In today’s post, I am looking at the idea of “sweeping-in” in Systems Approach. “Sweeping-in” can be described as the process of opening up the inquiry of a system by expanding its boundaries. Churchman discussed sweeping-in in several works, including “Thought and Wisdom” ^[1] and “The Design of Inquiring Systems” ^[2]. Churchman credited his teacher, E. A. Singer, for the concept of sweeping-in.

The “sweeping-in” process was introduced as a method for incorporating diverse concepts and variables from various sciences to resolve inconsistencies in measurements or observations. Churchman wrote:

the problems we humans face are so closely interconnected so that the only way we can study a system is to recognize the need to be comprehensive. ^[1]

there are no simple questions and the process of addressing a specific question will eventually require answers to more and more questions, i.e., require the “sweep-in” process. ^[1]

The sweeping-in process consists of bringing concepts and variables… into the model to overcome inconsistencies… ^[2]

In Systems Approach, sweeping-in requires us to expand our inquiry to incorporate a wide range of perspectives and variables. It demands that we examine the larger system and understand the ethical implications of our approach. This is a continual process that necessitates a cross-disciplinary approach. When addressing a situation, we must bring in knowledge and perspectives from multiple stakeholders and look at broader contexts. This means looking beyond the immediate problem to understand the larger systems and contexts in which our system exists.

Singer argued against the idea of simple and directly knowable facts from observation. He thought that there are no simple facts of nature that we can know directly, and that even seemingly simple observations are actually complex. In this regard, when we set out to find an answer to any question of fact, we realize that we must learn more and more about the situation. The original question becomes increasingly complicated, not simpler. Singer advocated not trying to reduce observations to simple elements, but instead following a sweeping-in process where our inquiry expands to include more context and interconnected systems.

The sweeping-in process is anti-reductionistic. Churchman explained this when he wrote about the strategies of inquiry ^[2]:

Which is better, to reduce the system to its elements or to expand the system? A system-science reply would be that since there are no simple, elementary questions, the first strategy is based on illusion and the second is the one to be followed.

The sweeping-in process requires us to embrace the complexity of the situation at hand. This demands epistemic humility. Reality is already complex, which means that our initial framing of the situation is often too narrow, resulting in premature solutions that are not effective and may cause more harm than good in the long run. We may ignore important interactions and relationships, leading to unintended consequences.

Sweeping-in involves examining our current system from the perspective of the larger systems it is part of. This is one of the basic ideas in systems approach – to understand the function of a part, we must look at it from the standpoint of the larger whole. There is a hint of Godelian thinking here. A great example from Russell Ackoff, a renowned Systems Thinker and student and friend of Churchman, is that of the automobile. No matter how much we understand an automobile and its parts, we will never understand why we drive on the right side of the road in the U.S. unless we consider the larger context—the historical, social, and cultural norms that shape American driving practices.

The reader might now wonder about the use of cybernetics in the title of the post. Churchman wrote that sweeping-in is a process of adding in and adjusting the results to improve our understanding of a problem ^[1]. This is a means to perform error correction in our understanding. This will be a never-ending process since we lack the variety to completely understand the external world.

Sweeping-in cautions against over-simplification. This does not mean that we need to make a situation artificially more complex for the sake of it. As I mentioned before, reality is already complex. We need to acknowledge our limitations and account for enough perspectives and variety to match the variety of the situation at hand. In Cybernetics, complexity is explained via variety. To achieve a requisite understanding of the situation, we need to have requisite variety. One of the most important ideas in cybernetics is Ross Ashby’s law of requisite variety. I welcome the reader to explore this further here.

The complexity that we are “adding” through sweeping-in is not arbitrary. We are attempting to include aspects that are needed but might not have been considered in the initial framing. This could include perspectives from other stakeholders, longer-term consequences, ethical considerations, or the influence of broader contexts such as social, political, or environmental factors.

Our basic instinct is to simplify when faced with situations that seem complex. This process is known as attenuating external variety in cybernetics. While simplification can effectively achieve requisite variety, excessive attenuation signals ignorance, which in cybernetics is referred to as the “lethal attenuator.” Our attempts to simplify can often create blind spots, causing us to overlook less obvious but influential factors. Therefore, sweeping-in serves as a reminder to deliberately resist oversimplification.

Having epistemic humility and being aware of our cognitive blind spots are important notions in second-order cybernetics. Second-order cybernetics reminds us that any system’s functioning includes the observer and their interactions with the system. Here, the feedback loops include the observer as a participant, influencing the dynamics and adding new layers of complexity to the situation. This recursive process highlights the interdependence of the system and the observer, making it illogical to separate the two.

This reflexive approach means that reality is constructed on an ongoing basis through the interaction between the observer and the system. Most importantly, this approach incorporates ethics, one of the key points of Systems Approach, by recognizing that the observer’s involvement in a system carries responsibility. Since observers influence systems and construct reality through their interactions, they must be aware of the consequences of their actions. This promotes a constructivist view, where knowledge and reality are not discovered as objective facts but are constructed through interaction in a social realm. Observers are responsible for the realities they help construct. This practical aspect challenges the implications of relativism. While multiple perspectives may exist, the ethical responsibility of observers grounds our understanding of “truth” and “reality”, emphasizing that our participation in systems has meaningful consequences.

Churchman used the examples of a prison and a hospital to explain the ethical considerations further^[1]:

The planner should search not for ways to make the prison or the hospital run more smoothly, but for the reasons why we have things like badly run prisons and hospitals. The reasons turn out to be political, as much as economic; hence, the planner needs to “sweep-in” the causes of the existence of the troubled organization, and these causes like in other systems.

Another notion in sweeping-in is the need for challenging assumptions. Here we should ask questions such as WHO defines the system, WHOSE perspectives are included or excluded, and WHAT ethical considerations should be taken into account etc. The path forward, as advised by Churchman, is to utilize idealistic thinking. We must look at what an ideal solution would look like, not just accepting the current “realities”.

There are no final solutions in this approach, only provisional solutions. There is only continuous feedback and adaptation. This is also an important aspect of second-order cybernetics. The emphasis is on “less wrong” solutions rather than correct solutions. Each action taken informs the next round of understanding and action. Thus, the emphasis is on improving our understanding, or “understanding understanding”, another notion in second-order cybernetics.

Churchman was a pragmatist. From this perspective, the practical payoff comes from improving the depth and quality of decision making by acknowledging our limitations and inherent complexity of the situation. The goal is better informed action. I will finish with a great passage from Churchman that shows his true pragmatist spirit ^[2]:

When all is going well, and data and hypothesis are mutually compatible, then is the time to rock the boat, upset the apple cart, encourage revolution and dissent. Professors with well-established theories should encourage their students to attack them with equally plausible counter-theories. This is the only pathway to reality: whenever we are confident that we have grasped reality, then begins the new adventure to reveal our illusion and put us back again in the black forest.

But the process is dialectical, which means that two opposing processes are at work… One is the process of defending the status quo, the existing “paradigm” of inquiry, with its established methods, data, and theory. The other is the process of attacking the status quo, proposing radical but forceful paradigms, questioning the quality of the status quo.

Singer… called the “real” an “ideal” and we can see why. The idealist is a restless fellow who sees evil in complacency; he regards the realist as a hypocrite at times because his realism is unrealistic. The realist, on the other hand, accuses the idealist of being impractical, because his insistence on destroying the value of the present way of life precludes positive action. The Singerian inquiring system does not seek to resolve the philosophical dispute, but, on the contrary seeks to intensify it.

Always keep on learning.

[1] Thought and Wisdom, C. West Churchman (1982)

[2] The design of inquiring systems, C. West Churchman (1971)