The shifting question: from control to emergence

Within this field of research, the concept of Complex Adaptive Systems (CAS) was developed. A CAS consists of multiple autonomous actors who make local decisions based on limited information, adapt to their environment and to one another, and collectively generate patterns that cannot be reduced to individual behavior [3].

An important characteristic of CAS is that the system learns. Actors adjust their behavior based on experience, causing the system as a whole to change. This implies that cause–effect relationships are unstable and that interventions themselves become part of the dynamics they seek to influence.

Emergence and non-linearity

Emergence is a core concept in complexity science. It refers to the emergence of system behavior that is not explicitly designed and cannot be explained by individual intentions alone [4]. Emergent behavior is often recognizable, but difficult to predict precisely.

Non-linearity means that small changes can have large effects, while large-scale interventions may sometimes have little impact. In societal systems, a small change in rules or incentives can lead to fundamentally different behavior, while extensive reforms are neutralized by adaptive responses [5].

From prediction to experimentation

The recognition of emergence and non-linearity leads to a different approach to design and governance. Instead of defining optimal solutions in advance, attention shifts toward experimentation, observation, and adjustment [6].

Within complexity science, agent-based models are used to explore possible dynamics. These models do not provide predictions, but serve as instruments for gaining insight into how local rules can give rise to global patterns.

Limitations of complexity thinking

Although the complexity approach explains much about dynamics and adaptation, it also has clear limitations. It offers little normative guidance for questions of legitimacy, justice, and power. The fact that a pattern emerges does not mean that it is desirable [7].

Without additional institutional and normative frameworks, appeals to complexity can lead to laissez-faire governance or indecision. Complexity thinking is therefore necessary, but not sufficient, for designing robust forms of collective action.

Implications for collective action

For collective action in complex systems, this means that governance cannot consist of fixing desired outcomes in advance. What is possible is the design of conditions under which desirable behavior becomes more likely and undesirable dynamics become visible in time.

This presupposes transparency, feedback, and room for adaptation. The complexity approach thus underpins the importance of learning systems and forms a second building block for the broader theoretical framework.

References

[1] Kauffman, S. (1993). The Origins of Order. Oxford University Press.

[2] Simon, H.A. (1962). The Architecture of Complexity. Proceedings of the American Philosophical Society.

[3] Holland, J.H. (1995). Hidden Order: How Adaptation Builds Complexity. Addison-Wesley.

[4] Mitchell, M. (2009). Complexity: A Guided Tour. Oxford University Press.

[5] Arthur, W.B. (1999). Complexity and the Economy. Science.

[6] Epstein, J.M. (2006). Generative Social Science. Princeton University Press.

[7] Pettit, P. (1997). Republicanism: A Theory of Freedom and Government. Oxford University Press.