Four Faces of Resilience
From rebound to sustained adaptability: four ways to understand resilience
Every so often I use this space to surface papers worth reading. My bar is simple: short enough to finish in one sitting, clear enough to sharpen your vocabulary, practical enough that you can try something new next week, and durable enough to matter outside its original domain. The best ones give you a horizon you keep reaching for.
This is one of those papers. In 2015, David Woods, a professor at Ohio State and a founder in the fields of cognitive systems engineering and resilience engineering, published a brief article that still feels fresh today. Woods has spent decades studying how complex systems succeed and fail, from nuclear plants and aviation accidents to healthcare and cyber-physical systems. His core question is straightforward: why do some systems fall apart under stress while others adapt and continue?
His answer starts by clearing up our language: “resilience” is not one thing, it is at least four.
The Four Senses
Rebound : resilience as bouncing back. Some systems recover faster after disruption, but the idea of a clean return to “normal” is misleading. Adapting changes the system.
Robustness : resilience as withstanding disturbance. Useful when disturbances are well modeled. Outside that envelope, robustness often hides brittleness.
Graceful extensibility : resilience as the opposite of brittleness. When surprises hit, some systems stretch capacity at the edges and improvise without collapsing.
Sustained adaptability : resilience as architecture. With the right network properties and governance, systems preserve the ability to adapt across repeated cycles of change. Think of evolvability over time, not just recovery in the moment.
The Real Stakes
The first two senses, rebound and robustness, gave us useful history. But they do not prepare us for what lies ahead. The future is in the latter two.
Graceful extensibility asks how to avoid brittle collapse when the unexpected arrives.
Sustained adaptability asks whether a system can keep adapting as pressures, contexts, and constraints shift over years or decades.
Both depend on how systems handle trade-offs. This is where Eric Hollnagel’s ETTO principle (Efficiency–Thoroughness Trade-Off) matters. Every system rocks back and forth between efficiency and thoroughness. Push too hard toward “faster, cheaper, leaner,” and you cut away the slack that makes adaptation possible when surprises arrive. Many of us experienced this very challenge when the COVID pandemic hit and disrupted supply chains. The effects of unexpected interruptions were exacerbated by tightly-defined control systems that couldn't "flex".
I make a similar point in my Conway’s Law at a Distance talk: as systems grow, the rate of change accelerates and expectations rise. Left unmanaged, those forces create intractability. Put that together with ETTO and the stakes are clear. Optimize too aggressively for today’s efficiency and you set up tomorrow’s brittleness.
In Woods’s framing, resilience is not recovery to previous equilibrium. It is the capacity to navigate surprise continuously, often resulting in a "new normal", while managing the trade-offs that shape complex systems.
Resilience is not snapping back or hardening up. It is designing for systems that can stretch without breaking and sustain their ability to adapt over time. That applies to ecosystems, software platforms, organizations, and economies alike.
If we want resilience that lasts, design not just for strength, but for graceful extensibility and sustained adaptability.
If this piece made you stop and think about where AI is taking design, you’ll love Signals. It’s where I explore how systems evolve and how we can evolve with them. Subscribe to join thousands of designers and developers building for change.