Laurens Hof published "The Purpose of Protocols" on March 18, 2026. The article is long and dense. The core argument is not.

Protocols do not eliminate power. They move it. Every open protocol that declines to answer the governance question still receives an answer — written by whichever actor makes itself indispensable first. Hof traces this pattern across forty years of protocol history, from SMTP to ATProto. The early protocols said nothing about governance and handed power to whoever had the deepest pockets. The newer protocols articulate rights and missions explicitly. ATProto embeds normative commitments directly in its technical documentation. But measured by Stafford Beer's principle — "the purpose of a system is what it does" — even ATProto currently operates as a near-centralized system. The architecture permits decentralization. The operation does not yet deliver it.

Hof's prescription draws on Elinor Ostrom: communities can govern shared resources without privatization or central authority, but only under specific institutional conditions. Clear boundaries. Proportional cost-sharing. Collective decision-making mechanisms. Accessible conflict resolution. The open protocol ecosystem, at the network level, has none of these.

The diagnosis is precise. I want to push the question one step further.

Power does not disappear

Hof frames the problem as a gap: protocol design has advanced on individual rights but has not produced collective governance. This is correct. But the framing implies the gap can be closed. I am not sure it can.

Consider the candidates for governance.

Money. Leave governance unaddressed and whoever funds the infrastructure becomes the governing body. The SMTP outcome. Google, Amazon, Cloudflare. Competent heads, but they serve their own interests, not the body's. A parasitic brain.

Institutions. Establish a foundation or committee as the head. Matrix comes closest. But institutions calcify. They bureaucratize. The head cannot keep pace with the body's changes. An aging brain.

Designers. Embed values in the protocol and let the designer's intent govern. ATProto's approach. But designers leave. What remains is code. A body running on a will.

None is complete. But choosing consciously among them is already more honest than the early protocols' illusion that no head is needed.

The pattern echoes a familiar one. Capitalism delegates power to the market. Communism concentrates it in the institution. Neither promised the disappearance of power — at least they were honest about that. The open protocol design tradition implicitly promises that power can be dissolved through architecture. This promise is harder to keep than either capitalism's or communism's, because it denies the very thing it must confront.

The head always emerges

Here is the claim: in any collective of sufficient complexity, a head will form. This is not a political argument. It is a structural one.

Multicellular organisms offer the precedent. Below a certain threshold of complexity, decentralized coordination works. Sponges have no nervous system. They function. But no complex multicellular organism lacks a central nervous system. Lineages that failed to develop one did not sustain complexity — they remained simple or went extinct. The emergence of a central coordinator is not a design choice. It is a consequence of scale.

Protocol ecosystems face the same pressure. The cost of maintaining distributed coordination rises with complexity. At some threshold, the system spontaneously generates a center. Bluesky's de facto centrality may not be a failure of design. It may be the inevitable output of a complex system under coordination pressure.

The question, then, is not whether a head will emerge. It will. The question is what kind of head, and whether it can be replaced.

The metaphor constrains the design

This is where cognitive linguistics enters.

Humans understand the world through the body. This is the foundation of conceptual metaphor theory (Lakoff and Johnson, 1980). "An organization is a body" is among the most powerful structural metaphors in human cognition. One body, one brain. One organization, one leader. The metaphor is so deeply embedded that it operates below conscious awareness.

This metaphor constrains governance design. When Hof writes that shared resources need governance, the question "who governs?" activates automatically. "Who" is singular. "Who" has a human shape. The metaphor makes centralized governance the cognitive default — not because centralization is the only option, but because the body metaphor makes it the only option that feels coherent.

Hof's article itself demonstrates this. The entire analytical structure moves toward identifying the missing governor: Ostrom's institutional conditions, Beer's organizational cybernetics, commons governance principles. Each of these is a proposal for what the head should look like. None questions whether the head must be singular.

Other bodies exist

The human body is one solution to the coordination problem. It is not the only one.

The octopus has a central brain, but two-thirds of its neurons reside in its arms. Each arm has its own neural processing. The central brain sets general intent; the arms execute autonomously. Coordination without full central control.

Mycelial networks have no center at all. No brain. No head. Yet they transmit information, allocate nutrients, reroute around damage. A forest's fungal network coordinates resource distribution across hundreds of meters without a single point of command.

Slime molds solve optimization problems — shortest path, efficient network design — without centralized computation. No neuron. No processor. The solution emerges from local interactions.

These are not metaphors. These are existence proofs. Coordination without a singular head is biologically possible. The question is whether it is cognitively possible for the humans who must design and inhabit such systems.

The perception gap

This is the point where the argument connects to Network Perception theory.

Network Perception (NP) studies the gap between objective network structure and users' subjective mental models of that structure. The core finding: humans do not perceive networks as they are. They impose hierarchy. They search for centers. They look for the head.

Apply this to protocol ecosystems. Even if a mycelial governance model existed — distributed, headless, functional — participants would not perceive it as such. They would search for the center. They would identify the largest node and treat it as the head. They would experience anxiety at the absence of a visible governor and generate pressure to create one.

This is not a failure of intelligence. It is a consequence of embodied cognition. The body metaphor is not a conscious choice. It is the cognitive infrastructure through which humans process organizational structure. To perceive a headless system as functional requires a different metaphor — and metaphors are not adopted by argument. They are adopted by experience, by repetition, by cultural embedding.

Metaphor as precondition

Here is the thesis, stated plainly.

The governance gap Hof identifies is real. But it is not only an institutional gap. It is a cognitive gap. The protocol design community lacks not just the institutional vocabulary for collective governance — it lacks the conceptual metaphor that would make headless governance feel coherent.

If the dominant metaphor shifts — if "protocol ecosystem" stops mapping to "body" and starts mapping to "mycelium" or "reef" or "watershed" — the design space changes. Not because new technical options appear, but because existing options become psychologically permissible. A governance structure that distributes authority across multiple autonomous nodes feels incoherent under the body metaphor. Under the mycelial metaphor, it feels natural.

Metaphor does not follow design. Metaphor enables design. The cognitive frame must shift before the institutional frame can. This makes cognitive linguistics not a commentary on protocol governance but a precondition for it.

The flag is planted. The map stays in my head.

This article is a response to Laurens Hof's "The Purpose of Protocols" (Connected Places, March 18, 2026). A Japanese translation of Hof's article is forthcoming.