Avalanche: The power of the feedback loop

In this post, I want to get across the importance of feedback loops. In particular, that a feedback loop can cause the opposite reaction to the expected one.

Continuing to explore my many stories about complex systems. See this first post for an introduction

Looking at Complex systems, expect to see some aspects of the system that don’t have simple explanations. For those aspects, there is no one simple story that is the One Truth; the story that makes all other stories false.

In this post, I want to get across the importance of feedback loops. In particular, that a feedback loop can cause the opposite reaction to the expected one. A surprise. A contradiction. Or sometimes a bit of both, some progress towards the goal you expect, and some movement away. In a simple world, logically opposite stories can’t both be true. In complex systems, both those stories can have some truth in them.

Much of the world is straightforward. If you do A then B happens,, then C and so on. A rigid fall of dominoes, with a visible end result. Such systems satisfy well our need for a definite answer, a predictable result. The apex of such systems is the digital computer. I boot up my Mac, and a billion operations later, the expected home screen appears. A technological miracle of logical predictability.

Some of the world is complex. Annoyingly or delightfully, depending on your viewpoint, such systems defy our thirst for the definite and the predictable. A straightforward system will go forward when you push it forward. The response of a complex one can be the opposite of what you intended. An avalanche of unexpected consequences.

The driver of such behavior is the dynamic feedback loop. Component A affects the behavior of Component B; but B is also linked in some way back to A. In a dynamic loop, B can affect A before it has time to settle into a new and definite state; then A changes B again before it has time to settle. It is this lively dance of connection that defeats a simple analysis. A tangled web of dynamic interaction.

This complex interaction can produce results that in a simple system would cancel each other. In complex systems, there is often some truth in the obvious story, and some in its opposite. For example, creating a new tool can have both a positive and a negative effect at the same time. The new AI tools are a current example. Much energy is expended on deciding if they are good or bad. I think the realistic perspective is say both stories have some truth in them. New tools empower and disable us at the same time.

Forcing one story to be the One True Story that makes all others false is out of touch with the reality of complex systems. Not only that, it promotes unneeded conflict. Claiming your story is the one true story makes all who disagree with you wrong. But they know, reasonably, that their story has some truth in it too. And not surprisingly get angry to be told they are completely wrong. Pointing out the opposite story has some truth in it has a better chance of being heard.

Declaring your opponents story has no truth in it creates a religious conflict about True or Not True. Recognizing that your opponents story has some truth in it turns that into a disagreement on balance and viewpoint. You may still come to opposite conclusions on the best policy. You may even violently disagree on policy. But you are less likely to demonize your opponent as driven by malice not by viewpoint.

The Economy is a good example of a complex system. Rarely is one story about the Economy agreed to be the true story. Economic policy is full of opposing arguments held strongly. Deficit spending vs Austerity, Free Trade vs Protectionism, Monetary easing vs tightening, Market vs Planned Economies.

Many economists will concede their opponents have some truth in their story. Some economic debates however, like Market vs Planned, can become a religious war. The Market economy may be declared the true path, and the Planned economy the false path. I see that as an example over simplification. In my view, the American economy is a mixture of Planned and Market competition. Most of the planned economy happens within big corporations, so is just less visible.

Recognizing that complex systems have both positive and negative effects implies taking responsibility for both the good and bad effects of your actions. It is more comforting to claim the negative effects are due to “bad people”. Some truth in that, but I believe much of the trouble in the world is caused by unintended consequences. I live a richer life due to all the energy and materials I use, but that is at some expense to the biosphere I live in. I should take responsibility for that.

Complex systems can, unlike simple systems, go in opposite directions at once. This alarming behavior is a consequence of the feedback loops in system. It is not due to a “malfunction”. Once I learnt this, life became a lot easier. Trying to “fix this malfunction” rather than adapt to this feature of reality saved me a lot of frustrated effort. I hope this viewpoint does the same for you.