Conversely, a system with only negative feedback relationships will die quickly. It is smothered in negativity and a lack of sustained energy. Some positive feedback relationships are needed in order to provide both direction and energy. It usually takes only a bit of negative feedback to keep a system operating in a reasonable, sustainable manner; however, quite a bit of positive feedback is usually needed to launch and sustain a system—especially if it is large and complex (think of the energy needed to launch a rocket heading into space).
There is one other feedback-related features that is critical to any full appreciation for the dynamic operation of complex adaptive systems. This feature is the Delay that occurs between the provision of the feedback and its receipt. Wars have been started and lost as a result of delay (in the transmission of information (as well as supplies).
At a more mundane level, we find that delays in the receipt of information regarding customer needs or even inventory can damage the operations of a retail store. Similarly, delays in feedback regarding rates of tax revenues can incapacitate a government’s capacity to plan for the initiation of specific human service programs. Even more important is the pattern or response to delays that can occur in a system. If there are major delays, then there are likely to be major swings (oscillations) in the size of a retail store’s inventory or a government’s quantity (and quality) of human services.
Some of those who study complex systems—adaptive and otherwise—believe that delays in a system account for more of its unique features than either the size or type (positive or negative) of feedback operating in this system. As an architect of system dynamics concepts, Donella Meadows, 2008, pp. 151-152) concluded that:
“Delays in feedback loops are critical determinants of system behavior. . . Delays that are too short cause overreaction, “chasing your tail” oscillations amplified by the jumpiness of the response. Delays that are too long cause damped, sustained, or exploding oscillations, depending on how much too long. Overlong delays in a system with a threshold, a danger point, a range past which irreversible damage can occur, cause overshoot and collapse.”
As Meadows has noted, there are systems in which delays are very short. The feedback is proximal (neighborly) with agents working closely together. Elements of the system are tightly coupled. Each element reacts immediately to the actions taken by a neighboring element. Such is the case with the flocking of birds. Tightly coupled agents (such as birds) work with minimal delays. They can be quite agile in their response to outside challenges—but can also be quite jumpy (“trigger-happy”) as Meadows observes. Other systems can operate with major delays in the flow of information and either positive or negative feedback. The agents operating in these systems are loosely coupled. The feedback is distal (foreign) and this system runs the risk of responding too late, with too little effort, and in the wrong direction to an impeding challenge.
It is important, in conclusion, to suggest that failure to recognize and appreciate the delay function operating in virtually all complex systems typically results in actions that yield results which are quite different from what was desired and anticipated. Often a set of “counter-intuitive” steps must be taken to produce the desired results. As Jay Forrester (the initial architect of system dynamics) has often proposed: “don’t just do something, stand there!” In other words, seek to carefully understand the dynamics operating in a complex system before trying to change it.
