112

­

­

Why are such systems called chaotic? The reason is, they are only predictable for a

short time. This is because biological systems, but also the weather and chaotic systems in

general, are not controlled linearly. That is, a small change in control, just like a small error

in description, doubles with each time step. For example, if I have only 1 per mil error in

the description, just ten time steps later I have more than 100% error and can no longer

predict the system state. The time scale on which this no longer describability happens

varies among systems and is a characteristic time. However, the result is the same for all

chaotic systems. Even for relatively short periods of time, one no longer knows what their

concrete state is, since one never knows the starting state with infinite precision, and small

errors always build up exponentially (the definition of a chaotic system). On the other

hand, controlling such a system is very effective (the so-called butterfly effect, since even

the smallest changes are always amplified exponentially). Finally, we now also know that

the result space of the system sets clear bounds, as does the climate of a place. Even if I

can’t predict the system in the short term, I can predict what the system will stay within in

the long term based on the attractor. For the same reason, stronger disturbances of the

system are very dangerous. Then it can happen that the system not only gets out of bal­

ance, but permanently leaves its previous system state and changes into a new “sick” state

(crossing a “tipping point”, see Chap. 16).

9.4

System Credentials: Emergence, Modular Construction,

Positive and Negative Signal Return Loops

­

9  Complex Systems Behave Fundamentally in a Similar Way