Thinking in Systems
Thinking in Systems
Donella Meadows
What is a system?
Self Organization, the ability of a system to adapt and make itself more complex
Hierarchy, the formation of subsystems with their own functions and goals
Why Systems Surprise Us
A system is a mental model, it never fully represents reality
Events and behaviors are what we naturally pay attention to, but they’re not
necessarily representative of the underlying structure of the system
Boundaries are set to simplify a model but everything is part of a giant system
without boundaries. We make rational decisions based on the information we
have...but we often don’t have the right information
Non-linear relationships, systems often don’t behave in a linear way like we might
expect, especially when they encounter the limiting factors
How Systems Go Wrong
Policy Resistance - Fixes that Fail
Goals of subsystems are different, inconsistent or competing against each other. More effort = more
resistance. Align the goals of the competing parties with a better overarching goal that everyone supports.
Addiction
Shifting the burden onto a solution that doesn't actually improve the root problem, and actually harms the
ability to deal with the root problem. Help it help itself.
Seeking the Wrong Goal - You Get Exactly What You Ask For
When the goal is not representative of what will actually fix the problem. You get exactly what you ask
for...and it might make things worse! Confusing effort with result.
Leverage Points
A leverage point may be intuitive to find to those familiar with the system, but it is
often pushed in the wrong direction. They're counterintuitive.
Buffers
Stocks which are big relative to their flows are more stable, but less flexible
They can have an impact as leverage points...but are often disqualified by physical limitations...e.g. the
carrying capacity
Physical Structures
Can have huge impact on a system, but once it is established (all roads in Hungary lead to Budapest) it is
very hard to change
Leverage Points
Delays - Between information and change
It's impossible for a system with long delays to respond to fast changes, and a system with too short delays
causes overreaction. Delays can have a huge impact on the system, but are often things cannot be made
faster. Usually easier to increase a delay.
Information Flows
Link the missing information that matters and will change behavior to a compelling position.
Example: Town's water inflow pipe downstream of waste outflow pipe
Leverage Points
Rules - Incentives, punishments, constraints
The rules and who has power over them. Constitution, Laws of Physics, etc.