Systems Basics and History

Systems Thinking, Week 2

Terms

This week we got right into the basics of systems thinking, both conceptual and historical. Some terminology to start (this article from Daniel Kim is a good deeper dive): a bunch of elements thrown together (whether they are marbles, car parts, cells, buildings, or government agencies) constitute a collection. A collection becomes a system when the elements are organized in such a way that they work together to create behavior or outcomes that aren’t present in the individual elements but arise from their interactions (putting the car parts together properly allows you to start it and drive somewhere); that is, behavior is explained as a function of the structure of the system.

The implication is that the order or placement of the elements in a system matters a lot; in a classic example, if you take a cat apart into its constituent parts and then put it back together differently, you would find a much lower level of functioning (make it a bicycle if that’s too unpleasant an idea). That emergent behavior is sometimes termed the purpose of the system, though that strikes me as either too much anthropomorphizing or too much assumption of design; in addition, even when systems are created with an explicit purpose, the outcomes seldom align closely with that intent.

Since most systems in the real world¹ interact with their environments, they are open. It’s not a huge logical leap to argue that the boundaries we put around systems to define in vs out are arbitrary, or, less nihilistically, that the definition of system boundaries are a choice that the investigator must make instrumentally based on the problem being solved. A system exists as a component of larger systems, and the components of a system are usually systems in themselves, so picking the right level of abstraction is an art.

History

Systems pioneer Russell Ackoff has provided a solid high-level history of the intellectual development of systems science. Essentially, the Renaissance and Enlightenment were founded on the ideas that the universe was totally knowable by breaking down phenomena into their component parts and analyzing cause and effect. This analytical/reductive approach enabled a huge outpouring of knowledge and understanding for several centuries, and led to the Industrial Revolution (i.e. the mastery of machines represents humanity’s application of its understanding of the natural world to the betterment of the species), but it also had its limits and shortcomings. For one, the environment was frequently ignored as a constraint; for another, this paradigm had difficulty integrating advances like the radical uncertainty of particle physics.

Starting in the 1920s and continuing through today, the field of systems thinking (that is, thinking about the interactions and emergent behavior of systems rather than continuously breaking things down into smaller parts for analysis) has grown as a companion and corrective to the reductive approach, from cybernetics in the 1940s² and information theory in the 1950s to chaos theory in the 1980s and network science in recent years³⁴. The focus on the interplay between elements, the importance of the environment to outcomes, and the treatment of overall system behavior as the main focus of study serve as important correctives to the limitations of the prior paradigm. Per Ackoff, the socio-technological analogue to this, based on the observation, manipulation, and communication of symbolic information, is the Information Age, where we automate intelligence in the same way we mechanized work (from the earliest use of computers to speed up calculations to the AI applications we are seeing today).

Final Thoughts

One of the interesting things about the literature introducing systems thinking is the heavy use of duality to compare and contrast with the reductive approach:

• Analysis vs synthesis

• Knowledge vs understanding

• How vs why

• Part vs whole

The most important idea is to realize that systems thinking is never the right way to think about a problem, but it might be useful⁵. It adds a dimension to our perception of the world beyond what is offered by the analytical approach, and can help us understand more and more of the world around us.

1. Put a mental bookmark here - the discussion of whether systems actually exist in the real world or are just a useful way to think about the world is an interesting question I intend to talk about next week.

2. This field is the origin of cellular automata, showing how simple rules and elements can lead to complex outcomes.

3. If you think this sounds wildly biased toward Western history, I agree! However, it does a decent job of laying out the intellectual lineage of the modern age (and, for the record, Ackoff does cite the exposure to the Muslim world during the crusades as a major impetus for the Renaissance).

4. Also worth mentioning, this timeline is aligned with the development of Foresight as a field; they essentially grew up together as sister disciplines. In fact, there are a few very clear points of overlap: Kenneth Boulding, one of the founders of general systems theory, was married to Elise Boulding, the sociologist who learned Dutch so she could translate Fred Polak’s Image of the Future, the seminal futures work; Jay Forrester did the systems dynamics work for Limits to Growth and is claimed by both groups.

5. In this way, it connects directly to the American philosophical orientation of Pragmatism. For a delightful in-depth treatment of this, check out Louis Menand’s The Metaphysical Club.