Resilience is the ability of a system to survive, absorb, and recover from perturbation — to persist and continue functioning in the face of stress, surprise, and shock. It is distinct from productivity, efficiency, and stability, and Meadows argued that it is persistently undervalued in human management of both natural and social systems.
A resilient system is not necessarily a stable one. A forest is resilient if it can recover from fire; a resilient economy can absorb shocks and restructure; a resilient person can weather adversity and return to functioning. Resilience does not require staying the same — it requires retaining the capacity to reorganize, adapt, and persist. A forest fire may look like failure; if the ecosystem can regenerate through self-organization, the system is resilient.
The tradeoff Meadows identified is the central practical problem: resilience is often purchased at the cost of short-term productivity, efficiency, or stability, and therefore it is routinely sacrificed in systems managed for short-term performance. Monoculture agriculture maximizes yield per acre but loses the resilience that diversity provides. Just-in-time manufacturing eliminates inventory buffers and thereby eliminates the resilience those buffers provide. Financial systems optimized for return on capital reduce redundancy and thereby lose the resilience that redundancy provides.
This tradeoff makes resilience invisible until it is gone. Because resilient systems look inefficient (spare capacity, redundant pathways, diverse strategies, buffer stocks), the pressure to optimize tends to eliminate the features that provide resilience — and the cost only becomes visible during the shock that the now-optimized system cannot absorb.
Meadows connected resilience to the structure of feedback-loops. A resilient system has multiple feedback pathways — if one is disrupted, others compensate. It has buffer stocks that absorb shocks before they propagate through flows. It has self-organization capacity to create new structures when old ones fail. All of these are expensive in good times; all are invaluable in bad times.
The connection to leverage-points is direct: maintaining the resilience of a system is among the high-leverage structural interventions because resilience enables all the other responses to work. Meadows argued that sustainability-indicators should explicitly measure resilience — the capacity of agricultural, ecological, economic, and social systems to continue functioning — not just current output levels. This was part of the agenda of the balaton-group and her work at the sustainability-institute.
In limits-to-growth-1972 and its sequels, global resilience was the underlying concern: the world system's capacity to absorb the perturbations produced by exponential growth was being eroded by that very growth. The world was trading long-term resilience for short-term productivity in ways that would eventually produce overshoot-and-collapse.
Meadows was careful to note that resilience can serve bad ends as well as good: a system of oppression can be resilient. What matters is specifying resilience for whom and for what purpose — a question that connects to the paradigm and goal levels of her leverage-points hierarchy.