photo credit: www.carnegiehall.org

For those of us tackling societal problems, “systems change” has become a buzz word. Confronted with persistent problems — in education, healthcare, capitalism, immigration, public health, criminal justice, police… you name it — we’re told to engage in systems thinking.

We have to look beyond our silos, find root causes, and identify what drives outcomes. Only by changing the structure of the system can we solve our problem.

But what do systems thinking and systems change really entail? How do we know whether we’re doing it right?

Many of us in the United States were introduced to systems thinking through systems dynamics, and the two have become nearly synonymous.

Systems thinking is a generic name for an emerging set of study areas that includes systems dynamics, but also biomimicry, design thinking, complexity theory, human systems dynamics, and numerous others.

The basic idea behind all of these is to zoom out to focus on a system: a set of interconnected parts that form a greater whole.

How you understand a system depends on where you sit in it — from the inside, it’s so complicated that it’s virtually impossible to see the whole thing and all the parts at the same time.

Take the human body for example. Even though medical science has chopped up our understanding of the body into numerous specialties, the body itself is a system of interdependent parts. The liver is useless if there’s no heart.

What systems thinking helps us understand is that it’s less about the individual parts, and more about how the parts relate to one another.

Systems thinking via system dynamics

Trying to see the whole picture is where systems dynamics comes in. It creates a representational data-driven model to identify the key parts, and how they relate to one another (scroll down if you want to see example — but fair warning—your eyes may glaze over!). A system dynamics model helps us comprehend the wider system and what variables matter most.

Systems dynamics is an incredibly useful way to step away from using linear logic to solve problems. A good example of that is freeway expansion to alleviate traffic congestion.

Add more lanes, less congestion, right? That’s linear logic. But the feedback loops of systems dynamics tell us that open freeways entice more development (bringing more cars) and pretty soon we’re back to where we started with traffic.

They can help bring the system into view, but don’t rely on systems dynamics models to help you know what to do to bring about systems change. Here’s why in a nutshell.

Just because we know what matters, it doesn’t mean we know what to do

Systems dynamics is built on the premise that if we can see the big picture, we’ll know what to do. It helps us identify so-called leverage points: the key variables that matter a lot to how a system functions. The premise is that systems change occurs by manipulating the key levers.

If you’re unfamiliar with what a systems dynamics model looks like, here you go. This one describes the demand for maternal and neonatal health services, and as far as these models go, the language is pretty understandable.

Suppose you are part of the system depicted above. Can you see where you might make a difference to bring about systems change? Would you know where to start? Does it tell you will actions will be effective?

In my experience, systems dynamics can immobilize rather than inspire people. When confronted with the enormity of seeing the entire system laid out before them, many people feel overwhelmed:

• I’m just one little person in one little corner of this gigantic map — what on earth can I do that would be meaningful?
• So many things are important; how do I decide which to focus on?

Even if we know exactly what leverage point we want to focus on, systems dynamics doesn’t tell us what interventions are effective at changing the system. Leverage points are not actions. In the diagram above, we might decide to focus on maternal literacy (far right top). But what do we do about maternal literacy?

Even if we know what to do, it doesn’t mean we know how to do it

Hmmm. We know the levers for climate change and COVID-19, yet both are still raging. Knowing what to do is not the same as knowing how to do it. “What” is a technical consideration. “How” is a messy human consideration.

Simplification can distort reality in critical ways

Systems dynamics models are useful because systems are so complex we can’t see it in our heads — there are too many interrelated variables involved. So, like all models, real life has to be simplified in order to be modeled.

In order to simplify, systems dynamics models use “expected case” data. In general, children can eat peanuts; that’s the expected case. But for one in every fifty, eating peanuts can be fatal.

In complex systems, local conditions matter a lot. The diet for 49 children can include peanuts; but it better not for the 50th. Systems dynamics can provide you with a generalized description of how things work. But generalized descriptions of how things work don’t necessarily provide the best guidance on what to do in specific cases.

Let’s broaden this and see how it plays out with dominant and non-dominant cultures. In the expected case with criminal justice, a person is innocent until proven guilty. The police are helpful, fair, and exercise restraint. More policing reduces crime. Suppose Black Lives Matter drew a systems map of policing and public safety. Do you think the feedback loops would look the same?

Systems dynamics maps depend on who is constructing them, even if they use research data.

Systems are in constant motion

Systems dynamics models are actually static, despite being named dynamic. They are dynamic because cause and effect is not linear. They are static because the system as modeled never changes; it is locked in and always reacts the same to the same stimuli.

But we know that systems aren’t static. At first Black Lives Matter was ignored or even disparaged. Seven years later, it’s a clarion call. Why? What changed?

Every small break in the status quo — whether someone pays attention to yet another murder of a Black person or a Black commentator stresses the injustice of it—shifts our awareness and mindsets a little bit. Shift. Shift. Shift. Until the pattern of ignoring police killings of Black people is broken.

When the pattern shifts, the reactions in the system change significantly, but the systems dynamics model won’t pick that up unless it’s updated, and it is impossible to update fast enough to keep up with what is happening in the real world today.

Why does this matter? Because models can give you a false impression of how systems change actually occurs. Systems change rarely comes in the form of a thunderbolt. Instead, it is the accumulation over time of actions, large and small, that shift the potential for the system to change, and trigger it to change when it is ready.

The big reveal: systems change isn’t a grand plan

Some folks become overwhelmed by systems dynamics. Others spring into action, hatching a grand plan for systems change. They start searching for the “right” leverage points and the “right” actions, and lay them out in a detailed action plan.

I once walked into the coordinating office of a major federal community development project. This project was about systems change. The staff had covered an entire wall showing all the players and agencies and involved, and how they were linked. I chuckled at the hubris of it.

Not the mapping itself; we should know how things are linked. But the clear implication was that all of these people and agencies posted on the wall were going to be directed through this one office, as if it were a grand puppet show where the puppet master held all the answers and controlled all the action.

The grand plan won’t work. It won’t work because that’s not how change happens in complex systems.

However, you matter a lot. In the next post I’ll share some tips for how to see the big picture and become inspired instead of overwhelmed.