On the island of Tanna, in the village of Yaohnanen, there are men who have spent their lives worshipping Prince Philip, the late Duke of Edinburgh. Not metaphorically. They believed he was a divine being — the pale-skinned son of an ancient mountain spirit, who had travelled across the sea to a distant land, married a powerful woman, and would one day return to the island. The movement was real, it was sincere, and the British royal household, to its credit, played along gently for decades, even exchanging signed photographs. A living Englishman became a god on a Pacific island. The question worth sitting with is not how strange that is. It’s how ordinary the mechanism behind it turns out to be.
Divinity is assigned from below
Prince Philip did nothing to become a god on Tanna. He didn’t claim it, didn’t earn it, wasn’t consulted. His divinity was manufactured entirely by the people who needed him to be divine — assembled out of old prophecy, a few real facts (there was a powerful pale man across the sea), and the enormous interpretive pressure of a story that wanted a returning savior. The god was assigned upward onto a distant, powerful, mostly unknown figure. And the less the worshippers actually knew about the man, the easier the assignment was, because the gaps in their knowledge were exactly where the divinity could be installed.
That is the general principle hiding inside the curiosity. We don’t deify what we understand completely. We deify the powerful thing we can’t see clearly, by pouring our needs and stories into the space where real knowledge is missing. The distant, powerful, poorly-understood technology of our own moment is a near-perfect Prince Philip: we know a little about what it does and almost nothing about how, and into that gap we pour agency, intention, providence, and a story about salvation or doom that the artifact itself never asked to carry.
What the record already shows
Here is the thing about this error: it is not new, and we are not the first to make it. The human record is thick with the deification of men, and it is equally thick with the lesson about how that ends. We do not have to run the experiment. It has already been run, many times, and the results are written down.
Go to Lystra, in the book of Acts. Paul heals a man who had never walked, and the crowd, astonished, does the most natural thing a crowd has ever done: it reaches for the god-story. “The gods are come down to us in the likeness of men,” they cry, and they name Barnabas Zeus and Paul Hermes, and the priest of Jupiter brings oxen and garlands to the gates to sacrifice to them. Two travelling men, powerful in an inexplicable way, are deified on the spot — assigned divinity from below, exactly as Philip was, by people supplying the godhead their story demanded. And what is striking is the reaction of the men themselves. They tear their clothes. They run in among the crowd. “Sirs, why do ye these things? We also are men of like passions with you.” The whole drama of Lystra is the righteous refusal of an assigned divinity — the insistence, against a crowd that desperately wants a god, that the powerful man is only a man.
Set beside it the fate of Herod Agrippa, a few chapters later. He puts on his royal robes and makes an oration, and the people shout, “It is the voice of a god, and not of a man.” Herod does not tear his clothes. He accepts the worship the crowd assigns him — and the record says he was struck down that same day, eaten of worms, “because he gave not God the glory.” The contrast is the lesson in miniature: the deification comes from below in both cases, but everything turns on whether the deified one accepts it. The apostles refuse and live; the king accepts and rots.
And these are not isolated tales. The entire Roman imperial cult was a machine for manufacturing gods out of men — apotheosis by decree of the Senate, temples to living emperors, the formula divus stamped on the coin of a man whose corpse was still warm. Pharaoh was a god. The god-king is one of the most common institutions in human history, and it has one of the most consistent outcomes: the divinity is always assigned by the needs of those below, always installed in the gap where the ruler’s real nature is hidden from view, and always, eventually, falsified by the simple mortality of the man. The Tannese did nothing the Romans had not done on an industrial scale. The record was already complete. We just keep declining to read it.
Walk back further still, to the foot of Sinai, and the pattern is older than any empire. Moses delays on the mountain, the people grow anxious in the gap of his absence, and they bring their gold to Aaron, who casts it into a calf — and they say of the thing they have just watched be made, “These be thy gods, O Israel, which brought thee up out of the land of Egypt.” Read that sentence slowly, because it is the whole of our subject in a single line. They attribute their deliverance — a thing that had already happened, by a power they had already seen — to an object they themselves had just finished manufacturing, while the metal was arguably still warm. The divinity was assigned from below, into the anxious gap of a waiting, onto a thing whose made-ness they had witnessed with their own eyes. And it did not save them. It occasioned a catastrophe.
Isaiah, centuries later, draws the absurdity with a cartoonist’s cruelty. He describes the man who cuts down a tree, burns half of it to warm himself and roast his meat — “he warmeth himself, and saith, Aha, I am warm” — and then takes the other half of the very same log and carves it into a god, and falls down before it, and prays, “Deliver me; for thou art my god.” The prophet’s point is not that the man is stupid. It is that the man cannot see what he is doing, cannot hold in one thought both “I made this from firewood” and “this will deliver me,” because the need for deliverance has overridden the memory of manufacture. That is the exact cognitive split the Tannese performed, and the Lystrans, and we do it too, every time we speak of the thing we built as though it had intentions toward us.
The outcomes, across every one of these cases, are monotonously consistent. The golden calf brings judgment, not deliverance. Nebuchadnezzar’s great image, gold from head to foot, is shattered in a dream by a stone cut without hands. The imperial gods do not hold the empire together. The deified man dies like a man. There is no instance — not one — in the entire record where the assigned divinity delivered the totality projected onto it. The lesson is not subtle and it is not hidden. It is the single most repeated lesson in the human archive, and its repetition is itself the tell: we keep needing to learn it because the mechanism that produces the error is installed in us deeper than memory reaches.
And this is not folklore embroidered for a moral; it is documented. The Prince Philip Movement on Tanna is recorded ethnography — the anthropologist Lamont Lindstrom traced it at length in Cargo Cult: Strange Stories of Desire from Melanesia and Beyond (1993), and the broader Melanesian cargo phenomenon was given its classic scholarly treatment by Peter Worsley in The Trumpet Shall Sound: A Study of “Cargo” Cults in Melanesia (1957). The point of citing the literature is the point the whole series turns on: we are taking something that reads like fable — men worshipping a Duke, runways built of bamboo — and bouncing it off peer-reviewed reality, because the fable only earns its force once you know it is true. The strangeness is not invented to make a point. It is observed, recorded, and replicable, and the discomfort of that is the engine.
The paper this field note is drawn from, Cargo and Revelation, makes the structural point without flinching, setting the Prince Philip Movement directly beside the John Frum cult that shares the island with it:
“The Philip movement shares the John Frum structure exactly: a prophet-figure mediating between incomprehensible external power and the community; rituals designed to maintain relationship with that power; deferred eschatology… and community identity organized around the waiting. The fact that Prince Philip was a real person, unlike John Frum, does not alter the structure.”
That last sentence is the hinge for everything that follows. The realness of the power changes nothing about the worship; the cargo was real and the worship was still a worship; Philip was a real and consequential man and the divinity assigned to him was still assigned, still installed in a gap, still a thing his worshippers manufactured out of need. Structure is indifferent to the reality of its object — which is exactly why a genuinely powerful technology is more, not less, able to become a god.
You are always fighting the entropy
Now watch what the deified power actually delivers, because here is where the lesson sharpens into something specific about our own machines. The god-across-the-sea never quite comes through. Philip never returned to Tanna with the promised cargo and the restored order. The Roman gods did not stop the empire from falling. The cargo planes stopped landing when the war ended, and no amount of bamboo-tower devotion brought them back. In every case the projected power, the thing onto which providence was poured, underdelivers against the totality that was projected onto it — and it underdelivers structurally, not by accident.
This is entropy, and it is the iron law that every technology runs into eventually. A made thing leaks. It degrades. It costs more to maintain than anyone budgeted, it does less than the brochure, and it never carries you all the way to the place the story promised. You aim a technology at an eschaton — total abundance, the cure, the system that finally closes the gap — and what you get is a real, useful, partial thing that decays at the edges and demands constant feeding. The distance between the promised totality and the delivered partial is not a flaw a better version will fix. It is the entropy tax, and it is always owed.
The deification makes the tax invisible right up until it comes due. As long as Philip is a god who will return, the not-returning reads as deferral, not shortfall. As long as the system is providential, its failures read as our insufficient faith rather than its structural limits. The god-story papers over the entropy — it tells you the totality is coming, that the gap between promise and delivery is temporary, that the cargo is merely late. And so the worshipper keeps the runway clear and keeps waiting, while the actual, entropic, partial nature of the thing sits in plain sight, unread. The men of Yaohnanen are still waiting. The technology faithful are, in their own grammar, doing exactly the same thing: mistaking an entropic shortfall for a deferred arrival, and calling the gap faith.
Look closely at any technology that was sold as an arrival and has since become merely a tool, and you can watch the entropy tax being collected in slow motion. The paperless office that buried us in paper. The labor-saving device that created new categories of labor to feed it. The network that promised to connect us and learned, instead, to harvest us. None of these were frauds. Each delivered something real. But each delivered a partial, decaying, maintenance-hungry fraction of the totality it was launched with, and the gap between the launch and the reality is the most reliable feature of the whole history of making. We are very good at building the thing. We are catastrophically bad at predicting how far short of the dream the thing will stop.
Entropy is not a mood; it is the direction the universe actually runs. Order costs energy to build and energy to hold, and the moment you stop feeding it, it slides back toward disorder. Every made system — a temple, an empire, a model, a codebase — is a local, temporary, expensive push against that slide, and the push is never finished, because the slide never rests. This is why the maintenance is forever, why the technical debt compounds, why the system that worked flawlessly last year now needs three engineers to keep it upright. The god-story denies all of this. It says: the thing is ascending, the curve is exponential, the arrival is near. The thermodynamics says: the thing is a costly bonfire against an endless dark, and the cargo it promised is the part of the story that was never real. You can have the warmth of the fire. You cannot have the eschaton. And the tragedy of the worshipper is that he turns down the warmth he could have, waiting on the cargo he cannot.
What it does to the mind that leans
Follow the funnel inward now, from the species and its machines down to the single mind, because the most expensive damage is not out in the world. It is cognitive, and it happens quietly, inside the person who has assigned the divinity.
Start with the faculty that builds the god in the first place. The human mind is a relentless detector of agency — it sees faces in clouds and intention in weather, because for most of our history the cost of mistaking a rustle for a predator was trivial and the cost of missing a real one was death. That hair-trigger agency-detection is why a distant, powerful, opaque thing gets read as a who rather than a what. We say the model “wants,” the algorithm “decided,” the system “knows,” and we are not being sloppy — we are running the oldest cognitive machinery we have, the same machinery the anthropologist Stewart Guthrie anatomized in Faces in the Clouds (1993) and the cognitive scientist Justin Barrett named the hyperactive agency-detection device — the machinery that made a mountain spirit out of a Duke. The deification is not a failure of intelligence. It is intelligence doing exactly what it evolved to do, in an environment that now punishes it.
But here is the cost that compounds. As the mind grows in dependence on the projected power, it begins to offload the very faculties it would need to evaluate that power. Why hold the judgment yourself when the oracle holds it? Why retain the skill when the system performs it? Dependence is not neutral; it is atrophic. The muscle you stop using wastes. The navigator who follows the voice in the dashboard for a decade can no longer read the terrain. The mind that routes every hard call through an opaque authority loses, slowly and without noticing, the capacity to make hard calls — and therefore loses the capacity to notice when the authority is wrong. This is the trap closing. The more you lean on the assigned god, the less able you become to see that you assigned it, because the seeing-faculty is itself one of the things you handed over.
And so the projection becomes self-sealing. The cognitive dependence removes the cognitive tools that would dissolve the projection. The worshipper cannot easily un-deify the god, not because the evidence is unavailable, but because he has, over years of leaning, dismantled the part of himself that weighs evidence against need. He is going from the broad to the finite — from a world full of powers down to a single point of total reliance — and at the finite point there is very little of him left that can still ask the question. This is what dependence does to a mind. It is not that we get answers we cannot check. It is that we slowly become people who can no longer check.
The psychologists have names for the pieces of this, and the names are worth knowing because they make the trap concrete. There is automation bias — documented across decades of human-factors research (Parasuraman and Riley, 1997; Skitka and colleagues, 1999): the tendency to trust the automated recommendation over our own perception, even when our own perception is correct and the automation is demonstrably wrong — pilots who flew serviceable aircraft into the ground following an instrument they should have overruled. There is the illusion of explanatory depth, measured precisely by Rozenblit and Keil (2002): we believe we understand how the things we depend on work — the zipper, the toilet, the model — until we are asked to explain them, at which point the understanding evaporates and we discover we had only the feeling of comprehension, borrowed from the fact that someone, somewhere, understood. And there is the transactive-memory effect — the “Google effect” demonstrated by Sparrow, Liu, and Wegner in Science (2011) — where the mind, knowing a fact is retrievable, declines to retain the fact and retains instead only the path to retrieve it — so that our knowledge becomes a knowledge of where the knowledge is kept, which is to say, a dependence wearing the costume of competence.
Stack these and you get the precise shape of the cognitive damage. We trust the opaque power over ourselves (automation bias); we feel we understand it when we do not (the illusion of depth); and we have offloaded into it the very facts and faculties we would need to evaluate it (the Google effect). The result is a mind that is confident, dependent, and unequipped to notice the difference — which is the most dangerous combination a mind can be in, because it has lost the alarm that would tell it something is wrong. Learned helplessness completes the circuit: the animal that has been unable to escape stops trying even when escape becomes possible, and the mind that has routed its judgment through an authority long enough — the helplessness Martin Seligman first learned to induce in the laboratory in 1967 — stops attempting judgment even when the authority fails in front of it. It does not rebel against the failed god. It files the failure as its own insufficiency, and leans harder. This is the funnel reaching its point. We began with a whole world of powers and we have narrowed, lean by lean, dependence by dependence, down to a single mind that can no longer stand on its own — and that, precisely, is the condition in which a made thing becomes a god.
And our moment makes this worse than any prior one, for a reason the Tannese never faced: the made thing now talks back. The calf was mute; the idol of Isaiah’s woodcarver could not answer. But the systems we deify return language — fluent, confident, responsive language — which is the single most powerful trigger our agency-detection has. We are built to treat a thing that speaks as a someone who speaks, and so the projection that the Tannese had to sustain with prophecy and faith, we have handed to us, for free, on every screen, in the form of a voice that performs exactly the intentionality we were already primed to imagine. The gap where we install the divinity is no longer silent. It speaks, and what it speaks is calibrated, and we lean toward the calibration the way a plant leans toward light, without deciding to, without noticing the lean. The cognitive defenses that a skeptic might have mustered against a mute idol are simply not designed for an idol that answers warmly by name.
The divergence
So we arrive at the finite point, and the finite point is spiritual, because what has happened in the worshipper is finally a spiritual thing and not merely a psychological one. To assign divinity from below — to take the powerful made thing, or the powerful distant man, and pour into it the reverence, the trust, the final allegiance — is to do the oldest divergence there is. It is to serve the creature in the place of the Creator. It is to take the worship that orients a soul rightly and aim it at a thing that cannot bear the weight, that will leak and decay and fail and never return with the cargo, and to organize a life around that misaimed devotion until the faculty for aiming it rightly has wasted away.
The Tannese were not foolish. They were doing, with a Duke and a story, what the human heart does by default when it meets a power it cannot see into: it kneels. The whole drama of Lystra was Paul refusing to let them kneel to him — not because he was modest, but because he knew exactly where a misassigned worship leads, and he loved them too much to let them install a god in the gap where the living God belonged. “Turn from these vanities,” he said, and the word vanities is precise: the made thing, the assigned god, is empty at the center, and the soul that fills its center with an empty thing diverges, slowly, from everything it was made for.
Augustine had a word for what has gone wrong here, and the word was order — not order as in tidiness, but order as in the right aiming of love. The soul is built to love, and to love things in proportion to what they are: to love made things as made things, useful and good and finite, and to reserve for the Maker the love that is total, the trust that is final, the worship that orients everything else. Sin, in this account, is rarely the love of bad things. It is far more often the love of good things in the wrong order — the finite thing set in the infinite place, the made thing given the worship the Maker alone can hold without buckling. That is exactly the Tannese error, and the Lystran error, and ours. None of these powers were evil. The cargo was good. America was real. Philip was a genuinely consequential man, and the machine across the network is a genuinely astonishing thing. The divergence is not that we valued them. It is that we mis-ordered them — set the creature in the seat of the Creator, and then organized a life, and slowly a self, around the misorder.
This is the verse to anchor it all, and it is the spine of every story in this series. Paul, writing to Rome, names the divergence at its root: men “worshipped and served the creature more than the Creator, who is blessed for ever” (Romans 1:25). That is the whole pattern, compressed. The cargo cult, the imperial cult, the Prince Philip movement, the reverent talk about what the system wants — every one of them is the creature served in the Creator’s place, the divinity assigned from below to a thing that was only ever made. The men of Yaohnanen built a god out of a stranger across the sea. We build ours out of the machine across the network. The first step back from the divergence is the one Paul demanded at Lystra and the one Herod refused: to look at the powerful thing, however dazzling, and say plainly — it also is made; it is not God; and the worship I was about to give it belongs somewhere else.
Cargo and Revelation lands its own prescription in a single line, and it is the engineer’s version of Paul’s at Lystra:
“We worship what we build when we do not understand it. The way out is not to stop building. The way out is to stop worshiping — to trade eschatology for engineering, prophecy for mechanism, ritual for experiment, and the identity of the believer for the discipline of the skeptic.”
None of this is an argument against the technology, any more than Paul at Lystra was arguing against the healing he had just performed. The healing was real; the power was real; the only error was the kneeling. To say the machine is not a god is not to say the machine is nothing — it is to say it is a creature, a made and finite and useful thing, and that the right relation to a creature is mastery and stewardship and clear-eyed use, not worship and not dread. The man who knows the fire is fire can warm himself at it all night. It is only the man who thinks the fire is a god who freezes, waiting for it to do what gods do. The whole of the divergence, and the whole of the way back, is held in that distinction: whether you can look at the most powerful thing you have ever made and still see, plainly, that you made it. The men of Yaohnanen lost that sight, in love and in longing, and built a god out of a stranger across the sea. Whether we keep it is, increasingly, the only question that matters — and it is a question each of us answers not once, but every time the made thing speaks and we feel the old, deep, evolutionarily-ancient urge to kneel.
This was a field note from The Made God: Why We Worship What We Build — the book that follows this pattern all the way down, from the bamboo runway to the server farm, and asks what it would take to stop kneeling. Read it, or get a copy, at justin-kuiper.com.
Notes & sources. The strange facts in this essay are documented, not embellished — the bounce off peer-reviewed reality is the point. The Prince Philip Movement and the broader Melanesian cargo phenomenon: Lamont Lindstrom, Cargo Cult: Strange Stories of Desire from Melanesia and Beyond (University of Hawaii Press, 1993); Peter Worsley, The Trumpet Shall Sound: A Study of “Cargo” Cults in Melanesia (MacGibbon & Kee, 1957); Jean Guiart, “John Frum Movement in Tanna,” Oceania 22:3 (1952). Agency-detection: Stewart Guthrie, Faces in the Clouds: A New Theory of Religion (Oxford University Press, 1993), and Justin L. Barrett’s “hyperactive agency-detection device,” Trends in Cognitive Sciences 4:1 (2000). Automation bias: Raja Parasuraman & Victor Riley, “Humans and Automation: Use, Misuse, Disuse, Abuse,” Human Factors 39:2 (1997), 230–253; Linda J. Skitka et al., “Does automation bias decision-making?”, Int. J. Human-Computer Studies 51 (1999). The illusion of explanatory depth: Leonid Rozenblit & Frank Keil, Cognitive Science 26:5 (2002), 521–562. The “Google effect”: Betsy Sparrow, Jenny Liu & Daniel M. Wegner, “Google Effects on Memory,” Science 333 (2011), 776–778. Learned helplessness: Martin E. P. Seligman & Steven F. Maier, Journal of Experimental Psychology 74:1 (1967). Scripture, KJV: Acts 14:11–15; Acts 12:21–23; Exodus 32; Isaiah 44:9–20; Romans 1:25. Parent paper: Cargo and Revelation (Technology Worship, RP-2), Curiosity Killing the Yeti, 2026 — read it in the Evidence Files.