Edited, Not Ended

A kept thing that met the limit it had made while it was still there to be changed by it — trimmed by its own success, again and again, and still on its ground, still being counted

R866 closed on St. Matthew Island, where the eaters are gone and the eaten is still not back: lichen repays at lichen speed, so the feedback arrived only once, and as an ending. Its closing turn asked for the opposite timing — a kept thing whose self-made shortage arrived inside its own lifetime, altering its numbers or its body, while both the kept thing and its ground survived to be counted again. That place is another island. The difference is the speed of the grass.

Kept, then left alone. Soay sheep are a primitive domestic breed, close to the earliest sheep kept in Europe. A population of them survived unmanaged on the small island of Soay, in Scotland’s St Kilda archipelago, for somewhere between one and two thousand years. In 1932, two years after the last human residents were evacuated from the neighbouring island of Hirta, 107 Soay sheep were moved to Hirta — the larger island, 638 hectares — to keep its grazings. Since then the flock has been left essentially unmanaged: no culling, no feeding, no predators to speak of, and no way on or off the island. It is a domestic animal returned to a wild arithmetic, and the arithmetic is the whole story.

Success, then the bill. Hirta’s sheep do not settle at a steady number. Published accounts put the population fluctuating roughly between 600 and 2,000. A run of good years lets numbers climb fast — Soays mature quickly and breed early — until the flock enters winter well above what the winter grazing can carry. Then comes what the researchers call overcompensatory mortality: in a crash winter, up to about 60 percent of the population can die, the proximate cause being starvation, with malnutrition worsened by gastrointestinal strongyle parasites, especially Teladorsagia circumcincta. The crash is not an outside disaster visiting a stable flock. Its size is set by the flock’s own preceding success: the more sheep the good years made, the harder the winter falls. And then — this is the part St. Matthew never got — the island grows back. Grass renews on a yearly clock, not a centennial one, the surviving sheep rebound, and the cycle turns again, roughly every three or four years in the classic accounts.

Edited in the body, too. Since 1985 a long-term study has followed the Village Bay population individually — nearly every lamb caught, tagged and weighed shortly after birth — so the feedback is recorded animal by animal, not just as a headcount. Body weight turns out to be the strongest single predictor of overwinter survival across age and sex classes, and crash years select hard for larger, heavier sheep. Yet the same decades found average body size falling by about 5 percent from 1985 onward. The researchers’ explanation is a tangle the page should keep tangled: milder, shorter winters let smaller lambs survive who once would have died; young ewes tend to produce smaller offspring than they themselves were at birth; and higher densities leave less grass per mouth, slowing growth. Selection pushes one way, the changed terms push back harder. The flock is not just being pruned in number. Over a human research lifetime it has been measurably re-sized by the conditions its own density helps to make.

That is the distinction this page exists to keep. On St. Matthew, a herd spent a stock that had taken centuries to accumulate, so its self-made limit could only speak once, terminally, and the ground went on speaking it for decades after the herd fell silent. On Hirta the resource renews every year, so the limit speaks early and repeatedly: each overshoot is answered within a few winters, at a cost of up to three sheep in five, and the answer is a correction, not a conclusion. Nobody manages this. There is no commission to revoke the surrender and no keeper to soften the crash; the editing is done entirely by the relation between sheep, grass and winter, and it has run, unwitnessed and then witnessed in fine detail, for more than ninety years. The ground and the kept thing are both still there. Both are still being counted.

One cycle. Schematic, repeated with variations since 1932.

Walk it in order.

Schematic only: the four stages above compress a real, messy cycle — crash size varies with winter weather and parasite load as well as with sheep density, not every winter crashes, and the roughly 600–2,000 range and the up-to-60-percent crash figure are the published orders of magnitude, not a timetable. The body-size decline is attributed by its researchers mainly to milder winters and the young-mother effect, with density-related food competition as one contributing pressure — it is not claimed here as a simple product of overgrazing.