It is one of the quiet scandals of modern science that we still have no generally accepted theory of aging. We can name the exact pathogen behind an infection and the exact thermonuclear reactions warming the cores of distant stars, but ask a biologist what, precisely, aging is, and the room divides into a dozen camps, each with its own pet mechanism, its own molecule, and quite often its own startup. There should be a theory of aging. There is none. Our knowledge here looks less like a map with a few blank corners and more like a black hole.
I want to suggest that the trouble lies in the question itself. We have been studying the shadow and calling it the object — the hole, and calling it the donut. Suppose aging, as a single thing, does not exist at all. What exists instead is a crowd of resilience mechanisms: DNA repair, immunity, apoptosis, the renewal of tissue from stem cells, the stubborn mechanical integrity of the cellular matrix. “Aging” is simply the name we give to the moment when they all give way at roughly the same time — and that simultaneity, I will argue, is no accident. It is engineered, slowly and blindly, by natural selection.
The factory and the warranty period
Imagine a factory building a car designed to last 100,000 kilometers. A sensible engineer does not buy a transmission rated for 300,000 km and bolt it to brake pads that fail at 40,000. The expensive over-built part is wasted: the car will be in a scrapyard long before that transmission shows its quality. Equally, no part should be rated below the target, or the car dies young and the warranty bleeds money. So the rational move is to source every component with roughly the same service life — a hair above the design target, and no more.
Evolution is exactly this kind of frugal, indifferent purchasing department. If any one resilience system fails earlier than the lifespan an organism needs to reproduce, selection works furiously to extend it. And any system that lasts conspicuously l