Systems thinking is best described as a synthesis between reductionism and emergence. Reductionism is the practice of analyzing and describing complex phenomenon on a simpler or more fundamental level while emergence is looking at the same phenomenon and describing it in the context of the whole. If you were to view systems thinking as a spectrum, at one end would be reductionism and then emergence would be at the other end. Both have their advantages and disadvantages, but to have the best description of the situation one should embrace the synthesis of the two, which is the essence of what it means to think in systems. By thinking about the overall system at once and how everything fits together, you are more likely to account for nuanced interactions between components that could lead to negative, unintended consequences if left unaccounted for.

Translating this description to decision making, systems thinking allows you to account for more nuances than either reductionism or emergence alone can account for, which then allows you to make better decisions. In general, the more quality information available to you about the situation, the better the decision(s) that can be made. Moreover, an effective systems thinker must be able to think gray. That is, to take take a step back and analyze the entire spectrum between emergence (white) and reductionism (black) and determine exactly where in the gray the optimal decision lies.

The Systems Iceberg

Systems thinking can be visualized in a broader context with the following visual known as “the iceberg.” This map has three levels, which are: events, patterns, and systemic structures [1]. Visualizing a system in this manner is referred to as Systems Mapping. Now, it's important to acknowledge that there are multiple ways in which a system can be visualized from analog cluster mapping to diagrams known as connected circles. All of these maps serve the purpose of visualizing the system in order to gain deeper insights into the interrelationships between the elements of the system. However, we will only address the iceberg model as well as causal loop diagrams within this article.

Events are the circumstances encountered on a day-to-day basis, such as your car breaking down, losing a customer due to poor customer service, a graduation in the family, etc. Patterns are a cognitive ledger of past events and when collated over time, can reveal recurring trends. For example, the stock market response to a bubble, the phases of a relationship, or the demand for cod by consumers during lent in the United States. The bottom of the pyramid is supported by systemic structures, which are the way in which the parts of a system are organized. These structures are the generators for the patterns and events that we observe. In the above example about losing a customer to poor customer service, perhaps these losses tend to only occur during specific times throughout the day. Why is this? Perhaps it's because of a shift change, lack in managerial oversight, or just overall poor training. Maybe a combination of events.

Note, systemic structures can be physical (e.g., workplace layout, supply chain morphology, etc.) or intangible, such as the organization of the workplace hierarchy or incentive structure. Further notice that we tend to live an event-oriented world. That is, we have a proclivity for noticing events over patterns and systemic structures even though it is the systemic structures positioned at the base of the pyramid that are driving these events. This makes sense as, in general, we are instinctively tilted towards responding to the immediate (i.e., immediate threats, dangers, concerns, etc.) than we are for long-term thinking. It is this long-term thinking or thinking at the systemic level that allows for optimal decision making.

Feedback and Thinking in Loops

Feedback is the transmission and return of information, which stands in contrast to linear cause-and-effect. That is, the linear view sees the world as series of cause-and-effect relations as demonstrated in the following image:

Here, A causes B, which causes C, which causes D, etc. Conversely, the feedback view takes the perspective that the world is a series of interconnected relations, where something affects something else, which then affects something else, etc:

Here, you have the linear cause-and-effect relation as before, but now we have the higher-order interconnections of D affecting C, which then affects A, and D also affecting B, etc. As you can see, the system is far more complex than a linear relation can fully capture.

To demonstrate the power of embracing a feedback versus linear view, let's consider the following example of a decline in sales (event A). This event is then responded to with a promotional campaign (event B), which is then followed by orders increasing (event C). This increase in sales is then followed by an increase in orders which need to be wait-listed (event D) as supply fails to meet demand. Then, due to the falling demand again (event F), a promotional campaign is again initiated (event G) in order to increase sales, etc. Linear thinking has lead to framing event B and event G as separate events even though they are clearly repeating events.

On the other hand, if we were to view this same example from a feedback perspective, the question “How do the consequences of my actions feed back to the system?” would be continually asked throughout the process. Thus, when sales go down (event A), a promotional campaign is launched (event B), orders increase (event C), sales then rise (change in event A), the number of wait-listed individuals increase (event D, which is an effect of event B), which affects orders and sales (change in event A and C), which leads to a repeat of event B.

But why is this important? The key insights gleaned from describing a situation in terms of systems versus linear thinking is that systems thinking forces you to describe the steps taken to get to where you currently are, which in turn then influences your actions moving forward. Referencing the most recent example, systems thinking in this scenario draws your attention to the inextricable relationships between events whereas you are more drawn to each cause-and-effect event pair with linear thinking. By being aware of the entire system and all the interrelationships involved in the problem, you're in a better position than if you were to only have noticed the individual cause-and-effect pairs.

Clearly, systems thinking helps us to better understand the complex behavior of the system, which then leads to better overall decision making for various scenarios that involve the system of interest. Pithily, the linear view describes what happens and when, while providing very little insight into how things happen and why, which is why it is inferior when juxtaposed to systems thinking.

Furthermore, drawing what are known as causal loop diagrams when analyzing a complex system can help to visualize all the interrelationships between variables. Once more, let's return to the example of a decline in sales. When drawn in a system loop, the behavior of events that unfold can be visualized in the following way [2]:

The exact same scenario is being described in this diagram as was before, but you get a better sense for how everything is interrelated as well as the repetitive nature of the system in particular. Now, in general, there are two different types of loops. There are reinforcing and balancing loops; each type describes how the system evolves in time. Reinforcing loops occur when the system reinforces more of the same leading to one element to dominate as time goes on. Conversely, balancing loops occur when the elements of the system balance out over time and you don’t end up with the scenario where one element dominates the other. The diagram here is an example of a balancing loop as no element begins to dominate over time.

Science and Systems Thinking - Complex Systems Science

The world, when you really take a moment to ponder it, is an incredibly complex place as we live in this system where everything is interconnected. However, due to cognitive limitations and ease, most of us have a tendency to only want to look at one part at a time. That is, we want to focus on one aspect and zoom in to learn the finer details. This is an ostensibly good strategy and is the basis of how science works, but if we fail to then zoom back out to see how this information fits into the bigger picture, our worldview isn't as accurate as it could be.

In regards to science, reductionism has been the dominant philosophy since its inception and has worked very well. In many instances, it continues to work very well to this day. However, scientists realize the value in systems thinking as there is information to be gleaned by thinking this way. For example, there are a number of interdisciplinary fields within science (e.g., systems biology, bioinformatics, ecological economics, behavioral neuroscience, etc.) that are using multiple fields of study to approach existing as well as novel problems. Due to the complexity of these problems, existing fields of study were inadequate (i.e., the thinking offered was too reductionist to solve the problem) to fully describe the dynamics of the system and a more integrated approach was developed. Complex systems science has emerged over the last 50 years as its own discipline within science that has its targets aimed at solving some of the most challenging problems that we face this century.

For example, consider the following diagram:

Here, the traditional reductionist disciplines within science are displayed vertically. These disciplines focus on understanding problems specific to their field. This is contrasted with the methods of complex systems, which cuts across these disciplines in an attempt to integrate their knowledge [3].

Another diagram to consider:

Here, each column contains three examples of systems consisting of the same components (from left to right: molecules, cells, people) but with different relations between them. Each row contains systems in which the relationship between the components is the same. For random systems, the behavior of each component is independent from the behavior of all other components. For coherent systems, all components exhibit the same behavior. Correlated systems lie between these two extremes, such that the behaviors of these system's components do depend on each another, but not so strongly that every component acts in the same way [4]. As you can see, complex systems science is using many scientific disciplines to approach problems in a unique way.

Key Concepts Summarized

Interconnections: Simply put, everything is connected. In systems thinking, this is a fundamental principle of life as the world is a dynamic, chaotic, inextricably connected place full of interwoven relationships and feedback loops.

Synthesis: In contrast to analysis, which is the paradigm of the reductionist worldview, the objective in systems thinking is synthesis. It is about understanding the entire system at once and finding the balance between reductionism and holism. Synthesis is the ability to see all of the interconnections.

Emergence: Essentially, emergence is the end product of synergy between the parts of the system. Through non-linear, chaotic interactions, a complex system emerges from the noise. Biologically speaking, emergence is the paradigm of life. That is, you start with the fundamental building blocks for life and through chaotic interactions over large time periods, the animate arises from the inanimate. This further continues as life forms then evolve through natural selection as they interact with one another as well as each other, which, eventually, gives rise to even more complex multi-cellular life who are bound by the same universal rules of natural selection that their constituent cells are.

Feedback Loops: With the countless interactions that infuse reality, there are constant feedback loops and flows between aspects of a system. Once the type of loop is identified, we can then begin to understand its dynamics and adjust the loop accordingly.

There are two primary types of feedback loops: reinforcing and balancing. Reinforcing loops occur when aspects of the system reinforce more of the same, such as global warming or compound interest. With these types of loops, an abundance of one element, which is continually growing and refining itself, can lead to it completely taking over. In practice, this can be good or bad. For example, consider global warming. As green house gases, C02 in particular, warm the planet this causes more green house gas emission from natural sources, such as the thawing of permafrost. Subsequently, as the Earth continues to warm, it will do so at an increasing rate as the warming causes the release of more green house gases, which begets more warming, and so on. Conversely, a positive feedback loop in your retirement account is a very good thing.

The other type of feedback loop is known as a balancing loop. As the name implies, this type of loop is where the components of the system balance out and you don't run into the situation where one component eventually dominates the system. The predator-prey model is a classic example of a dynamic system where there's a balance between the two competing species: one predator and one prey*. If the population of the predator gets too big, all the prey will get eaten and then the predator will no longer have any food and drive itself to extinction. If the population of the prey gets too large due to a lack of predators, the prey will exhaust their own food supply and subsequently experience a dramatic population decline or possibly even go extinct. As you can see, a balanced loop must be formed between the two species so as not to create a reinforcing loop favoring one animal or the other.

*This is clearly overly-simplistic as there are, in general, multiple predators for a prey animal as well as various types of prey that predators can choose from in an ecosystem. However, for clarity, we consider a system where there is only one predator and one prey.

Causality: In systems thinking, causality is being able to decipher how elements of the system influence one another. This leads to a deeper understanding of the feedback loops and allows perspective on the relationships between individual elements, which is fundamental to fully grasp the system dynamics.

Systems Mapping: There are many ways that a system can be mapped. From the iceberg model to interconnected circle maps, there are many different ways that systems can be mapped. Regardless of which map is chosen, the impetus for doing so is to visualize the connections between the elements of the system. By visualizing, you can better understand how elements are connected and interact with one another, which leads to insights and discoveries that can be used to develop new strategies or interventions. To recapitulate, visualizing the system leads to deeper insights into the interactions between elements, which leads to a better understanding of the system overall and, ultimately, the ability to make better decisions [5].

Conclusion

The world is a complex place. Everything is interwoven to a degree that it overwhelms and begs the question as to how it is that we actually know anything at all. Systems thinking is a mental model that helps to make sense of the complexity by encouraging a balance between reductionism and emergence. Further, while the ethos of reductionism has dominated science in recent history, this century will be dominated by fields where reductionism has been unable to provide all the answers (e.g., biology) by thinking in systems. While I don't readily prognosticate, it is my view that complex systems science will dominate the category of prominent scientific discoveries this century. By thinking about interconnections, synthesis, emergence, feedback loops, causality, and visualizing the complexity through systems mapping, deeper insights can be extracted about the system that could not be obtained through linear thinking. In the end, systems thinking leads to better decisions, which leads to better outcomes, and hopefully a better life.

References

[1] Klotz, Stefan. The Future of China's Software Outsourcing Industry - A Choice of Region to Source from. (2004).
[2] Kim, Daniel H. Introduction to Systems Thinking. Pegasus Communications Inc. 1999.
[3] Nigel Gilbert, Klaus G. Troitzsch, 2005, Simulation for the Social Scientist, Open University Press, MacGraw-Hill, (Maidenhead), ISBN 13 978 0335 21600 0
[4] Siegnefield, Alexander F., Bar-Yam, Yaneer. An Introduction to Complex Systems Science and its Applications. ArXiv. 2019.
[5] Acaroglu, Leyla. Tools for Systems Thinkers: The 6 Fundamental Concepts of Systems Thinking. Medium. September 7, 2017.

Systems Thinking: A Balance Between Reductionism and Emergence