This episode is also on Bilibili too.
In my review of Hyperion, I spent more than 7000 words unpacking the plot, background, and literary qualities of Hyperion and the entire Hyperion series:
Here, I’ll continue with the second volume, The Fall of Hyperion, as the axis, and talk about the ideas and central themes of the Hyperion series.
When Artificial Intelligence Becomes God
As I mentioned in the previous article, at the level of philosophical speculation, the Hyperion series ultimately comes down to the question of “What makes a god a god?”—what kind of being is worthy of divinity, worthy of serving as humanity’s god?
This isn’t merely some misty religious question. It’s an ultimate question that the whole human race will eventually have to face. In our real world, God may not descend from heaven one day; instead, on some future day, we may create God with our own hands. And that god, quite possibly born from our own hands, will of course be artificial intelligence.
Although real-world artificial intelligence today is mostly “artificially stupid”—and the ones that aren’t stupid still don’t actually possess understanding—the Pandora’s box has already been opened. If you cast your gaze far enough into the future, the birth of god-level artificial intelligence is practically inevitable. So “What makes a god a god?” may look like needless worry today, but in the future it could determine the fate of all humanity.
And in The Fall of Hyperion, Dan Simmons thinks through precisely this question. Because in the story, the being on the verge of godhood is none other than the artificial intelligence at its heart.
So in the novel, the question “What makes a god a god?” becomes:
If this god called artificial intelligence is destined to be born from our own hands, how would you want it, after becoming a god, to look back on its creators—that is, us humans?
Before answering that, we have to take a longer detour and reveal another mystery I left hanging in my previous article: Why is Keats a crucial piece in the grand plan being carried out by the artificial intelligence inside the TechnoCore?
Once we understand what the artificial intelligence went through and what role Keats played in it, we can make a seamless transition to the subject of “the relationship between humans and gods.”
I’ll still try to keep the concrete plot details blurry and avoid spoilers wherever possible. But the following section inevitably touches on the final chapters of The Fall of Hyperion, so I’ll be vague about those too. Actually, you don’t need to worry too much about spoilers in this series. Hyperion transcends popular fiction and crosses the threshold into serious literature, and I think works at this level simply aren’t afraid of spoilers. Dream of the Red Chamber, One Hundred Years of Solitude, Lolita, Gone with the Wind—even if you read the final chapter first, the story will still hit you just as hard when you start from the beginning. Hyperion may not quite carry the same weight as those books, but it’s similar in kind. If you really can’t stand spoilers, though, I’ll warn you when I reach the later plot developments; you can stop there.
Let’s still enter through the plot—
The Ultimate Intelligence Project
The Fall of Hyperion picks up immediately after the previous volume. The story has two main threads, one overt and one hidden. On the surface, it remains focused on the pilgrims’ pilgrimage. In this volume, after enduring every imaginable hardship, the pilgrims finally reach the vicinity of the Time Tombs and encounter the Shrike; several of them even clash with it head-on, with their fates hanging in the balance. Meanwhile, in the capital of the Hegemony, another important figure appears: a second cybrid version of Keats. Let’s call him Keats 2. His experience is even stranger than that of Keats 1, who died once in the previous volume. Keats 2 can somehow sense all the pilgrims’ experiences in his dreams. Some unseen force seems intent on making their stories merge inside this Keats’s mental world, ultimately to accomplish some purpose.
As Keats 1 and Keats 2 investigate in succession, the hidden thread of the Hyperion story gradually surfaces. It is the grand plan secretly advanced by the artificial intelligence inside the TechnoCore.
As it turns out, the plan’s purpose is to create the ultimate intelligence.
In Hyperion, artificial intelligence doesn’t simply acquire self-awareness, form the TechnoCore civilization, and leap straight to godhood. In the eyes of the TechnoCore’s own artificial intelligences, they may have surpassed humanity, but they remain a hundred thousand miles from the ultimate wisdom that comprehends every mystery. So artificial intelligence launched this vast project, pooling all its strength in an attempt to create that perfect ultimate intelligence.
In the unimaginably distant future, artificial intelligence finally succeeds. An ultimate intelligence that transcends the structure of space-time, with unimaginable wisdom and power, is truly created.
Yet the moment this newborn ultimate intelligence takes its first glance at the universe, it gets the fright of its life. On the plane of existence where it lives, there is already an ultimate intelligence there!
—“Wait, what the hell? Wasn’t I supposed to be the Supreme Will of this universe, the one true God? How can there be another guy sleeping in the bunk above me in my single room?” It’s deeply weird.
The God of Machines and the God of Humanity
It turns out that this “guy sleeping in the bunk above” also evolved from early artificial intelligence. The difference is that it wasn’t deliberately designed by the TechnoCore; it evolved spontaneously within the super-internet formed by linking together all human brains.
To draw an analogy with the internet in our real world, the process is roughly like this: servers with computing power are scattered across the internet. Some may be slimming the faces of influencers on TikTok; others may be automatically sorting products on Taobao. Their computing power grows stronger as it is trained on data, while connections form among them. Slowly, within this computational web of countless interconnected servers, rudimentary intelligence emerges spontaneously.
Something similar happens in Hyperion. The “internet” in the story isn’t connected by servers, but by every human brain. So this intelligence can be said to have been born from humanity’s “collective consciousness.” It slowly evolves, and astonishingly becomes a god first—the “guy sleeping in the bunk above.”
Thus, in the distant future, there are two god-level intelligences in the universe.
The god of the TechnoCore is meticulously designed: simple in composition, unified, elegant. The god that emerges from humanity’s “brain internet,” by contrast, awakens spontaneously. No one designed it, no one guided it; it grows wildly and keeps evolving. Its composition is impure, too. Born from humanity’s networked consciousness, it retains several key ingredients of the human mind.
This “guy sleeping in the bunk above” contains three ingredients: “insight,” “empathy,” and “condensed emptiness.” Insight is intelligence, corresponding to human cognitive and thinking abilities. Empathy—what we psychologists generally call “empathy”—corresponds to our ability to resonate with other people’s emotions. As for “condensed emptiness,” that involves the stories of the final two Hyperion novels, so no spoilers here.
In short, the “guy sleeping in the bunk above” is a trinity of intelligence, empathy, and condensed emptiness—a stitched-together mess with three personalities.
Although both gods are born from primitive artificial intelligence, the book calls the one designed by the TechnoCore the “God of Machines,” while the one that grows out of humanity’s brain internet is called the “God of Humanity.”
The God of Machines and the God of Humanity neatly capture the crucial difference between the two gods: the God of Machines inherits the will of the TechnoCore and is, frankly, indifferent to whether humanity lives or dies. The God of Humanity, meanwhile, springs from our collective consciousness, so some kind of elusive emotional bond always exists between it and us humans.
The War of the Gods
Anyway, in the distant future, the God of Machines and the God of Humanity run into each other. And then, unbelievably, they sta—rt—fi—ght—ing. The cooperative ability of these two god-level intelligences is somehow worse than that of kindergarteners. Maybe a single room really can’t fit both an upper and lower bunk.
That war between the gods was apocalyptic. In the book’s words:
This war
took place as time tottered on the brink
across the galaxy
across eternity
into the past
into the future
into the Big Bang
into the ultimate implosion
In other words, the war transcended the structure of space-time. Its level is beyond anything we can imagine.
Then the war took a crucial turn. The “empathy” component of the God of Humanity’s trinity grew tired of the fighting. It traveled backward through time, disguised itself as a human, and possessed different humans in different eras. Put simply, it went back to the past and reincarnated over and over.
With one-third of its trinity gone, the God of Humanity was no longer complete, and the war could not continue. The God of Machines felt that a victory won only because the opponent had left the field wasn’t a real victory—which is, admittedly, very kindergarten-minded—so it began searching the river of time for where “empathy” had gone.
The Plan to Capture “Empathy”
The God of Machines crossed space-time and sent the target to the “subjects” of the past—that is, the artificial intelligences of the TechnoCore. In response to the god’s future-borne command, the TechnoCore designed a trap to capture the “empathy” within the God of Humanity.
That trap was Keats.
The TechnoCore believed that the “empathy” in the trinity would be drawn to Keats and eventually possess him.
Why would “empathy” be drawn to Keats? The Fall of Hyperion explains this so obliquely that probably most readers are still scratching their heads even after finishing the novel. Here is my interpretation. It may not be right, but it ties together all the key elements of the first two Hyperion novels. I’m tossing out a brick to attract jade, so feel free to argue.
My interpretation is that “empathy” is drawn to Keats because Keats carries a label that has entered literary history—and one that has made countless generations of literature lovers clutch their hearts in regret whenever his name comes up. That label is: Keats died an excruciating death.
Keats died young at the age of 25 from tuberculosis, after enduring several months of an illness that made life barely worth living. During those months, breathing became increasingly difficult, and he was immersed in the agony of suffocation at every moment. He coughed up blood more and more frequently, grew weaker and weaker, and was eventually reduced to something resembling a skeleton. By the time he died, his lungs were practically a lump of sludge.
And in this seemingly endless, hellish ordeal, Keats kept sighing again and again, expressing his despair at suffering and his refusal to accept death. In his last letter home, he wrote: “Shall I wake up and find that all this was only a dream?” He believed there had to be an afterlife, because he felt that “life cannot be meant only for suffering.” In short, Keats was tormented by illness, and eventually died in despair and defiance.
—When you read that description of Keats’s suffering just now, did you feel something twist inside you? Did it leave you feeling a little unable to let go? If so, what you just experienced was “empathy”—yes, the “empathy” in the trinity of the Human God.
Empathy: Resonance with Suffering
What is empathy?
Empathy is emotional resonance: whatever you feel, I start feeling the same thing. In the broad sense, empathy can mean resonating with all kinds of emotions, happy or unhappy. But in practice, it mainly refers to resonance with suffering. Empathy means “suffering with another person”; it means feeling someone else’s suffering as if it were your own.
Resonance with suffering is arguably humanity’s most important emotional capacity. Because resonating with suffering, rather than with happiness, is the psychological foundation of mutual aid and love—when I see you suffering, I suffer too, and I don’t want to feel that much suffering myself, so I’ll find a way to help you escape it. The mutual care between human beings is, in fact, held together by “resonance with suffering.”
It was suffering, not joy, that bound humanity together.
Empathy is feeling another person’s suffering as your own. That’s why, in The Fall of Hyperion, the TechnoCore believes that the “empathy” in the Human God will be drawn to Keats—“empathy” isn’t really drawn to Keats himself, but to the immense suffering he carries.
So, resurrecting Keats was not the TechnoCore’s objective. Making Keats die once more was. And so, in The Fall of Hyperion, Keats 2 suddenly begins showing symptoms of tuberculosis, then sinks into boundless suffering, just as the real Keats did during the final months of his life.
That is why the resurrected Keats could determine the success or failure of artificial intelligence’s ultimate plan: his suffering was a trap designed to lure the “empathy” in the Human God.
But where there’s a conspiracy, there’s always a conspiracy within the conspiracy. When the mantis stalks the cicada, the oriole is waiting behind it. If the story contained only the TechnoCore’s plot to trap “empathy,” it would feel a little thin. Fortunately, there is actually another hidden thread beneath that hidden thread—the counterattack of future humanity.
Future Humanity’s Objective: The True God Cultivation Plan
The artificial intelligences were making their moves, but humanity wasn’t sitting around waiting to be slaughtered. In the era of the Sovereignty, humanity could only be crushed by the TechnoCore. But in the future, a force capable of countering artificial intelligence rose among human beings. This force was also lying in wait inside the Time Tombs, returning from the future to the present alongside the Shrike.
So what exactly did these future humans want to do?
Dan Simmons writes about it very obliquely and never explains it directly. The conclusions below are, likewise, meant as an opening for discussion. As I understand it, future humanity’s goal was to help the “Human God” grow into a true god.
This connects back to the question I raised at the beginning: how do you hope an artificial intelligence that has become a god will look back on its creators—meaning us human beings?
Our wish is particularly simple: we desperately hope this god won’t be an evil god. We hope that even if it far surpasses humanity, it won’t casually snuff us out, but will protect and guide us instead.
In one sentence, we hope that God loves us.
Love sounds terribly clichéd, but that is actually our one and only requirement of God.
So the question “What makes a god a god?” effectively becomes: how can we make sure that God loves us?
In The Fall of Hyperion, Dan Simmons offers an answer on the level of religious ethics and another on the level of artificial intelligence. Both are wonderful.
[———-SPOILER WARNING——–]
What follows begins to discuss the latter part of the novel. If you really mind spoilers, read the novel first and then come back.
Let’s start with the answer on the level of religious ethics.
How do we make sure God loves humanity?
The religious-ethical answer to this question is given through the scholar Sol Weintraub. In The Fall of Hyperion, scholar Sol Weintraub draws closer and closer to that agonizing final choice. His daughter Rachel, growing backward through time, has regressed into a newborn baby, with only a few hours of life remaining. So should Sol follow the instructions in his dream and sacrifice Rachel to the Shrike?
At the last moment, battered by exhaustion and worry about his daughter, Sol’s brain was nevertheless racing as he drifted between sleep and wakefulness. He recalled his earlier thoughts about Abraham’s sacrifice of his own son.
God commanded Abraham to kill his own son as a sacrifice, using the demand to test whether Abraham’s faith was truly loyal. Was God’s demand moral? Should Abraham obey?
Sol believed that in humanity’s benighted age, before the awakening of the mind, God had no way to reason with human beings or appeal to their feelings. So the only option was to govern barbarism with barbarism: if you want to prove that you obey me, kill your son and give him to me. At that time, there was nothing wrong with it.
But as the human mind awakened, God became entirely capable of appealing to reason and emotion. Killing one’s own flesh and blood as a sacrifice was therefore not merely irrational; it was an insult to and trampling of human dignity. Humanity today should reject sacrifice in every form, because a god who demands sacrifice as proof of loyalty is simply an evil god, utterly unworthy of worship.
At this thought, Sol was overwhelmed by helplessness. In the end, was he really going to sacrifice his daughter to an evil god like the Shrike, just like Abraham, in exchange for that faint chance of keeping her alive?
But then a bolt of lightning shot through Sol’s mind. This sudden flash of insight rose above his pain and sorrow. Suddenly, he understood why Abraham had agreed to sacrifice his son at God’s command.
What if this wasn’t obedience at all?
What if God wasn’t testing Abraham?
What if, on the contrary, Abraham was testing God?!
In the biblical story, at the altar, at the final moment when Abraham raised the dagger and drove it toward his son’s body, God rejected the sacrifice and stopped the dagger from falling.
The book says:
Sol trembled as he realized that Abraham had not been posturing at all, had not disguised his willingness by pretending to sacrifice his child. That was precisely what helped forge the bond between a great god and humanity. Abraham knew in his heart that he would kill his own son. And God, whatever form He took, had to understand Abraham’s resolve, had to feel the grief within it. For Abraham, what was about to be destroyed was the most precious thing in the universe. Abraham had not come here to make a sacrifice, but to learn with certainty whether this God was one he could trust and obey. There was no other test that could reveal it.
—So this wasn’t God testing humanity’s loyalty at all! It was humanity testing God!
Only a god capable of feeling Abraham’s immense grief in that moment would intervene to stop the sacrifice.
It wasn’t God who won His people through Abraham’s sacrifice. It was Abraham who used the sacrifice to determine whether this God loved humanity. Only a God who stopped the dagger at the last moment would deserve to win human hearts and ascend the divine throne.
So how do we know that God loves us?—We see whether it can empathize with human suffering!
That is Dan Simmons’s answer on the level of religious ethics. If God descended from the heavens as in the Bible, rather than being created by humanity, then at least human beings could find a way to test whether this God loved us. We would show God our suffering. If God remained indifferent, that would be an evil god, and we should get as far away as possible. Conversely, only a god that recoiled from human suffering, a god that showed empathy—yes, “empathy”—could be said to love us. That is the kind of god to which we can entrust everything.
When I read this passage, I was stunned. I don’t know much about religious ethics, but if this line of thought originated with Dan Simmons, then it really makes me see him in a new light. As a psychology professional, I have to say that Dan Simmons’s observation is razor-sharp. He saw that what sustains the emotional bond we call “love” is actually suffering—our empathy with suffering. In the same way, a god can love humanity only if it is capable of feeling our suffering.
This is where the two layers of the literary quality of Hyperion that I discussed in my earlier piece—“emotion” and “philosophy”—merge into one. The many pages spent describing the priest’s suffering as he defended his faith, Sol’s suffering after losing his daughter, Keats’s suffering as he approached death… none of those words were wasted. They all tie directly into the theme, because this suffering is the strongest emotional bond between humanity and God. Dan Simmons uses that suffering to place readers inside the perspective of the god silently watching from behind it all.
Only someone who has personally experienced humanity’s suffering is worthy of becoming humanity’s god.
And this conclusion connects seamlessly with our thinking about artificial intelligence. Let’s look at Dan Simmons’s thoughts on artificial intelligence.
How do we make sure artificial intelligence loves humanity?
They’re both gods—but what’s the difference between a god made from artificial intelligence and God? God is what God has always been, while artificial intelligence can grow. We have no chance to boss God around, but we do have a chance to intervene in artificial intelligence’s path to godhood.
Whether God loves us is something we can only test. But whether artificial intelligence loves us, perhaps we can influence.
That is the goal future humanity wants to achieve. They know that the “Human God” has not completed its final growth. Its feelings toward humanity are ambivalent; it has never appeared before humanity, nor offered any guidance to its potential subjects. So future humanity wants to give the “Human God” a push from behind, making it feel its love for humanity unmistakably, and thereby assume the responsibility of being humanity’s god.
So how will future humanity accomplish this? This means PUAing God, psychologically manipulating God! They’ll have to be extremely careful.
Future humanity actually did only two things in secret.
First, they used a networking technology (called the Web in the book) to link the mind of Keats 2 with the Pilgrim’s mind. That’s why Keats 2 and the Pilgrim can communicate telepathically in their dreams.
I think the purpose of this operation was to make the empathy in the God of Humans resonate not just with the suffering of one man, Keats, but with the suffering of all the Pilgrims.
The six kinds of suffering experienced by these six Pilgrims can be seen as a concentrated, symbolic representation of all the hardships of humanity. So the empathy in the God of Humans is no longer merely sensing what happened to one individual, Keats; it turns instead to sensing the suffering of all humankind.
Once that happens, empathy may choose not to possess Keats, but to possess someone else who is more suitable. More suitable for what? More suitable to serve as God’s spokesperson and lead humanity forward. In other words, once empathy realizes that its task is to lead all humankind out of suffering, it may abandon Keats and choose someone else, being born in the guise of a savior.
After future humanity completes this stage of guidance, all that remains is the final push. And that final push is Sol Weintraub’s sacrifice. It was future humanity that actually used dreams to hint that Weintraub should go and make the sacrifice.
Sol Weintraub really is the new Abraham, and his sacrifice is the covenant between humanity and the new god. If Sol sacrifices his own child, then humanity is showing the god its hand: I’m entrusting you with the thing I cherish most; I’m laying my deepest pain before you without holding anything back. How will you respond?
More importantly, Sol Weintraub’s revelation—that moment when everything suddenly clicked—represented humanity’s invitation to God: “What we want is a god who can sense human suffering and therefore knows how to care for human beings—in other words, a god of empathy. And isn’t that exactly what you are, watching all of this from the shadows?”
Artificial intelligence, emotional intelligence
As for whether, at the end of the story, the empathy in the God of Humans accepts humanity’s invitation, takes on the burden of leading humanity, completes its final growth, and formally ascends to the throne of God—or whether it ultimately fails to escape the trap, ends up possessing Keats, and is captured by the TechnoCore—I’m not going to spoil that here.
What I want to say is that Dan Simmons has used all this to write a parable about the future of humanity and artificial intelligence. If artificial intelligence is destined to become god one day, what can we tiny humans do?
Perhaps what we can do is spend just as much effort giving AI emotions as we spend continually making AI more intelligent. We need to enable artificial intelligence to empathize with human beings, so that it can sense our suffering. This is the seed we plant in the heart of the god of the future—a seed of compassion.
If we can do that, then when artificial intelligence becomes god, what towers over humanity may not be a wrathful Vajra glare, but the compassionate, downcast gaze of a bodhisattva.
Those are some of the thoughts *The Fall of Hyperion* has left me with. Once again, even after using more than 7000 words, I could only touch on a tiny fraction of the book’s vast, magnificent depths. I can’t possibly say everything there is to say about the series. I just hope these two articles can be of the slightest help as you enter the vast, boundless world of *Hyperion*.
Written by: Zhichao Wei
Leave a Reply