Systems Thinking
The structure of a bathtub system—one stock with one inflow and one outflow, as diagrammed by Donella Meadows.
In times of trouble I am especially inspired to see the world in a way that helps me understand why things are that way and what we could do about it. I see systems theory as a helpful tool for understanding complex situations, like what the world is experiencing now. Here is an excerpt of something I wrote in grad school:
Donella Meadows, a great systems thinker, describes a system as “ an interconnected set of elements that is coherently organized in a way that achieves something” (Meadows 11). Systems thinking seems to elude a single definition, but Wikipedia states: “Systems thinking is not one thing but a set of habits or practices within a framework that is based on the belief that the component parts of a system can best be understood in the context of relationships with each other and with other systems, rather than in isolation.” This statement fits with much of Meadow's work. In her book Thinking in Systems, A Primer she describes how elements work with interconnections and a function or purpose to form a system.
According to Meadows, elements are often the most visible aspects of systems. Elements of a digestive system might be teeth, stomach, and intestines—more obvious than the chemical signals regulating them. Not all system elements are tangible, though. In a social system a general attitude, like school spirit, may be an element. Elements are related through interconnections, which are system flows and their controls. Flows may be literal, as in the flow of water in a tree, or figurative, like the flow of information in a school.
Often the most difficult to see aspect of a system is its function or purpose. Meadows notes that a system's function is not always what it purports to be or what its constituents desire, but is best ascertained from what the system actually achieves. In designing and modifying systems, then, it is important to remember that simply intending an outcome will not, in itself, lead to that outcome. The rest of the system must be analyzed.
To understand systems behavior, it is helpful to think of them in stocks and flows. Meadows has a diagramming system (see above) based off an analogy of a bathtub, with controls monitoring water going in and out and, between, a volume available for the collection of water. A household heating system provides another example of sinks and flows. A thermostat is set to a certain temperature, which the furnace tries to match, and, if the outside air temperature is colder than the thermostat setting, the heat from the house will leak out. Insulation and air barriers may control the rate at which it leaks.
In the example of the household heating system, the furnace is relying on a feedback loop with the thermostat. Feedback loops can only inform future decisions—in this case, whether the furnace should turn on or off. They cannot correct past behavior. The length of time it takes for the thermostat to register the actual temperature, to compare it to the set temperature, and for the furnace to turn on or off all affect the length of the feedback loop. Since the feedback information can only affect future behavior, if it takes a long time to communicate the information, then the goal—set point—may be overshot. The furnace may overheat the house or let it get too cold. Another way to say this is that the system will experience oscillations. Meadows draws graphs, like the one shown below, to understand system behaviors over time.
Once the actual dynamics of the system are understood, leverage points can be located for altering the system. For instance, the graph of room temperature may show that, as the difference between inside and outside temperatures increases, so does the discrepancy between desired and actual room temperatures. If the purpose of the system was to maintain an even temperature, insulation and air barriers could be added to the house, a larger furnace could be installed, or the set temperature could be lowered. There may be more than one purpose for a system—for example, the intent in the house may be to keep a comfortable temperature at minimum cost—and systems can quickly become quite complicated.
Alterations of the system goals have a greater affect on the system's performance than the individual parts, flows, or feedback loops. In the bathtub example, if the goal of the system is not to be a bath at all, but rather to drain the hot water heater, one can imagine that the results will be different—water will run through without being stored in the tub (except to the degree that the infill flows faster than the outflow). Meadows has compiled a list of system leverage points—places in a system where an intervention has significant impact—in ascending order of their effectiveness (see quote below). The list is meant as “an invitation to think more broadly about system change” rather than as a recipe for finding leverage points (Meadows, 1997).
“9. Numbers (subsidies, taxes, standards)
8. Material stocks and flows
7. Regulating negative feedback loops
6. Driving positive feedback loops
5. Information flows
4. The rules of the system (incentives, punishment, constraints)
3. The power of self-organization
2. The goals of the system
1. The mindset or paradigm out of which the goals, rules, feedback structure arises”
She also has broad advice for those working with systems of any scale. Below is her condensed list from the last chapter of the Primer.
“- Get the beat of the system
- Expose your mental models to the light of day
- Honor, respect, and distribute information
- Use language with care and enrich it with systems concepts
- Pay attention to what is important, not just what is quantifiable
- Make feedback policies for feedback systems
- Go for the good of the whole
- Listen to the wisdom of the system
- Locate responsibility in the system
- Stay humble and stay a learner
- Celebrate complexity
- Expand time horizons
- Defy disciplines
- Expand the boundary of caring
- Don’t erode the goal of goodness”
Meadows, Donella H. (1997) “Places to Intervene in a System.” Whole Earth, issue 2091, article 27.
Meadows, Donella H. (2008). Thinking in Systems, A Primer. White River Junction, Vermont: Chelsea Green Publishing.
When designing your project, it is helpful to think about what it is you really want, and why. Getting past assumptions about what is “good” and how things work is the first step. The second step is to witness what is and how things really work. We might ask questions like, “Where is the weather actually coming from on this site? What is a comfortable reach for me? Will heating the air here make me more comfortable, or just lead to draftiness? Would heating a surface in the room be more effective? Would I actively use more storage, or just use it as a way to avoid dealing with things?” Then, a design team (say, you and I) can cleanly apply actual goals and get results.
I find that when we know what we want—really know, from first-hand experience and from listening to what the whole system wants—then we tend to commit to our projects, and they tend to happen with wholeness, goodness, and true value.