I have spent an embarrassing fraction of my adult life playing Civilization, and I say this with love: the tech tree is the game's most seductive fiction. Not because the technologies are wrong, but because the shape is. In the game, knowledge is a ladder. Each technology sits in its box, waits for its prerequisites, and unlocks — discrete, ownable, permanent. Research is what a society does when it has spare capacity, and once a box is filled in, it stays filled in forever. [author assertion]
Real technology compounds in almost the opposite way. The printing press did not matter because a single inventor climbed to it; it mattered because paper, ink, urban literacy, cheap transport and postal routes were already lying around, waiting to be combined. Remove any two of those complements and the famous unlock fizzles into a curiosity. In my judgement, most technologies that look like steps are actually intersections — they happen when several slow, unglamorous lines of development cross, and they fail when the crossing is incomplete. [author assertion]
The tree also never forgets, and societies do. Techniques for making Roman concrete and Damascus steel were lost for centuries, not because people became stupid, but because the knowledge lived in practices, guilds and supply chains rather than in boxes. Knowledge that is not continuously re-enacted decays. A tech tree with no forgetting is a comforting model of progress; it is not a description of any civilisation I have studied. [author assertion]
What binds is rarely the idea. The same invention can sit idle in one society for a hundred years and transform another in a decade, and the difference is usually institutional: who may own the returns, who can borrow against them, which standards body blesses the approach, which guild is allowed to object. When I teach this, I tell my students that the scarce resource is seldom ingenuity. It is the surrounding tissue that lets ingenuity be used, financed, repaired and taught. [author assertion]
To be fair to the game, it gets two deep things right. The first is opportunity cost: research capacity is scarce, and choosing a line means not choosing another, with consequences that only become visible eras later. The second is path dependence: early choices genuinely do foreclose later ones, in the game and outside it. A civilisation that optimises for the next unlock and one that invests in complements are making different bets, and history grades both. [author assertion]
If I were redesigning the tree, I would build a web of complements with decay: technologies that fade if unused, revive if recombined, and unlock at different depths depending on what a society already practises. It would be a worse game and a better model. The tree teaches us to feel like strategists, and feelings are not nothing. But we should notice the myth it smuggles in — that knowledge accumulates like a savings account. In my experience it behaves more like an ecosystem: it compounds when the conditions are right, collapses when they are not, and is nobody's property for long. [author assertion]
