selfish herd

The selfish herd is a model of animal aggregation in which individuals gather not to serve the group, but to reduce their own chances of being taken by a predator. Developed by W. D. Hamilton, it turns the familiar protective appearance of a herd into an account of self-interested positioning: the group forms because each animal seeks the relative safety of other animals around it.

The model and its premise

Hamilton’s model begins with a predator that tends to attack the nearest prey individual, a strategy that reduces the predator’s energy expenditure. If prey can detect a predator at a distance, they can flee; if it appears suddenly from cover, each animal may still lower its own risk by changing position within the aggregation. The model asks how such movements could produce a herd without assuming that any individual acts for the common good.

The central idea is a “domain of danger”: the area of ground nearer to one prey animal than to any other. If animals are spaced in a regular formation, an individual away from the edge might have a roughly hexagonal domain. A predator lurking within that area is especially likely to encounter that individual. Those on the edge are more exposed, because their domains extend outward into a wide open area rather than being bounded by neighbours.

From the edge to the centre

An animal that minimizes its domain of danger will try not to remain at the herd’s edge. If it finds itself there, it moves inward; if the animals begin loosely dispersed, each has an incentive to settle into a gap between others. No one needs to direct this movement. The separate efforts to avoid being the nearest prey can draw individuals together, turning a scattered group into an increasingly dense aggregation.

The model therefore explains a collective pattern through individual incentives. Each animal benefits from occupying a position surrounded by others, while the exposed edge remains unavoidable for someone. The inward movement is ceaseless in the model because each individual’s preferred outcome is to leave the risky position to another.

Limits and significance

The concentration cannot increase without limit in real life: opposing pressures prevent the animals from collapsing into a single mass. The model does not claim otherwise. Its value is that even a simple set of assumptions—predators tending to attack the nearest prey and prey tending to reduce their own exposure—can generate aggregation. More elaborate models may describe real circumstances more closely, but that does not remove the explanatory usefulness of Hamilton’s simpler account.

The selfish herd is not a theory of cooperation. In the model, there is no altruism and no individual actively protecting the group; each animal exploits the positions of the others to improve its own safety. The phrase consequently names a herd of selfish individuals, not a herd acting as a unified selfish agent. Its wider significance lies in showing how a group-level appearance—animals clustered together against danger—can emerge from selection acting on individual advantages, without requiring a shared purpose or benefit to the population as a whole.