Against Optimization

For most of the big decisions we make—about how to govern our societies or how to structure our individual lives—there is a better, wiser strategy for us to follow. Topple the churches to the god of Optimization. Replace them with shrines to a wiser, more caring deity: Resilience.

To see why, we need to draw on lessons from unexpected places: the shells of molluscs, the carefully engineered robustness of ant colonies, and by debunking the mistaken interpretations of evolutionary biology that have infected the dominant—but incorrect—view as to how our world works.

The popular reduction of evolutionary principles to “survival of the fittest”—with overtones of relentless, flawless optimization—is a tragic mistake. (Many incorrectly attribute the phrase to Charles Darwin, but it was first coined by Herbert Spencer). While it is true that evolution does often fine-tune species to greater fitness over time through natural selection, the ultimate engine of evolution is survival and reproduction—which often requires robustness and the ability to adapt to uncertainty.

A hyper-optimized species that can only survive in one environment will get wiped out if that environment changes. That’s one reason why evolution routinely works in unexpected ways, through what the brilliant evolutionary biologist Zachary Blount calls “the genomic junk drawer.” The specific evolutionary path that a species took—along with plenty of accidental, contingent events along the way—leaves extra stuff in the genome that might at first appear to be junk.

The awe-inspiring genius of our natural world is that evolution provides a mechanism to repurpose that genomic “slack” into something more useful when the environment changes. It’s the evolutionary wizardry of resilient adaptation. That’s why, as Daniel Milo argues, a huge range of lasting species are defined not by optimal solutions, but by “good enough” ones. It’s not survival of the perfectly optimized, but survival of the resilient, as only the most robust inherit the Earth.

For example, nacre, or “mother of pearl,” is one of the oldest and most unchanged biomechanical structures on Earth. With a stunningly beautiful lustre, it gives pearls their sheen and adorns the inner shell of some molluscs. It is largely the same structure from when it first emerged roughly 530 million years ago. (Modern humans have been around for only about 250,000 years, so we might have something to learn from this longstanding byproduct of evolutionary pressure).

Nacre persists because nature is an engineering marvel, producing an ingenious structure that offers a parable for us. The short version is this: at the nano-level, the nacre on mollusc shells has a series of flawed, interconnecting parts that are decidedly un-optimized. The flaws lock together in an irregular brick and mortar pattern, where the “mortar” is organic material that, if needed, can be squeezed out when the material is put under strain.

Moreover, the unique structure creates discontinuities, so if one part cracks, the damage is contained, isolated, decoupled from the rest of the material. To an untrained eye, the structure looks woefully inefficient, wasteful, badly designed. Instead, it’s one of the strongest substances in the world.

This structure provides two initial key lessons for humans—both in our social systems and in our lives. Resilience can often be produced by systems that feature:

Diversity (lots of different kinds of components that work together are more robust than a uniform single structure, just as the Estonian power supply was augmented by a wide array of other electricity sources when one cable was severed);

Redundancy (systems that are designed to work even after an unexpected failure or setback are more robust, illustrated by the Suez Canal, which had no backup option when the route became blocked).

The third lesson comes not from molluscs but from ants. It’s resilience from what I call decoupled connectivity, the idea that robustness comes from interconnected support networks—but also that one needs to be able to sever a destructive node when it becomes toxic. Connectivity allows a system to repair itself when under strain, while decoupling allows isolation to contain a devastating cascade.3

When ant colonies face a disease outbreak, for example, they exhibit ingenious behaviors. If the outbreak is merely of a mild fungal infection, then connectivity saves the colony, as “nurse” ants are swiftly deployed to administer “a formic acid antimicrobial poison to their patients whilst grooming them.” Without the connected network, a mild outbreak could become a devastating epidemic.

However, if the outbreak does become more severe, maintaining connectivity could prove fatal. Then, the colony will pursue more extreme strategies, either of isolation—keeping infected ants away from the healthy ones—or of killing diseased individuals, pruning that node off from the colony altogether.4 Since outbreaks don’t happen all the time, keeping these mechanisms in place could be thought of as a form of inefficient slack. But when disease strikes, it’s the slack that saves the colony from existential risk.

One of nature’s overarching lessons is this: what may look to a naive human eye as waste, or inefficiency, or under-optimized slack is often evolution’s secret weapon, providing the adaptive resilience to survive in an ever-changing world.

- More Here