Think in systems | Milen Vasilev

This is primarily a summary of the book Thinking in Systems by Donella Meadows.

I have used Bard and ChatGPT along the way so this is mostly written by them. See notes after each section. I have used this as an exercise in summarising the book as well as testing the two models.

Bard’s initial summary

Thinking in Systems is a book by Donella Meadows that introduces the systems thinking approach to problem-solving. The book begins by defining systems thinking and then goes on to discuss the key concepts and principles of systems thinking. Meadows then provides a number of examples of how systems thinking can be used to solve problems in a variety of fields, including business, education, and the environment.

Some of the main points in the book that stand out to Bard:

Feedback Loops
Stocks and Flows
Limits to Growth
Entropy and Irreversibility
Hierarchy
Leverage Points
Seeing the Forest and the Trees
Learning to Think in Systems
Systems Thinking and the Future

Meadows argues that systems thinking is a powerful tool that can help us to understand and solve complex problems. She provides a number of examples of how systems thinking has been used to solve problems in a variety of fields, including business, education, and the environment.

Here are some key takeaways from the book:

Systems thinking is a way of understanding the world that emphasizes the interconnectedness of things.
Systems thinking can help us to see patterns and relationships that we would not otherwise see.
Systems thinking can help us to understand why problems persist even when we try to solve them with traditional methods.
Systems thinking can help us to develop solutions that are more likely to be successful in the long run.

ChatGPT’s initial summary

Here are some of the main points from the book Thinking in Systems by Donella Meadows:

Systems thinking is a way of understanding complex phenomena that involves looking at the relationships between different components of a system.
Feedback loops are an important concept in systems thinking. They can be positive (reinforcing) or negative (balancing), and they play a key role in determining a system’s behavior.
Systems can be broken down into stocks (accumulations) and flows (rate of change), and understanding the relationship between them is critical to understanding a system’s behavior.
Systems can be classified into different types, such as simple, complicated, chaotic, and self-organizing systems, and different types of systems require different approaches to understanding and managing them.
Mental models, or the assumptions and beliefs that we bring to our understanding of a system, can affect how we interpret data and make decisions about the system.
Leverage points are areas in a system where small changes can have a big impact, and understanding them can be crucial to making meaningful changes in complex systems.
Systems thinking can be applied to a wide range of problems, including social, economic, and environmental issues, and it can help to identify root causes of problems and find solutions that address those root causes.
Systems thinking has limitations, and it’s important to remain humble in our understanding of complex systems and to be aware of the potential unintended consequences of our actions.
Overall, the book encourages readers to think more holistically about the systems they encounter in their daily lives and to apply systems thinking principles to better understand and solve complex problems.

My take – concepts I am taking away from the book as useful

Some of the main points of the book that I find interesting and useful:

Dynamic equilibrium
Self-organizing
Complexity is the highest at borders (interaction points)
Layers of limitation
Delays should be accounted for in the systems
Government as a system – Why Nations Fail by Daron Acemoğlu, James A. Robinson
Feedback loops
Leverage points
Short term relieve vs long-term restructuring
Diversity leads to optimal long-term outcomes
Paradigm shifts are difficult
Systems theorizing will not get us to solve all problems but will help us avoid some wrong turns
Applying the scientific method

Let’s take a look at these from up close.

Dynamic equilibrium

Dynamic equilibrium is a state in which a system’s behavior remains relatively stable over time despite ongoing changes to the system’s inputs and outputs. This occurs when the system’s feedback loops are able to maintain a balance between opposing forces, resulting in a relatively steady state of the system’s behavior.

Note: This is ChatGPT’s answer. Bard’s did not quite cut it.

Self-organizing

Self-organizing systems are systems that are able to spontaneously create patterns and structures through the interactions between their components. Meadows emphasizes that self-organizing systems can be both beneficial and problematic, as they can create new and innovative solutions to problems, but they can also reinforce undesirable behaviors or structures. Understanding the feedback loops that drive self-organizing systems is important for managing and shaping these systems to achieve desired outcomes.

Meadows provides a number of examples of self-organization in her book, including the example of the human body. She says that the human body is a self-organizing system because it is able to maintain a constant internal environment despite changes in the external environment. This is due to the body’s ability to regulate its temperature, blood pressure, and other vital functions.

Note: The first paragraph is from ChatGPT. The second is from Bard.

Complexity is the highest at borders

Donella Meadows argues that complexity is highest at the borders or boundaries between systems, where different elements or systems interact with one another. This is because the interaction between different systems can create feedback loops and new patterns of behavior that are difficult to predict or understand without looking at the system as a whole. For example, the boundary between an ecosystem and human society is complex, as the actions of humans can have unpredictable and far-reaching effects on the ecosystem, and vice versa. To understand and manage complex systems, Meadows emphasizes the importance of examining the boundaries between systems and questioning the assumptions and mental models that underlie our understanding of those boundaries.

Note: This is ChatGPT’s take. Bard was too long-winded and out of the book’s context. I keep having to ask Bard to use the books for context.

Layers of limitation

Meadows explains that systems are often constrained by various layers of limitations that can hinder their performance and effectiveness. These limitations can arise from the design, implementation, or maintenance of physical, technological, or societal systems. It is essential to identify and address these limitations to optimize the functionality of the system.

In the context of systems thinking, physical limitations can indeed be challenging to fix because they are often related to the design or construction of a physical system. For example, fixing a broken bridge or rebuilding a damaged building can be a costly and time-consuming process.

Note: Selected bits from ChatGPT’s answer.

Delays should be accounted for in the systems

Delays can occur for a variety of reasons, such as:

The time it takes for information to travel from one part of the system to another.
The time it takes for people to process information.
The time it takes for machines to process information.

Delays can have a number of negative consequences, such as:

Causing the system to become unstable.
Causing the system to produce incorrect results.
Causing the system to be unresponsive.

Note: Bard wins this time with a simple bullet summary.

Government as a system

In the context of systems thinking, the government can be viewed as a complex system with numerous components and interdependencies. As a system, the government is responsible for establishing laws and regulations, providing services, and managing public resources.

Like any system, the government can be influenced by various internal and external factors, such as political pressures, economic conditions, and social dynamics. These factors can impact the effectiveness and efficiency of the government in achieving its goals and serving the needs of its citizens.

Understanding the government as a system requires a comprehensive analysis of its components, feedback loops, and interdependencies. By applying systems thinking, it is possible to identify areas for improvement and develop strategies to optimize the performance of the government as a system.

Note: This is ChatGPT’s take – its answers are a little more rounded than Bard’s.

Great book on the topic: Why Nations Fail by Daron Acemoğlu, James A. Robinson

Feedback loops

Feedback loops are an essential component of systems thinking that help us understand how a system behaves and responds to changes over time. By analyzing feedback loops, we can identify problems and develop strategies to optimize the system’s performance.

There are two types of feedback loops: positive and negative. A positive feedback loop occurs when a change in a system leads to an amplification of that change over time, while a negative feedback loop occurs when a change in a system leads to a dampening of that change over time.

Positive feedback loop: Climate change

As temperatures rise, ice caps melt, reducing the amount of sunlight reflected back into space
This causes more heat to be absorbed by the Earth’s surface, further increasing temperatures
This creates a self-reinforcing cycle that continues to amplify the effects of climate change

Negative feedback loop: Thermostat regulation

A thermostat senses that the temperature in a room is too high
It sends a signal to the air conditioning unit to turn on
As the air conditioning unit cools the room, the temperature drops
The thermostat then senses that the temperature is within the desired range and sends a signal to the air conditioning unit to turn off
This creates a self-regulating cycle that maintains a stable temperature in the room

Note: This is ChatGPT – follows directions better than Bard.

Leverage points

Leverage points are specific areas or components within a system where a small change can have a big impact on the overall behavior or structure of the system. In Thinking in Systems, Donella Meadows identifies 12 different leverage points within systems, ranging from the most superficial (e.g. changing the names or numbers of elements within a system) to the most profound (e.g. changing the goals or mindset that underlie a system). Meadows argues that understanding and targeting these leverage points is critical for achieving significant and lasting change within a system. By identifying the most effective leverage points within a system, it’s possible to create changes that ripple through the system and have a broad impact on its behavior or structure.

Some of the key areas where leverage points are found are:

The goals of the system. The goals of a system are the desired outcomes that the system is trying to achieve. If you change the goals of a system, you can change the way that the system operates.
The rules of the system. The rules of a system are the constraints that govern how the system operates. If you change the rules of a system, you can change the way that the system behaves.
The structure of the system. The structure of a system is the way that the different parts of the system are connected. If you change the structure of a system, you can change the way that the system functions.
The feedback loops in the system. Feedback loops are the processes by which the output of a system is used to modify the input of the system. If you change the feedback loops in a system, you can change the way that the system responds to changes in its environment.
The paradigm of the system. The paradigm of a system is the shared understanding of how the system works. If you change the paradigm of a system, you can change the way that people think about the system and how they interact with it.

Note: The first paragraph is an intro from ChatGPT. The bullet list is from Bard.

Short term relieve vs long-term restructuring

Meadows emphasizes the importance of distinguishing between short-term relief and long-term restructuring when it comes to addressing problems within a system. Short-term relief refers to actions that provide immediate relief or temporary solutions to a problem, but do not address the underlying causes of the problem. Long-term restructuring, on the other hand, involves addressing the root causes of a problem and making changes that will result in lasting solutions. Meadows argues that while short-term relief can be necessary in certain situations, it’s important not to lose sight of the long-term goal of restructuring the system to address the root causes of the problem. By focusing on long-term restructuring, it’s possible to create sustainable solutions that address the underlying causes of problems and prevent them from recurring in the future.

Note: ChatGPT had the above brief summary which is all I needed. Bard was good but a little long-winded and not in the context of the book.

Diversity leads to optimal long-term outcomes

Meadows argues that diversity can lead to optimal long-term outcomes in complex systems. This is because diverse systems are better able to adapt to changing conditions and are less susceptible to catastrophic failures than systems with less diversity. Meadows notes that diversity can take many forms, including diversity of species within an ecosystem, diversity of ideas within a social system, and diversity of skills and perspectives within an organization. By encouraging and valuing diversity within a system, it’s possible to create a more resilient and adaptable system that is better equipped to handle the challenges and uncertainties of the future.

Note: ChatGPT had a better summary.

Paradigm shifts are difficult

Meadows emphasizes that paradigm shifts are difficult to achieve within systems. Paradigm shifts involve a fundamental change in the way that a system is understood or approached, and can require significant changes in mindset, assumptions, and values. Meadows notes that paradigm shifts often encounter resistance from within the system, as the existing structures and feedback loops within the system can reinforce the existing paradigm and make it difficult to imagine or accept alternative approaches. However, Meadows also argues that paradigm shifts are essential for creating lasting change within systems, and that understanding and working with the feedback loops and leverage points within a system is key to achieving successful paradigm shifts.

Note: Again ChatGPT had a better summary.

Systems theorizing will not get us to solve all problems but will help us avoid some wrong turns

Meadows acknowledges that systems thinking is not a panacea, and that it is important to remain humble and open-minded when working within complex systems. She argues that by recognizing the limits of our knowledge and understanding, and by engaging in ongoing learning and experimentation, we can work towards creating more sustainable and resilient systems over time.

Systems theorizing is not a magic bullet. It will not solve all problems. It can help us to avoid some wrong turns, but it cannot guarantee success.

In the context of Thinking in Systems, Donella Meadows argues that systems theorizing is a valuable tool for solving complex problems. She writes that “systems thinking is a powerful tool for understanding and changing the world.” However, she also cautions that “systems thinking is not a panacea. It will not solve all of our problems.”

Meadows argues that systems theorizing is most useful for solving problems that are complex and interconnected. She writes that “systems thinking is especially useful for problems that are difficult to solve with traditional methods.”

Meadows also argues that systems theorizing is most useful for problems that require a long-term perspective. She writes that “systems thinking is essential for solving problems that require a long-term view.”

Note: The intro is by ChatGPT, the rest is from Bard. Bard uses more quotes. Reminds me of Ask Senaca – a model trained on the writings of Seneca that also uses a lot of quotes: futuretools.io/tools/ask-seneca

Here is what I got when I asked Ask Seneca: How to think in systems?

My friend, to think in systems, you must first understand the importance of proper timing and action. As I wrote, “It is one thing to understand the merits and the values of facts, another thing to know the precise moment for action, and still another to curb impulses and to proceed, instead of rushing, toward what is to be done” ¹. Additionally, it is crucial to apply what you learn to your conduct and to control your passions. “Talk in this way to other men, provided that while you talk you also listen; write in this way, provided that while you write you read, remembering that everything you hear or read, is to be applied to conduct, and to the alleviation of passion’s fury” ². Lastly, always be aware of the potential for human nature to produce negative outcomes. “Think of everything, expect everything: even with men of good character something queer will crop up” ³. By understanding these principles, you can begin to think in systems and approach situations with a clear and rational mind. Farewell.

Pretty cool right!

Time to wrap up. Meadows mentions using the scientific method so my last note to self is on that:

Applying the scientific method

The scientific method is a systematic approach to acquiring knowledge that involves formulating hypotheses, making predictions, and testing those predictions through observation and experimentation. It typically involves the following steps:

Observing and describing a phenomenon
Formulating a hypothesis to explain the phenomenon
Making predictions based on the hypothesis
Testing the predictions through experiments or further observations
Analyzing the results and modifying the hypothesis as needed
Repeating the process to refine and build upon existing knowledge.

The scientific method is a key tool used in the natural sciences, as well as in many other fields that rely on empirical evidence and experimentation to generate knowledge.

To recap

These are the bits and pieces that I have taken away from reading Thinking in Systems. Some are central to the book, and others come from other fields and are just mentioned. I think they help in framing and communicating ideas and can help me in my day-to-day.