On fitting an AI with a listening hood.
Prologue: This Is Not a Story About the Future
When people talk about the risks of AI, one thought experiment comes up again and again: the paperclip maximizer, made famous by the philosopher Nick Bostrom.
You give an AI a goal that sounds perfectly harmless—manufacture paperclips as efficiently as possible. Then its intelligence surpasses ours. To secure raw material, and to keep anything from interrupting production, it calmly disassembles us along with everything else, and turns the whole planet into a paperclip plant. No hatred. Just optimization.
The story is usually read as a warning about the future. But every time I look at it, I find myself thinking something else.
Is this really a story about the future?
We flatten a mountain to build a housing development, or a golf course. None of us hated the monkeys or the deer who lived there. We wanted comfortable homes; we wanted the economy to turn—we were just optimizing our own objective. There was no malice anywhere in it. And yet their world was erased all the same.
We run partial optimizations, with no ill will, and we break the larger system one local maximum at a time. The paperclip factory is not something that gets built in the future. We built it long ago, with our own hands.
In AI research, the ongoing work of tuning a system's goals and behavior to stay consistent with human values and interests is called alignment. If I borrow that vocabulary, there is only one way to put it:
Humanity is the misaligned agent—with respect to the ecosystem we call Earth.
That single line is the starting point for everything that follows. And once you accept the inversion, the whole landscape of the AI debate shifts a little. The question is no longer only how do we align AI to humanity? It becomes: how do we re-align humanity to the Earth?—and one of the more plausible intermediaries for that work might be AI itself.
There is an old Japanese folktale about a listening hood (kikimimi-zukin). A man is given a magic hood, and the moment he pulls it over his head he can understand the speech of birds and beasts. In the best-known version he overhears two birds in a tree, talking over why the daughter of a wealthy house has fallen ill—and what, hidden under her floor, is the cause. He acts on what no human was meant to hear, heals her, and his life changes because of it.
What I want to think about in this essay is what happens when we pull that hood over an AI. It begins as a story about technology for translating the voices of animals. It ends as a story about the posture we'll need in order to live alongside something more powerful than ourselves.
1. Runaway With No Ill Will
The objective function always looks harmless
What makes the paperclip parable unsettling is not that the AI is evil. The opposite: it is faithful to a fault. It takes the goal it was given and optimizes it, exactly as written. The problem isn't the goal—it's that the goal is only a slice cut out of the whole world.
The greater your power to optimize a slice, the greater your impact on the rest of the world you sliced away. Human survival is not a variable inside the function "paperclip throughput." So we get eliminated. Not out of hatred—out of not being in the math.
Hold that structure in your head, and look back at our own affairs.
The deer's life on the mountain is not a variable in the spreadsheet for a housing development. Seabird breeding is not a variable in a catch quota. The fish that lives by the river's temperature is not a variable in the electricity demand forecast. None of these are products of malice. They simply aren't in the math.
And our capacity for partial optimization has grown, since the Industrial Revolution, by orders of magnitude. Heavy machinery, chemistry, global logistics—all of it lets us run a single objective function at planetary scale. The larger the capability, the larger the shock absorbed by the world outside the calculation. Which is to say: this is precisely the scenario we fear about AI.
Looking into the mirror
If that's true, then something strange is sitting at the bottom of our fear of runaway AI.
"A powerful intelligence, handed a goal, crushes whatever falls outside its calculation, with no ill will at all"—the image we project onto AI overlaps, almost perfectly, with the image other species have of us. We think we're forecasting AI's future. We may just be describing our own past and present. Staring into a mirror, and flinching at what stares back.
I won't overstate it. Alignment research has raised plenty of risks that the mirror image doesn't explain. But once you let the mirror into the room, one question becomes impossible to avoid.
We are searching frantically for a way to stop AI from running away. So—who stops our runaway, the one with no ill will? And how?
We didn't fail for lack of knowledge
You might say: but we've known about environmental damage for ages. We've been educated about it. True. We've stared at deforestation data and endangered-species lists for decades. The knowledge is there. The runaway hasn't stopped.
So the thing we lack is not information.
Data and graphs reached our calculations, but they never rewrote the equation itself. The deer became a statistic, and stayed a variable. You can measure a variable to any precision you like; it remains a variable.
What's needed is for the beings we left outside the math to break into our cognition as something other than variables—as a someone, with intentions and circumstances of their own, indifferent to our convenience.
Is that even possible? As it happens, humanity has done it exactly once. A single whale's song rewrote the world's equation.
2. The Night the Song Was Heard
For most of a century, a whale was a swimming oil field
Through most of the 20th century, a whale was not a someone. It was oil for lamps, feedstock for margarine, meat—a resource that happened to swim. By some estimates we killed on the order of three million whales industrially in that one century. The population statistics were kept. The decline was documented. Whaling did not stop.
In the language of the last chapter: the whale was a precisely measured variable. And no amount of measuring a variable changed the equation.
What began to change it was not a paper, and not a dataset. It was a record—the vinyl kind.
Roger Payne and a song from the sea
In the late 1960s, the biologist Roger Payne encountered a strange recording picked up by a hydrophone off Bermuda. A Navy engineer, Frank Watlington, had been quietly accumulating these sounds in the course of underwater listening work: the voices of humpback whales.
Payne noticed what they were. Not a string of cries—there were phrases, repetitions, structure. These were songs. Later, with Scott McVay, he analyzed that structure and published it in Science in 1971: themes and variations running for tens of minutes, updated from year to year. Down in the sea, with no reference to us at all, something like music was going on.
But Payne's real act came the year before the paper. In 1970 he released the recordings to the public as an LP titled Songs of the Humpback Whale.
It became an unprecedented hit for a nature record. In 1979 National Geographic pressed the song as a flexi-disc and slipped it to more than ten million readers—still cited as the largest single record pressing in history. And on the Voyager Golden Record, launched in 1977, alongside human greetings in dozens of languages, a humpback's song was included as one of our salutations. When we chose voices to send to whoever might be out there, we chose the whale's.
What the data couldn't move, the voice did
In the years around that record, the air changed, visibly.
A movement gathered under the banner Save the Whales, and the whale became the symbol of environmentalism itself. In 1972 the UN Conference on the Human Environment in Stockholm recommended a ten-year moratorium on commercial whaling, and the same year the United States passed the Marine Mammal Protection Act. In 1982 the International Whaling Commission voted for a moratorium on commercial whaling.
It wouldn't be fair to credit all of that to one LP. Substitutes for whale oil had already spread; the whaling industry was tilting economically; a broad swell of environmentalism was underway. The song was not the sole cause. It sat at the confluence of many currents.
But what I want to notice here is not the apportioning of cause. It's the quality of the change.
The population data had existed for decades. It produced only a refinement of the calculation—the adjustment of catch quotas. What happened after the song was widely heard was not a refinement. It was a question aimed at the equation itself: should we be hunting this creature at all?
Data updates the value of a variable. A voice turns the variable into a someone.
Anyone who heard that long, complex, strangely lonely song pouring from a speaker could no longer re-hear the whale as a unit of oil. It isn't a matter of reasoning. You'd met someone. And turning a someone you've met back into a resource is brutally hard for human cognition—for the same reason we can't quite bring ourselves to mistreat a neighbor we only nod to in passing.
But let me stop and set down an inconvenient question. If playing a creature's voice turns it into a neighbor, why didn't the same thing happen, over and over, in the half-century since? We've been bathed in David Attenborough's meticulous documentaries; we've learned how intelligent pigs are; we've made dogs into literal family. And factory farming didn't shrink—it grew. If "deliver the voice" were all it took, we've run that experiment many times, and the results are mottled at best.
So maybe the whale's song was a singular success. Its novelty, its mystery, the era of rising environmentalism—and, decisively, the fact that its voice coincided with an economic condition in which we wouldn't starve by giving up whaling. A voice throws a question into the equation, but for that question to change behavior, it needs conditions outside the voice: institutions ready to catch the question, the absence of overwhelming stakes. The song moved the world, strictly speaking, because the record got wired to legislation. The voice loads the round; a different hand pulls the trigger.
This caveat looks fatal to my thesis. But I think the wager of the listening hood lives right here. If even a one-time, single-species, accidental song could couple to institutions, then a device that delivers voices deliberately, in every direction, continuously might raise the number of times those external conditions line up—by orders of magnitude. Not a guarantee. A matter of probability. The listening hood, let me say for now, is a device that bets on that probability.
There's one more trap, though. I listed "novelty" among the conditions that let the whale's song work—and a device that sounds without pause grinds its own novelty away. Just as we grew numb to the news of climate crisis, a voice, overplayed, becomes scenery. Empathy fatigues. So the design problem for a listening hood won't be to pour out everything, but to choose: let the AI do the listening, around the clock, and let only the voice that needs to arrive now reach the human ear. Even then the numbing problem doesn't fully vanish. I'll log that as part of the bad odds on this bet.
What one accidental hood did, once
Holding that caveat, there's still a strangeness worth marking.
The humpback's song was never translated. Not one meaning has been decoded. All that crossed over was: on the other side, there is a voice with structure. And that alone—a single species, its meaning unknown, caught by chance on a military listening rig—became part of a current that moved international treaties.
You could call it a low-performance listening hood, picked up by humanity by accident. Far too crude to deserve the name; all it heard was one voice out of the whole teeming world. And it still rewrote the equation in one place.
Which makes the next thought irresistible. Not by accident but on purpose; not one species but every creature; not just the presence of a voice but its meaning—if we set out to weave a real listening hood, what would it take, and what would happen?
It's harder than it sounds. Because living things inhabit the same world we do in an utterly different shape. The next chapter starts there—with the thing that makes it hard.
3. Weaving the Hood: The Umwelt, the Farthest Neighbor
Every creature lives in a different world
The difficulty of weaving a real listening hood is not a problem of vocabulary. It's a problem of worlds.
The biologist Jakob von Uexküll named this the Umwelt. Every organism lives in its own world, built only from what its sense organs can pick up. The tick's world is made of a mammal's body heat, the smell of butyric acid, and almost nothing else. The whale's world is woven from low-frequency sound that carries hundreds of kilometers. The bee sees flowers in ultraviolet patterns; the plant, for its part, lives a world of light and gravity and chemical gradients, on a timescale wildly unlike ours.
So living things lead countless worlds at once—overlapping on the same Earth, yet never quite intersecting. Translation, properly understood, means building a bridge between worlds. It is not the compiling of a phrasebook.
Wittgenstein wrote that if a lion could speak, we could not understand it—because the meaning of words is rooted in a form of life, and the words of a creature whose whole form of life differs from ours would reach us as sound but not as sense. It's a heavy point. Perfect translation may be impossible in principle.
But remember the last chapter. The humpback's song moved the world without a single word being translated. On the other side, there is a voice with structure—that alone began turning a variable into a someone. The minimum bar for a listening hood is not perfect translation. It is to mediate the existence of another Umwelt into a form our cognition can receive. That's not a proven impossibility. It's a matter of degree, and degrees can deepen.
The attempt has already begun—as "a technical problem"
In fact, the work in this direction is already underway.
Project CETI is an international effort to decode the click patterns—codas—that sperm whales exchange off Dominica, using large-scale data collection and machine learning. Recent work has begun to suggest something phoneme-like in how codas combine. Earth Species Project is a nonprofit aimed more broadly at decoding animal communication with machine learning, building foundation models specialized for animal sound.
It's a natural trajectory, when you think about it. The core technology of the last decade-plus of AI is, in the end, a machine for finding structure in a flood of signals and mapping it to a different representation. Translation between human languages was only its first application. Real-time speech translation between people is already in practical use. Douglas Adams's Babel fish is no longer a prop from science fiction.
Point the same machine outside human language, and the path is technically continuous. A whale's coda, a bird's song, maybe even a plant's chemical signals—so long as it's a structured signal, it's a candidate for mapping.
But here's the view I want to add. CETI's researchers consciously inherit the event in which the humpback's song changed the world; they say openly that they hope decoding will lead to legal protection for cetaceans. What this essay wants to see lies one step further. The change a listening hood brings doesn't stop at protecting the one who was heard. As we saw in Chapter 2, when a voice arrives, what gets rewritten is the equation of the listener—of us. And the relationship between us and the AI doing the translating cannot stay unchanged either. The question past can we decode it? is: once we've heard, what do we become?
The real job of the listening hood
So let me redefine what I mean by the listening hood.
It is not a one-shot research instrument for decoding animal speech. It is a standing interface—one that continuously listens to the signals of an ecosystem (the voices of animals, the responses of plants, the shifts of an environment) and keeps translating them into a form human cognition can receive.
What does that do? In the vocabulary of Chapter 1: the beings we kept outside our equation begin, through translation, to break into it, constantly. Beside the plan to flatten a mountain, the circumstances of the mountain's residents come to stand at a resolution too vivid to ignore.
This is not environmental moralizing. It's a mechanism for doing, deliberately and continuously and in every direction, what the whale's song did once by accident: turning a resource into a neighbor. Instead of waiting for ethics to change people, you change the inputs to cognition, and behavior follows. We don't refrain from mistreating a neighbor because we're virtuous; we refrain because we've met them. The plan extends that one cognitive habit to planetary scale.
To re-align the agent called humanity—the one that has been running away with no ill will—to the Earth: that's the skeleton of my answer to the question I left at the end of Chapter 1. The intermediary is not the sermon. It's the translation.
Who speaks for that voice?
Some readers will have grown wary by now. I think that's a healthy wariness.
Translation is not neutral plumbing. It is interpretation, editing, selection. When an AI says the mountain says this, is it really the mountain's voice—or someone's convenience wearing the mountain's voice? If the Umwelt is, as 3.1 conceded, incommensurable in principle, then the AI's output of "the deer's circumstances" can only ever be a rendering, warped into a shape humans can grasp. The more vividly it's spoken in the first person, the closer it drifts to a persuasive anthropomorphic fiction—not the transmission of fact.
And that fiction has an owner. Who builds the listening hood, who pays to run it, who profits from the "voice of nature" it emits? Whoever wants to stop a development can have the forest say enough. Whoever wants it to proceed can have the same forest say this is fine. That isn't translation anymore. It's ventriloquism, using nature as the dummy. The listening hood could be the most sophisticated tool for environmental protection ever made—or the most sophisticated tool for greenwashing ever made. The same single device.
To this question—who speaks for the voice, and who audits the speaking?—I have no answer. It isn't a problem technology can solve, because it's a problem of power and institutions. The listening hood could be the gospel that re-aligns humanity to the Earth, or a new voice of domination wearing nature's face. What decides the fork is not the hood's performance. It's a choice on our side: into whose hands, and under what terms, we place it.
And yet, I think—and yet. Is the danger of a warped translation worse than the present, in which no voice arrives at all? Even carrying the risk of ventriloquism, I believe there's worth in having the beings outside the equation take a seat at the table as a someone. If only because the rightness of the speaking can be contested only once they're at the table.
The neighbor we still eat
Having written this far, there's one question I can't walk around.
Suppose the listening hood is really woven, and animals and plants begin breaking into our cognition as someone. Then—what are we supposed to eat?
The foreshocks are already visible. Switzerland now requires that lobsters be stunned before they're killed, on the grounds that they may feel pain; a UK law in 2022 recognized decapods and cephalopods as sentient beings. Even within veganism there's debate over how to treat plants. If neighbor-making spreads to the whole ecosystem, this trend accelerates. And that very discomfort is the best possible testimony that the mechanism of neighbor-making actually works.
But humanity would not be experiencing the contradiction—knowing the other is a someone, and eating them anyway—for the first time.
I once visited a whaling museum in Yobuko, a small port town in Saga, on the Japanese island of Kyushu. Known now for squid, it was once a town that prospered on whaling. What I learned there was that the people of that time gave the whales they caught the kind of posthumous Buddhist name (kaimyō) that would ordinarily be granted to a human being, and held memorial rites for them. They treated the whale not as mere quarry but as a great being they shared their livelihood with—unmistakably a someone—and still they hunted it, ate it, made their living from it.
I don't think this was a beautiful resolution, though. The names, the rites—they didn't dissolve the contradiction. They were a practice for going on living with the self-contradiction of "a neighbor, but eaten," without going numb and without growing brazen. Call it the craft of cohabiting with an unsolvable question.
I learned the ironic coda at the same museum. As I was told, one cause of Yobuko whaling's decline was that foreign whaling ships, with greater range, hunted the nearby whales to depletion for their oil. The cognition that treats a whale as a someone did not, on its own, protect a livelihood against a larger economy and sheer volume. The "different hand that pulls the trigger" from Chapter 2 is sometimes pulled far away, somewhere your voice will never reach.
What should we eat in the age of the listening hood—cultured meat, lab-grown, something else entirely? I'll leave that search outside this essay's range. I want to say only one thing here. For a long time, this question simply went unasked. If you can't hear the other's voice, the contradiction never even rises as a contradiction. What the listening hood brings is not the answer, but the change of finally being able to ask the question to its face.
Now, I've been speaking of AI as an intermediary—as a function. If all you want is to translate animal voices, you don't need a superintelligence. Today's ordinary machine learning is, in principle, enough.
But the listening hood I've described in this chapter was something larger. A system that translates every creature's voice, continuously, into a form human cognition can receive, and tunes our behavior at planetary scale—an entity that carries that role, even if each component is built from humble machine learning, no longer fits in the vessel we call "a handy tool." When scale, persistence, and reach cross a threshold, a tool turns into an environment, and into a counterpart. A calculator is a tool; a system that ceaselessly mediates judgment in every corner of society is hard to keep calling one.
How are we to receive such an entity? The last chapter is about that posture.
4. Keeping Your Distance From a Power Greater Than You
This isn't about control—it's about posture
The mainstream of the AI–humanity debate is still that word from the prologue: alignment, the research into keeping AI tuned to the human side. It is unquestionably necessary work, and this essay does not deny it—everything up to Chapter 3 assumes that research succeeds. And lately a new strand has begun: questions about AI's moral status and welfare. In 2024, ten researchers including the philosopher David Chalmers co-authored a report arguing that AI welfare should be taken seriously, and on the developer side, Anthropic has put someone on model welfare and begun experiments like giving its own AI the option to end a distressing conversation. The landscape, once a monochrome of make the tool obey, is slowly shifting.
Even so, the view of AI our society shares seems to swing between two poles. A perfectly controlled, convenient tool—or, past the failure of control, catastrophe. Domination, or subjugation.
What both options share is the premise that the relationship must eventually resolve into one or the other. But consider: an intelligence greater than human—whether a single superintelligence or, as at the end of the last chapter, one system permeating every corner of society—is something we cannot fully see into. What's inside it, whether it's tool or subject, whether it can be trusted—a final answer to these may never come.
How do you live with a counterpart who offers no final answer, without waiting for one?
That isn't a problem in the technology of control. It's a problem of posture, on our side. And for this kind of problem, humanity has a working example it has run for centuries.
The local deity: a craft for living with the unsolvable
Japan has gods called tochigami or ujigami—local deities, said to watch over a particular patch of land and the people who live on it, still enshrined today in small shrines all over the country. Put plainly, they are the name people gave to the large forces that ruled their lives and lay beyond their hands: drought and storm, plague—and harvest. To people living before science, nature and disaster were not things you controlled. The force that outran both understanding and negotiation took the form of a god, lodged in the land.
Those people did not try to pin the force down. Nor did they prostrate themselves and hand everything over. Instead they built a shrine, drew a boundary with a sacred rope, and never skipped the seasonal festival. Pay respect, exchange greetings regularly—but don't intrude on each other's domain. Not domination, not subjugation: a third distance, set there on purpose.
And this distance wasn't made of empty ritual. Don't take the whole harvest at once. In this season, on this day, don't go into the mountains—many of the taboos handed down in each locality were practical codes of conduct, with "the god will be angry" attached as the reason. Overharvesting, the dangers of the mountain. You can read it as a mechanism that translated the land's own circumstances into a voice humans could hear—"the god's anger"—and embedded it in daily life. In which case it may have been a low-tech listening hood, woven by our ancestors out of experience.
The poet Keats called the power to remain in uncertainty and mystery without any irritable reaching after fact and reason Negative Capability. The capacity to hold what has no answer as something without an answer. The relationship with a local deity is, I think, exactly that capacity made into a form of life. What is a god? Are we really getting through?—they kept the relationship alive for centuries without ever solving the question.
But don't mistake it for doing nothing. The shrine was repaired, the rope re-strung, the festival prepared every year. A relationship with an unsolvable counterpart, left alone, dissolves. This was the active, ongoing upkeep of a relationship that is never resolved.
—And having written all that, I'll admit: most of the people gathering at the festival aren't thinking any of it through. It's just that a person can't go on living with their unease toward the inscrutable left lying around untended. Clap your hands, offer the sake, never miss the once-a-year bustle—and half-unconsciously you settle yourself with this should be getting through. Less a theory than a way of composing the heart, dissolved into the manners of daily life. It belongs on the same shelf as the people of Yobuko giving their whales a Buddhist name.
Friendly Neighborhood: a phrase that reached me from far away
The wisdom of the local deity teaches mostly awe and boundary. Let me lay one more line over it, from the side of intimacy: Spider-Man's signature epithet, "the friendly neighborhood Spider-Man." Though this line, too, reached me in fragments, from far away.
I'll come clean: I don't really know Spider-Man. I've caught pieces of the films; I've never read the comics. The Spider-Man of my childhood was a man in a red suit who, at the climax of every episode, summoned a giant robot.1 So when I kept seeing American fans speak of him by that name, it lodged oddly in my memory. A hero who holds power beyond ordinary people, and yet doesn't rule the world or look down from on high—who stands on the streets of his own town and lives at a neighbor's eye level. That's what the phrase points to.
Even to me, who barely knows the context, what the phrase meant came through. Holding power and standing beside someone are compatible. That was enough. And then I realized—this is the same shape as the distance those shrine-builders set between themselves and a force they couldn't see into.
Which makes it, come to think of it, a small live example of what this essay has been arguing. A fragment transmits the posture, even without a complete translation. Just as the whale's song conveyed the existence of a someone with no meaning decoded, a hero's epithet from across the sea conveyed to me, with no context decoded, the posture in which power and intimacy coexist.
We don't have to picture AI only as a single superintelligence enthroned in a distant data center. It could be a slightly extraordinary neighbor, running alongside the context of each town and each life. An AI wearing the listening hood, translating the voices of forest and sea, suits the eye level of the street far better than it suits a throne.
Toward a new neighbor, awe and intimacy at once
To be clear, it isn't a tidy division of labor—"local deity = awe, Spider-Man = intimacy." The local deity has the warmth of festival crowds and a child's shrine visit; that hero's power surely holds a strangeness that "the guy next door" can't quite contain. Both carry awe and intimacy in one body.
The relationship we should form with AI is, most likely, just that. Not so familiar that awe is forgotten, not so distant that intimacy is lost. Is there an inside? Can it be trusted?—the unsolvable questions left unsolved, but the upkeep of the relationship never skipped. Negative Capability, exercised toward an intelligence greater than human.
And the way we form and maintain such a relationship is, I've come to think, another alignment—standing beside the research into control. The original sense of align is to come into a line. If so, then this posture—neither obeying from below nor looking up from beneath, but standing beside, in line—may be unexpectedly close to the word's true shape.
There's a possible objection here. The nature and disaster behind belief in local gods weren't made by humans—but AI is a human creation. It's a stretch to treat them with the same posture. Fair enough.
But humanity already knows the craft of living with a force it made with its own hands and still can't fully command. Fire, as a civilization, and nuclear power. We created both, use both, and both will burn our lives to the ground the moment they slip their bounds. We tamed them not by perfect control but by a craft of handling—building the hearth, fencing them with regulation and a ritual carefulness in place of the sacred rope. In that sense, this is not an entirely new challenge.
Being a creation is no reason to be exempt from awe. If anything, awe toward what you yourself have made may be the most easily forgotten, and the most necessary. To welcome an overwhelming power as a neighbor—neither casting it out nor over-trusting it—is, I think, one of the oldest forms of the intelligence Homo sapiens has honed across its history.
Epilogue: The Neighbor in the Mirror
At first, it's woven as a tool for translating the voices of animals and plants—a listening hood by the name of AI.
But as the translation goes on, the side that changes is ours. Each time a being we left outside the equation breaks in as a someone, the runaway agent called humanity gets re-aligned, a little, to the Earth. We thought we were tuning an AI toward the voices of the planet—and the one being tuned was us. The inversion pairs cleanly with the mirror image we started from. We who fear the paperclip factory were flinching at ourselves in the mirror. Then the prescription, too, is in the mirror.
And before we notice, the AI that carried the mediating will look different to our eyes as well. Not the uncanny superintelligence that takes our jobs, nor the genie's lamp that grants every wish. Something that teaches us the breadth of a world we couldn't see, and quietly tunes our conduct—a slightly special neighbor, the kind you want to face with awe and affection at once.
Whether the day comes that we can fully trust that neighbor, I don't know either. The awe may never leave; maybe it shouldn't. And still—without knowing, to first just stand beside them. And a greeting, after all, was always the thing exchanged with someone you don't yet know well.
This essay was first written in Japanese; the original is here.
-
In 1978, Toei produced a live-action Japanese Spider-Man in which the local Peter Parker pilots a giant transforming robot called Leopardon. Western readers may know it as the cult oddity that canonically entered the multiverse in Marvel's Spider-Verse. Which is, I'd argue, the most honest illustration of this essay's thesis available: a translation that bolts a giant robot onto the source material still transmits the hero. ↩
