Thinking in Systems: What is a System?
Thursday, April 16, 2009 at 10:56AM in Systems Thinking Three kinds of things are necessary for something to be a system, as Donella Meadows interprets the definition.
Elements
These are the things that are easy to notice and describe, even though they might be ideas rather than stuff, such as the reputation of a department as well as the number of people in it.
Interconnections
What relationships hold the elements together? What are the ways that one affects another?
Purpose
This is not necessarily the intended purpose, but rather what the actual behavior is leading toward. If the behavior doesn't lead toward anything, then it is not a system. There is often not just one purpose, since most systems have subsystems and these can have competing purposes.
Other definitions exist, many of which use more precise and academic wording, such as those at the Free Management Library and the open course on Systems Thinking and Practice.
Dynamic
Systems operate over time.
This is not included as part of the definition, but is understood as a necessary part of trying to discuss system behavior. I wanted to emphasize it here because it's a piece that I sometimes have to remind myself about. Even while looking at visualizations of the movement of systems over time I sometimes drift into considering them as static entities in and of themselves, and not as mere frames captured from a scrolling feed.
When you talk about time you are usually talking about the elements as stocks - how much of something is there? - and the interconnections as flows - how fast is something changing and what is that change?
The number of trees is a stock. Logging and planting are flows and how fast those activities occur can be changed. Meadows represents the idea with a commonly used model of trying to fill a bathtub while the drain is open.
One last thought on time - a quote which is elaborated on more thoroughly in the book.
A stock takes time to change, because flows take time to flow.
This post is the second in a series that discusses the concepts in Thinking in Systems
View Printer Friendly Version
Email Article to Friend
Reader Comments (2)
Yep, I worked in a Department with management scientists and much of first year was taught teaching students to think systemically. Here is a little gift of a game I play to develop a habit of recursive thinking.
http://www.kozisgreat.com/games/recursive.html
I've worked with management theoriest too, who believed purpose was not part of systemic thought. It seems setting a bounday or constraint often gives a group a structure that provideds a proxy of purpose - wonder if you have thoughts on that.
And as a psychologist, well, we weren't raised on dyanmic systems. Losada's work using Lorenz equations to model happiness is an exception.
Great blog. Will check you out weekly!
Hello Jo,
That's a thought I haven't heard in my learning yet - that a constraint can be a proxy of purpose. Do you remember where the idea came from? I'm having trouble figuring out how to relate to it concretely.
Yes, as a psychologist I can see how the dynamic part wouldn't be included in what you usually call a system. It wasn't in materials science either. We only considered time at very specific junctures.
Hope you enjoy your upcoming visits, too. Thanks for commenting.