Lloyd Taylor

The Cognitive Origins of Authority


Something moves in the darkness outside.

Your Pleistocene ancestor doesn’t know what it is. Only that something shifted — a sound, a displacement of air, a change in the quality of the silence. The body has already responded before the question forms. The question, when it arrives, is not what moved. The question is who is out there.

You exist because your ancestor assumed an agent. The brain that asked what instead of who did not survive to be your ancestor. It belonged to someone who waited a half-second too long to discover whether the darkness contained intention, and that half-second was the last decision they ever made.

Evolution does not optimize for accuracy. It optimizes for survival. The hardware that survived was the hardware that assumed agency first and corrected only when the agent could be found and falsified — and in the conditions that shaped it, that was usually soon enough. Your ancestor heard a sound in the dark and thought: someone or something is there. You are the consequence.

The system has no off switch. This is not a metaphor. The neural architecture that scans for agents in ambiguous signals runs continuously, below the threshold of conscious control, and it has one design constraint that determines everything that follows: it cannot return a null result. Failure to find the agent was not an acceptable outcome in the environment that built it. Failure to find the agent meant death. So the system does not stop at no agent found. It searches until it finds one. And in the dark, with the wind moving through the trees and the crops failing and the children dying of causes no one can name, it always finds one.

For most threats, the system could verify. The tiger either appeared or it didn’t. The search terminated with evidence. But some threats were different in kind. Drought did not appear once and retreat. Disease returned the following season. The river flooded in the same month three years running. Recurrence implied something the one-time threat did not: a pattern. And a pattern, to hardware that could only model the world in terms of agents, implied intent — the same neural architecture that detected agents in ambiguous physical signals ran on temporal patterns too. Recurrence was the signature of a mind behind the phenomenon, because in the environment that built the hardware, recurrence without a cause was not a category the system had learned to recognize. Intent implied a disposition — something that could be pleased or displeased, appeased or provoked. The tiger had no preferences about your ancestor’s behavior. The thing behind the drought did. Your ancestor had no framework for stochastic variation in a complex adaptive system. The hardware provided the only available translation: someone is doing this to us.

There was no one. There was weather, and microbes, and the fluid dynamics of rivers, and none of it had preferences about anything. But the hardware did not have a category for consequential pattern without author. The search ran, as it always ran, until it found an agent. And when the agent could not be located in the physical world — could not be caught, cornered, or killed — the system did what it was built to do: it kept searching. The unfindable agent was not evidence of no agent. It was evidence of an agent that had not yet been found. An agent that was, by its very unfindability, more powerful than any tiger. Your ancestor did not choose to believe. The hardware made the only move it knew how to make. The god was not invented. The god was the inevitable output of a system that could not return a null result — applied, for the first time, to something that wasn’t there.

If the agent had preferences, it could be influenced. This was not a theological leap. It was the same cognitive move your ancestor made with every consequential relationship they had ever navigated. You learned, early, that the people who controlled your survival could be appeased. That offerings worked. That timing mattered. That there were right words and wrong words, right gestures and wrong ones. The hardware that modeled other minds — that ran simulations of what the powerful wanted and adjusted behavior accordingly — was the same hardware now pointed at the drought. Sacrifice, ritual, propitiation: not the inventions of credulous minds, but the direct application of social intelligence to the only available explanation. The medium was invisible. The relational logic was identical to every negotiation that had ever kept your ancestor alive.

But here the new agent differed from every prior relationship in one respect that changed everything. The tiger could be seen. The chief could be approached. The agent behind the drought could not. It produced no direct signal. It could not be observed, questioned, or read. The hardware that ran on facial expressions and tone of voice and the micro-signals of bodies in proximity had nothing to work with. What individual hardware produced as private inference, the community converted into shared necessity. And yet the relationship had to be maintained, because the stakes — crops, rain, the lives of children — were higher than any relationship your ancestor had ever managed. Since the tribe could not find the agent behind the storms, they created one they could see: the interlocutor, the shaman, the priest.

The one who spoke for the silence held something no one else had: exclusive access to the most consequential agent in the community’s world. Not military strength. Not accumulated food. Not even kinship — the oldest form of social leverage. Something older than any of those, because it was prior to all of them. The crops grew or failed depending on whether this relationship was properly maintained. The rain came or didn’t. The children lived or died. The one who managed the channel between the community and the force that controlled all of this was not powerful in the way a strong man was powerful. The strong man could be beaten. This person could not be replaced, because the role was not about strength. It was about access. And access, when it is singular and unverifiable, is the original form of power. The strong man’s authority required continuous demonstration. This authority required only that no one could prove it absent.

Here is where the ascent makes its most consequential step. The agent who explained the storms was also, it turned out, available to explain everything else. Why did the harvest succeed this year and fail last year? The agent had a preference. Why did this family prosper and that one suffer? The agent was expressing a judgment. Why did this man have the right to lead, and that one not? The agent had designated him. The force that your ancestor’s hardware had generated to explain the weather turned out to be infinitely extensible. Every question that the community needed answered — every question about order, about legitimacy, about who had the right to do what to whom — could be routed through the same channel. The agent behind the drought became the agent behind the throne. The explanation for the rain became the foundation of the law. The detection artifact became load-bearing.

This is not a claim that every institution is theology in disguise, or that stable coordination requires active belief. Bureaucracies run on procedure. Markets run on incentive. Legal systems run on enforcement. These can persist for long periods without anyone consulting the original story. The claim is narrower: when human beings ask not whether an authority is powerful but whether it is legitimate — whether it has the right to bind rather than merely the capacity to compel — the chain has historically terminated in assumptions the system cannot ground from within itself.

This was not conspiracy. No one decided to use the invisible agent for political purposes. The same hardware that needed an agent behind the storm needed an agent behind the social order. Why does the strong man rule? Why does this boundary hold and not that one? Why is this promise binding? The hardware ran on those questions too, and produced the same output: someone authorized this. The religion and the politics were never separate because they were never produced by separate instruments. They were the same move, made by the same hardware, on adjacent domains. The priest-king was not a fusion of two things. The priest-king was the visible form of a single underlying function — the human proxy for an agent the community had created because it could not survive without one.


The house that this built is the one you live in.

Nearly every durable system of legitimate authority that has governed large human populations — every court, every throne, every constitution, every binding promise between strangers — traces its authority, if you follow the chain far enough, back to the agent your ancestor’s hardware generated to explain the rain. The chain is long. The steps are real. The institutions are genuine. But the anchor is a detection artifact produced by a system that could not afford to be wrong about tigers.

This is not a small claim. It means that the architecture of human legitimacy contains a Gödelian problem at its foundation — constructed from inside the instrument, by the instrument, incapable of examining the instrument’s calibration from within the axiom set the instrument produced. The fish cannot see the water. In this case, the water is the tiger.

The Enlightenment saw the problem, or part of it. If the authority of kings derived from a god, and the god was a human fabrication, then the authority of kings was a fabrication built on a fabrication. The solution seemed clear: replace the divine anchor with reason. Ground legitimacy in the rational consent of the governed rather than the preferences of an invisible agent. It was the most ambitious re-founding in the history of human thought.

It never fully solved the binding problem. Not because the thinkers were insufficient — they were among the finest minds the species has produced. Reason, examined carefully, has no authority to obligate anyone to anything. It can describe. It can recommend. It cannot bind. The social contract theorists sensed this and spent enormous energy trying to locate the source of legitimate authority in something other than the divine. What they produced was a hypothetical agreement among rational agents who never met, never signed anything, and never consented to the terms. The theorists knew this — the contract was never meant as history. It was meant as logical ground. But logical ground is not the same as binding force, and the gap between them is precisely the gap the divine anchor had filled. They replaced one invented agent with another invented process. The hardware was still running. The null result was still unavailable.

Nietzsche saw the structural consequence more clearly than anyone. God is dead, he wrote — meaning not that a being had ceased to exist, but that the foundation on which Western civilization had built its entire architecture of meaning, morality, and legitimate authority had been removed. And the architecture was still standing. People had stopped believing in the foundation while continuing to live in the house. He called what remained the shadow of God — the moral vocabulary, the sense of binding obligation, the feeling that some things were genuinely wrong and not merely inconvenient — persisting in a civilization that had quietly removed the only thing that had ever given those feelings their force. He named the problem with perfect precision. He did not solve it. No one has. The shadow persists not because people still believe the original story, but because nothing has been built that doesn’t need a story of that kind.

The shadow is still there. It falls across every courthouse and every parliament, every oath of office and every declaration of rights. The language has changed. The anchor has not been replaced. What holds the structure up, in the absence of the agent that was always its foundation, is a question that modernity has been loudly not answering for two hundred years — filling the silence with ideology, nationalism, and the performed confidence of institutions that know, somewhere below the level of official acknowledgment, that the ground beneath them is the same ground it always was.

Your ancestor heard something move in the darkness and assumed an agent. The assumption was wrong. It was also the most consequential cognitive event in the history of the species. Everything followed from it — the shaman, the priest-king, the divine right of kings, the moral law, the social contract, the rights of man. The whole vast apparatus of human civilization, built level by level on a foundation that was never what it claimed to be, and that cannot be replaced from inside the house it built.

From the cave to the constitution, we had to find, or create, the agent.


References

  • Barrett, Justin L., Why Would Anyone Believe in God? (AltaMira Press, 2004)
  • Boyer, Pascal, Religion Explained (Basic Books, 2001)
  • Durkheim, Émile, The Elementary Forms of Religious Life (1912)
  • Epley, Waytz & Cacioppo, Psychological Review (2007)
  • Frankfort, Henri, Kingship and the Gods (University of Chicago Press, 1948)
  • Guthrie, Stewart E., Faces in the Clouds (Oxford University Press, 1993)
  • Nietzsche, Friedrich, The Gay Science (1882)

Copyright © 2026 – Lloyd W. Taylor – https://lloydwtaylor.com – ltaylor@netelder.com