Biology Proves Hierarchy Is Optional

When you argue that human societies could be organised with less hierarchy, the reply is usually some version of: hierarchy is how coordination works at scale. Above a certain number of participants, someone has to be in charge, or nothing gets decided and nothing gets built. Every human society has had hierarchies; that near-universality is treated as evidence of necessity.

The reply is available and it is empirical. Nature coordinates enormous numbers of agents without command structures, routinely, and has been doing it far longer than we have.

Cells

The clearest case is the one you are made of.

Communication between the cells of your body is not hierarchical. Cells do not issue orders to one another. There is no cell that decides and other cells that comply.

The obvious objection is hormones: surely that is a central system issuing instructions. But look at what a hormonal signal actually is. It is a broadcast into a shared medium, which cells respond to according to their own receptors and their own local state. Two cells receiving the identical signal do different things, or nothing, depending on what they are and where they are. That is not a chain of command; it is a shared environment that participants read.

And beyond even that, cells communicate locally, with no reference to any central system at all — direct contact, local gradients, immediate neighbours. Much of what your body does is settled entirely between adjacent cells with no signal from anywhere else.

The result is an organism of tens of trillions of components maintaining itself for decades. Whatever is achieving that, it is not a command hierarchy.

Ant Colonies

The classic example, and worth restating because the name misleads.

There is a queen, so people assume there is a monarch. The queen does not direct the colony. She lays eggs. She issues no instructions, receives no reports, and has no view of what the colony is doing. Remove her and the colony continues functioning until it runs out of workers.

What actually coordinates the colony is local rules plus a shared medium — pheromone trails, contact rates, simple response thresholds. Each ant follows rules referencing only its immediate surroundings. Out of that comes foraging, defence, brood care, waste management, and the allocation of labour across tasks in proportions that shift with need.

No ant knows the plan. There is no plan. There is a set of local rules whose aggregate behaviour looks like planning from the outside.

The Argument This Supports

Now the inference, and I want to keep it modest, because there is an overreach available here that would discredit the whole point.

I am not claiming that human societies can be organised like ant colonies, or that biology gives us a blueprint. Ants are not people; cells have no interests of their own; the analogy breaks down quickly if pushed.

What biology establishes is narrower and sufficient: hierarchy is not a requirement for coordinating large numbers of agents. It is one solution among several. The claim that above some scale command structures become necessary is simply false as a general proposition, and we have overwhelming evidence against it.

Moreover, these are not marginal or fragile arrangements. They are what evolution converged on, repeatedly, across unrelated lineages, and they have been running for hundreds of millions of years. If distributed coordination were unstable at scale, this is where we would see it fail. We see the opposite: the systems that have been running longest are the ones with no one in charge.

Where It Bites

I have argued elsewhere that the evils of hierarchy look intractable partly because no human society has ever lacked them — that the universality of the thing suggests we are up against something deep.

Biology is the existence proof that cuts against my own earlier reasoning. The universality of hierarchy among humans is a fact about humans, not a fact about coordination. Something about our species, our history, or our institutions selects for command structures. Nothing about the problem of organising many agents requires them.

That is a considerably more hopeful position than the one I started from, and it relocates the question. The problem is not that hierarchy is necessary. The problem is finding out what, in our case specifically, keeps producing it.