanding the limits of reductionism is vital to understanding the importance of major systems concepts (which I will get to later).

Traditional science is expressed in a variety of ways. Peter Checkland defines it as having three integral characteristics: reductionism, repeatability and refutation (1999). Midgley calls it mechanistic science, being characterized as a method relying on observations to produce objective knowledge about the world (2000). Russell Ackoff describes the period for traditional science as the “machine age” characterized by a focus on reductionism and analytical thought, mechanism and determinism (1973). He describes it as deterministic because the effects observed are directly determined by causes. These characteristics do not sufficiently cope with complex situations. Social sciences, for example, usually involve complex problems that are not easily (or ethically) experimented on. Social problem situations (this is a systems term to describe your area of research) often involve a variety of actors with varying worldviews who can alter findings and ultimately the outcome of the study (Checkland, 1999). Science attempts to play a neutral social role in these situations, and does so very ineffectively (Resnik, 1998). This is because in complex social situations it is essentially impossible to remain completely objective and value-free. Thus instead of attempting to solve a problem with an inadequate methodology, it makes more sense that a new one should be created. This includes having scientists involved learn about ethical choices and morality and applying these dimensions (Resnik, 2009). Thus the act of observation should that take into account the value systems of actors as well as researchers themselves (Midgley, 2000).

Systems Theory as a New Paradigm

Thomas Kuhn’s famous work The Structure of Scientific Revolutions (1970) argues that science undergoes different phases. Whatever the current phase is, it is practiced by most scientists and is referred to as “normal science”. The “normal science” of a particular time period is associated with certain methods, theories, concepts and beliefs. This is referred to as the current scientific paradigm. He argues that as science progresses within the paradigm, limitations and anomalies will arise within the methods, theories, etc. At some point individuals decide if they will accept these limitations, or move to a new system. Eventually the anomalies, limitations and scientists abandoning the old paradigm becomes too numerous, or to use a systems term – reach a tipping point – the entire system revolutionizes and the new paradigm becomes normal science.  Ackoff suggests that this happened when switching from the machine age to a systems age (1973). The new systems age is characterized by expansionism, teleology and a new synthetic mode of thought. Expansionism is the idea that all objects and events are a part of a larger whole and teleology is the study of purposeful behaviour. Synthetic thought is essentially synonymous to early, an immature, definitions of systems thinking.

Ackoff describes a system as “a set of interrelated elements of any kind” (1973, pg 663) and as something that is “more than the sum of its parts; it is an indivisible whole” (1973, pg 664). Fritjof Capra raises an interesting question about this new systems approach:

“This new approach to science immediately raises an important question. If everything is connected to everything else, how can we ever hope to understand anything? Since all natural phenomena are ultimately interconnected, in order to explain any one of them we need to understand all the others, which is obviously impossible.” (Capra, 1997, pg 41)

His response, which I am quite fond of and takes some getting used to, is that this systems approach produces “approximate knowledge”. However never having complete or definitive answers is not problematic; it is a fact that researchers and scientists need to acknowledge while attempting to create as complete pictures of reality as possible. To do this, various actors and viewpoints should be taken into consideration when developing knowledge about a problem situation. It should be remembered that it is nearly impossible to create a fully comprehensive description of a complex situation and the world.

Previously I mentioned that understanding the limitations of science is important because it allows for the development of major systems concepts. From the idea that observation cannot be value-neutral and that understanding cannot be based on reductionist methodology, key concepts arise such as emergence, hierarchy and boundaries. Emergent properties are properties that are meaningless and not understood in a language that is appropriate to only lower levels in a hierarchy (or rather, they don’t make sense/can’t be understood out of context). They are the result of interactions of a system rather than one of the parts on its own (the problem is a system,one part does not stand separate). Midgley gives the example of drunk driving, which has the emergent properties of death, injuries, confrontations and public outrage (2000). Checkland describes emergence in the following passage:

“It seems convincing to everyone to describe the knowledge we have of the world in terms of different levels of complexity laws which operate at one level seem to be higher order with respect to those of lower levels. This is the kernel of the concept of “emergence”, the idea that at a given level of complexity there are properties characteristic of that level (emergent at that level) which are irreducible” (Checkland, 1999, pg 52)

Hierarchy theory is thus concerned with the inherent differences between the different levels of organization and the relationships that happen between them (Checkland, 1999). Additionally, once it is accepted that no worldview or knowledge base can ever be fully comprehensive or contain the whole truth about reality, the concept of boundaries becomes fundamental (Midgely, 2000). Boundaries are the scope of one’s research or problem definition. A simple example, provided by Midgely, is the world represented as a rectangle and the boundary as a circle in the middle of it. The circle, as the boundary, is defining what is in the scope of this particular project. Of equally importance is what is outside the scope of the circle, as it is being intentionally excluded (Midgley, 2000). Prior to Churchman­­, systems analysts assumed that boundaries were a “given” structure within reality (Midgely, 2000). Churchman was one of the first to demonstrate that boundaries are needed to limit knowledge within analysis (1970).

Some Last Thoughts on Systemic Philosophy

Midgley defines the ‘systems idea’ as: “If something is described as ‘systemic’ it is, as far as possible, comprehensively understood.” (2000, pg 34). This idea is strong in general systems theory, which is grounded in the idea that it is conceivable to have a mutual language amongst various academic communities (von Bertalanffy, 1956). Von Bertalanffy claims that all modern science of his time had an ambiguous term of “wholeness” associated with it and that systems theory makes this “wholeness” distinctive (1950). It does this by transcending disciplinary boundaries to gain more thorough and rich knowledge (Midgely, 2000). If someone attempts to construct knowledge based on a system with disciplinary boundaries, they are potentially hopelessly restricting themselves from the freedom to allow outside specialist languages to contribute to their own knowledge. Additionally, the original researcher’s knowledge may contradict other disciplines, and another professional could easily assist in preventing these contradictions. However, for this type of research to avoid becoming overwhelmed with knowledge, the previously described boundaries must be involved to limit the overall scope.

Systemic theory’s opposition to methods of reductionism and mechanism implies an inherent conflict between systems theories and subject/object dualism (Midgley, 2000). This refers to the separation of the observer (subject) and the observed (object) in a research situation. A dualist believes that the subject can remain independent of the object by standing outside of it. By doing so the subject does not influence the object in any way, and once the subject does influence the object than any objectivity vanishes. This naïve objectivism assumes that independence between the subject and object is possible. This, however, is inherently problematic with any anti-reductionist view point. In the systems view of the world everything is interacting with everything else and thus independent observation resulting in “objective” knowledge is impossible. This is important for developing one’s epistemological viewpoint on knowledge. Coming from a traditional science background it may be challenging to cope with the idea that knowledge produced has to be stated as being purely subjective in virtually all cases.

So that was all a bit more scattered than I had hoped it would be but I think it touches on important philosophical ideas behind systems theory. In the next part I will write about method, why I chose to use the method of systems that I did and how I’m using it. Part three, will look at why the philosophy and method are BOTH important to understand to have a holistic approach and understanding to ones research…and meaningful intervention.

Works Cited In The Series (and all great books on systems theory):

Ackoff, Russell L. (1973). Science in the Systems Age: Beyond IE, OR, and MS. Operations Research 21(3): 661-671.

Bateson, Gregory. (1972). Steps to an Ecology of Mind. Chandler Publishing Company: Philadelphia, PA.

Bertalanffy, Ludwig von. (1950). An Outline of General System Theory. The British Journal for the Philosophy of Science 1(2): 134-165.

Bertalanffy, Ludwig von. (1956). General Systems Theory. General Systems Year Book 1: 1-10.

Capra, Fritjof. (1997). The Web of Life: A new scientific understanding of living systems. Random House of Canada: Mississauga, ON.

Checkland, Peter (1981) Systems Thinking, Systems Practice. John Wiley and Sons, Ltd: West Sussex.

Checkland, P. and Scholes, J. (1999). Soft Systems Methodology in Action. John Wiley and Sons, Ltd: West Sussex.

Churchman, C. West. (1968). The Systems Approach. Dell Publishing: New York, NY.

Churchman, C. West. (1970). Operations Research as a Profession. Management Science 17: b37-53.

Fitzgerald, L.A. (1999). Why there’s nothing wrong with systems thinking a little chaos won’t fix? A critique of modern systems theory and the practice of organizational change it informs. Systemic Practice and Action Research 12: 229-235.

Forrester, J. W. (1961). Industrial Dynamics. MIT Press: Cambridge, MA.

Gleick, James. (1987). Chao: Making a New Science. Penguin: London, UK.

Jackson, M.C. (1991). Systems Methodology for the Management Sciences. Plenum: New York, NY.

Kuhn, Thomas S. (1962). The Structure of Scientific Revolutions. The University of Chicago Press: Chicago, IL.

Midgley, Gerald. (2000). Systemic Intervention: Philosophy, Methodology and Practice. Kluwet Academic/Plenum Publishers: New  York, NY.

Resnik, David B. (1998). The Ethics of Science: An Introduction. Routledge: London, UK.

Resnik, David B. (2009). Playing Politics with Science: balancing scientific independence and government oversight. Oxford University Press: New York, NY.

Previous post

Awesome Sauce Music Friday! Something Funky with the Universe Edition

Next post

Teen Skepchick's Reality Checks 8.27



Katie is a graduate student from Canada studying the environment and systems theory. She also loves dinosaurs and baking cupcakes. Follow her on twitter @katiekish

1 Comment

  1. September 13, 2012 at 11:45 am —

    Reblogged @katiekish “Systems Thinking, Part 1: Philosophy” on Systems Sciences group at from

Leave a reply