THE IDEA
The dance that never ends
Nothing evolves alone. Every living thing, every organisation, every technology exists in relation to other things that are also changing. When those changes are responses to each other - when each system’s evolution alters the environment the other is adapting to - that’s co-evolution.
The idea comes from biology. Flowers evolved to attract pollinators. Pollinators evolved to exploit flowers. Each adaptation by one party changed the selection pressure on the other. Over time, they shaped each other into forms that only make sense in the context of the relationship. A bee’s body is a map of the flowers it evolved alongside. A flower’s colour is a message written for a bee’s eyes.
But co-evolution applies far beyond biology. Regulation and innovation co-evolve - new rules create new incentives, which produce new behaviours, which demand new rules. Attackers and defenders co-evolve - in cybersecurity, in immune systems, in military strategy. Companies and customers co-evolve - products shape preferences, and preferences shape products. The critical insight is that neither party is evolving toward some fixed optimal state. They’re evolving toward each other, and the destination keeps moving because the other side keeps changing too.
IN PRACTICE
Adaptation that begets adaptation
Spam filters and spammers have been co-evolving for decades. Early filters caught emails with obvious markers - suspicious subject lines, known spam phrases. Spammers adapted: misspelled words, image-based text, compromised legitimate accounts. Filters adapted: machine learning, content analysis, sender reputation scoring. Spammers adapted again. Neither side is winning permanently because neither side is standing still. Each improvement by one becomes the selection pressure for the next adaptation by the other.
A parent and a teenager co-evolve their communication strategies. The parent sets boundaries. The teenager tests them. The parent adjusts - tighter in some areas, looser in others. The teenager recalibrates - more honest about some things, more secretive about others. Each adapts to the other’s last move. If the relationship is healthy, the co-evolution moves toward trust and autonomy. If it’s not, it can spiral into surveillance and evasion. The direction depends on the feedback, not on either party’s intentions alone.
A city’s transport system and its land use co-evolve over decades. Build a motorway and suburbs sprawl toward it. Suburbs generate traffic that demands more road capacity. More road capacity enables more distant suburbs. Meanwhile, public transport loses riders to cars, services are cut, and car dependency deepens. The transport system shaped the city, and the city reshaped what the transport system needed to be. Decades later, the pattern feels inevitable - but it was co-evolved, one adaptation at a time.
WORKING WITH THIS
Seeing the dance, not just the dancers
When you’re trying to understand why a system looks the way it does, ask: what has it been co-evolving with? The shape of any system is partly a record of what it’s been adapting to. An organisation’s processes are a fossil record of the problems it’s faced and the partners it’s worked with. A relationship’s dynamics are the accumulated result of two people adjusting to each other.
When you’re trying to change a system, remember that the other systems around it will adapt in response. This is why so many changes that look good in isolation fail in context. You changed one side of a co-evolutionary relationship and assumed the other side would stay put. It didn’t. It adapted.
The practical move is to widen your view. Don’t just ask “what will happen if we change X?” Ask “what will the systems around X do in response, and what will we need to do after that?” Plan for at least two rounds of adaptation, not just one. And look for co-evolutionary relationships that are working well - where each party’s adaptation is making the other better. Those are the relationships to protect and invest in.
THE INSIGHT
The line to remember
Nothing evolves in isolation. The shape of any system is a record of everything it’s been adapting to - and they’ve been adapting right back.
RECOGNITION
When this is in play
You’re seeing co-evolution when two systems seem locked in a pattern of mutual adaptation - each change by one provoking a change by the other. When a problem keeps “coming back” in a slightly different form after each attempt to fix it. When a regulation produces exactly the behaviour it was designed to prevent, because the regulated parties adapted. When two people, teams, or organisations seem to have shaped each other over time in ways neither intended. When the question “how did it get like this?” can only be answered by telling the story of two things changing together.