Systems and Research

I have been struggling for years now with issues about systems.  First, how do you introduce the basic principles of systems to students, or other people who are interested but have no previous background?  There have been many different books and articles written at different levels, for different audiences, but (for me) all of them still leave questions or create additional confusion about what is really different about understanding things in terms of systems.

The most basic writings tend to use simple examples, explaining that a system is a collection of parts that creates a “whole” entity.  You can have a box full of tires, spokes, chains, sprockets, metal tubing, etc.  Only when those parts are assembled in a certain way, though, do they make a bicycle, which then has functionality as a means of transportation.  The relationship of the parts to each other matters.  But this is still a simple, mechanical system and often leaves the question, “so what – how does this explain anything new?”

Other common examples include the fact that there is nothing about the “wetness” of water that can be explained by the properties of oxygen or hydrogen molecules.  It is only their formation into a system that allows this new property to emerge. The usual question is again, “so what?  Isn’t this just like evolution?  Over time, some things combined with others to form new, more complicated things?”

The more complex writings about systems start at the other end of the spectrum.  Most quickly delve into philosophy and questions about the nature of human knowledge.  Most also question traditional approaches and assumptions in science, while at the same time attempting to build on what is already known.  In addition, many of the founding theorists and writers in systems were trained as scientists, and as such used mathematical formulas as the description and justification for their findings.

All of this only became more complicated as I tried to work with students in using systems as an approach for research.  While there are a number of systems methods that make use of quantitative, computer-based models, they do not necessarily change the basic research approach.  Agent-based modeling, for instance, is considered by many to be a systemic method, but there are many people using agent-based modeling that have no orientation to systems, as such, at all.  So even in research and articles published in systems journals, or at systems conferences, there is often a looming question:  is this particular piece of research systemic, and if so, is it because of the particular approach or methodology (and is that obvious), or is it only because of the orientation of the researcher (the user of the tool) or is there no necessary difference?

My own orientation to systems is much more basic and pragmatic, and that might be a helpful place to begin.  (I have clearly drawn on and learned from many, many other theorists and writers; teachers and colleagues, but at this point is it much clearer for me to present my amalgamation of those ideas than to try to trace and credit every thread.)

Systems is for me a way of seeing and understanding; it is a way of framing or defining an issue.

For me, systems are patterns of activity.

This does not necessarily separate systems from traditional science.  In physics and chemistry there is dynamic equilibrium.  Atoms and molecules remain ever in motion, even within the most solid-looking substances, like granite.  They just remain in a well-enough constrained pattern that they appear to us not to change.  Most gasses, on the other hand, appear to have no substance at all.  (Experiencing a gale or a hurricane or tornado changes that perception, of course.)

The things that we experience in life are all patterns of activity of one sort or another, held in dynamic equilibrium to different extents.  Some are physical objects made of material elements; the chair that I sit in and the house in which I live.  The chair lasts for a time; the house I expect to last longer.  The water that I drink appears to be “gone,” but is only changed.

Many of the important things in our lives have strong psychological components.  The stability of my home is more important than the structure of my house.  I need for the people I love to remain “who they are” rather than becoming strangers.

This is where understanding – and research – become difficult.  What are these things that I find so important, and how do I really come to know them?  This is also where the break with traditional science often occurs.

I can understand many material objects by understanding the underlying components.  I can take a watch apart and see what makes it “tick.” I can dissect a frog in science class and see the organs, but they only give me a limited amount of information in that state.  (I can put the watch – a mechanical object – back together and it will work.  I have to let other frogs make new frogs – a very different kind of system.)

Students in organizational studies often want to understand very complex things.  Traditional scientific approaches normally imply either that (1) you can take it apart to see what makes it work, or (2) you can observe it like a culture in a Petri dish – and then in both cases you can measure and explain your observations with mathematical accuracy.  Social science approaches sometimes limit the accuracy that can be expected by working only towards measures of correlation rather than causality, but still with a goal of being able to say something about the phenomenon in general (i.e. how it applies to the larger population) not just about the subjects of the study.

In the first case, identifying the parts of an organization almost inevitably results in the idea that “organizations are made up of people.” That is somewhat correct, but mostly not.  All human social systems of every kind are comprised in some way of people, and the attributes and activities of people.  Saying that people constitute organizations is like saying that gears and springs constitute watches.  They sometimes do, but there are gears and springs in lots of things other than watches.  And when you take some watches apart there are no gears and springs – there are parts that look more like the workings of cell phones or computers.  The most unique and most common characteristic of a watch is its being an instrument for measuring time.  There are many other instruments added onto watches today (e.g., altimeters, thermometers, heart monitors, global positioning systems, etc.) but if the basic instrument LACKS the ability to measure time, it is probably not going to qualify as a watch.

Complexity often leaves us in a quandary about identifying exactly what something is, and therefore explaining and predicting how it will work or behave.  In that case, the best that we can often do is to explain what it’s like.  It can be helpful, but it also sometimes comes at a price because the way that we understand something greatly affects what we expect of it.  Gareth Morgan’s classic book, Images of Organizations, for instance, offers eight metaphors through which organizations might be seen (as a machine, as an organism, as a brain, etc.)  Fortunately, he is careful to point out that “any theory or perspective we bring to the study of organization and management, while capable of creating valuable insights, is also incomplete, biased, and potentially misleading” (p. 5).  Studying neural networks inside of Lehmann Brothers may provide useful insights, but it may not tell you anything about why it just filed for bankruptcy.

Most students, of course, don’t attempt to study an organization as a whole entity.  Many attempt to study attributes or characteristics which they believe significantly affect organizations – for instance, transformational leadership.  From a research standpoint, this still raises the same questions.  What is this thing, this phenomenon, and how should you go about studying it?  Most research courses throw students back to the approach of defining the variables of the phenomenon (taking it apart) and measuring them, with the hope that the measurements will add back up to the phenomenon.  Not surprisingly, there is an instrument (the Multifactor Leadership Questionnaire) which breaks down the attributes and supposedly will “determine the degree to which leaders exhibited transformational and transactional leadership and the degree to which their followers were satisfied with their leader and their leader’s effectiveness.” Personally, I think that this puts us back at the watch analogy.  The characteristics that get measured show up in lots of places (like gears and springs.)  Gears and springs are one way to make a timepiece, but they aren’t the essence of it.  You can make time devices with sand dropping through an hourglass, or using a sundial.  So if you want to study leadership, what is IT, and then what difference does it make?

One thought on “Systems and Research

  1. I myself have grappled with the issue of understanding, teaching and writing about systems. My solution has to been to avoid the use of the single word “systems”, and to combine the word with others to provide a context.

    In business and organization studies, I like to make the contrast between systems models and systems metaphors.

    Systems models, to me, are associated with systems science. Following the ideals outlined by Greek philosophers, science is a pursuit of truth. It requires phenomenon that are repeatable, observable, and measurable. Models are associated with theories — which are abstractions — and can be replicated and/or reproduced at varying levels of rigour.

    System metaphors, on the other hand, are much more how human beings understand the world. Gareth Morgan’s Images of Organization certainly follows this spirit. Jay Ogilvy once cited that human beings speak in 5.2 metaphors per minute. Scientific facts are hard to substantiate, and are generally too dry to consume. Motivational speakers apply rhetoric as an art.

    In management and organization, we need both systems science and systems metaphors.

    Science isn’t the best foundation when we don’t have a complete history of observations, and a decision for action is required. This is reflected in writings of Nassim Taleb (The Black Swan) and Jerome Ravetz (Postnormal Science).

    System metaphors can lead us astray because a mental image may not be tangibly possible or practical. Time travel makes great science fiction, but tends to ignore the second law of themodynamics (entropy).

    Having both models and metaphors share a common foundation of systems enables a strong foundation of language on which we can have discussions. From business consulting work that I had done in China, I discovered that management concepts based on systems basics (e.g. function, structure, process) were easily translatable and well-understood.

    Perhaps the biggest challenge with systems is that they’re so pervasive that it takes some effort to appreciate the patterns of the ordinary and the everyday.

Leave a Reply

Your email address will not be published. Required fields are marked *