Self-organization is the capacity of a system to create new structures, behaviors, and entities — to reorganize itself in response to changing conditions. It is the property that makes complex systems more than the sum of their parts, and in Meadows's leverage-points hierarchy it ranks fourth from the top: more powerful than rules, goals, information flows, and feedback loop strengths, but less powerful than changing paradigms.
The reason self-organization ranks so high is that it is generative. A system that can self-organize can produce new feedback loops, new rules, new structures — it can create the system properties that lie below it in the leverage hierarchy. Evolution is the canonical example: through self-organization, life has generated the extraordinary diversity of organisms, behaviors, and ecosystems that exist today, none of which were designed. Markets self-organize to allocate resources. The immune system self-organizes to respond to novel threats. Scientific communities self-organize to accumulate and test knowledge.
Meadows drew a direct connection to resilience: a system capable of self-organization can regenerate, adapt, and recover in ways that a rigid system cannot. Self-organization is what allows a resilient system to respond to surprises — to restructure rather than merely return to a previous state. The loss of self-organization capacity (through extreme monoculture, through rigid bureaucratic rules that suppress adaptation, through monopoly that eliminates evolutionary pressure) therefore represents a loss of resilience.
The practical implication for management is uncomfortable: systems managed to eliminate unpredictability tend to suppress the self-organization that produces resilience and adaptation. A tightly controlled forest (fire suppression, monoculture planting, elimination of species diversity) loses its capacity to self-organize — and becomes vulnerable to the very disturbances that natural self-organization would have handled. A tightly managed economy suppresses the creative destruction that self-organization produces, and thereby loses the adaptiveness that makes economies capable of responding to new conditions.
Meadows did not romanticize self-organization. It can produce bad outcomes as well as good: the self-organization of criminal enterprises, the evolution of antibiotic-resistant bacteria, the emergent coordination of exploitative market structures. The value of self-organization depends on the conditions under which it operates and the goals it serves — which connects again to the paradigm and goal levels of her leverage-points hierarchy.
For the balaton-group and Meadows's sustainability work, self-organization was central to the concept of a sustainable society: one that preserves the conditions under which natural and social self-organization can operate, rather than optimizing those conditions away in pursuit of short-term output.