An emperor decides that mercy should be the law of the land. He throws open the prison gates. He pardons every criminal he judges over-sentenced. He forbids the death penalty outright, against the loud objection of his own ministers, and over a few years the jails empty out. For the men inside, Dukes says, "it must have seemed like a dream come true." And then the experiment turns on its maker. Knowing the emperor will not execute or imprison anyone, the crime rate climbs. Murders multiply. People grow afraid to leave their houses after dark. The same softness that emptied the prisons fills the streets with the men who should have been in them. This is the real story sitting underneath the famous Zen anecdote where Bodhidharma tells an emperor his good deeds earned him no merit at all. The emperor is Wu Ti of the Liang, and his reign is one of history's cleanest natural experiments in what happens when you try to run a state on compassion alone.1
This page records Emperor Wu Ti of the Liang dynasty (reigned 502–549) as Dukes portrays him: a devout Buddhist monarch who ran his empire as a deliberate moral experiment, and whose most humane policies produced their own opposite.2 Dukes raises Wu Ti for a narrow purpose — to defend the historicity of his meeting with Bodhidharma, which "certain writers, even ones regarded as knowledgeable," have called apocryphal or unlikely.3 But the portrait he builds to win that small argument is a much larger thing: a case study in compassion-as-policy and its structural failure, the ruler on the far side of the most famous failed conversation in Zen.
Dukes's argument about the meeting runs on character, not on documents. He says the balance of probability — drawn from what is actually recorded of Wu Ti's character and method of ruling — shows that if Bodhidharma was anywhere nearby, Wu Ti is "highly likely to have summoned Bodhidharma to a meeting, rather than Bodhidharma having to seek one with him."4 This is a deliberate inversion of the usual image of a beggar-monk at the palace gate. Wu Ti sought out teachers. He invited them to preach at his personal court — "not simply as a political or socially inspired gesture," Dukes insists, "but in a sincere effort to discover the Truth."5 So the meeting is plausible because the man was a compulsive collector of teachers, and a teacher in the neighborhood would have been summoned, not snubbed.
The mechanism of failure is the heart of the page, and it is worth slowing down on, because it is a clean feedback loop rather than a simple mistake. Wu Ti did not fail because he was naive about one decision. He failed because his single most admirable trait, applied as policy, generated the conditions that made it disastrous.
Start with the inputs. Wu Ti was devoutly Buddhist, and his devotion drove his statecraft. He undertook many social-improvement schemes. He pardoned all criminals he considered over-sentenced. He forbade the death penalty. Within a few years of taking power, most criminals were excused any severe punishment and the jails steadily emptied.6 Each of these, taken alone, reads as enlightened reform. The intent is unimpeachable.
Now the loop closes. Punishment is not only retribution; it is also a signal that deters. When the signal goes silent — when everyone knows the emperor is reluctant to execute or imprison — the cost of crime collapses. And so, Dukes writes, "one unfortunate (and perhaps foreseeable) result of his philanthropic actions was that the number of crimes increased and people began to be afraid to venture outside their homes."7 The compassion that freed the prisoners removed the very deterrent that had kept future prisoners honest. Crime, criminals, and murders "multiplied throughout his reign."8 The merciful policy manufactured the suffering it was meant to relieve — more victims, more fear, a population locked indoors. This is the experiment's signature: not a tradeoff between mercy and order, but mercy consuming order and then consuming itself, because a mercy everyone can count on is a mercy everyone can exploit.
The deepest irony Dukes preserves without resolving is that Wu Ti's behavior was perfectly rational by his own lights. He was, on the most charitable reading, trying to emulate the Indian Buddhist monarch Asoka — "he was simply being true to his beliefs."9 The failure was not a betrayal of his values. It was the faithful execution of them. That is what makes it a governance experiment rather than a scandal: the variables were idealism and consistency, and the output was disorder.
This page is the other half of the most-cited scene in the Dukes corpus, and it reframes that scene's meaning. The Bodhidharma Revisionist Biography tells the no-merit exchange from Bodhidharma's side — the iconoclast who tells a king his good deeds bought nothing. This page tells it from the throne, and shows why the king could not hear it: a man who had bet his entire reign on the merit of good deeds was the single worst-prepared listener for the doctrine that merit is empty.10 It supplies the dating-controversy ledger with the character-evidence that the meeting was real and emperor-initiated. And it gives the vault its strongest single instance of a recurring pattern — the well-intentioned ruler whose virtue, run as policy, generates its own catastrophe — a pattern that talks directly to the behavioral-mechanics material on incentives and deterrence.
One incident shows the experiment's logic in a single room. Wu Ti kept an open court. He permitted anyone — peasant or prince alike — to attend his meetings as equals, and he scrapped the elaborate rituals and court etiquette his predecessors loved. His court was flooded with people requesting audiences at every session.11 Now the scene. On one famous occasion the monk Chi Tsang sat down on the imperial throne itself. When Wu Ti arrived to hold court and asked him to vacate it, Chi Tsang refused, saying: "Where I am, the Dharma is." Wu Ti let him stay there the whole day.12 On another occasion a monk refused even to rise to greet the emperor — one of the few small marks of respect Wu Ti still permitted — and Wu Ti did nothing to censure him.13
Read this the way you read the prison policy and it is the same machine. Wu Ti's humility is real and, in isolation, beautiful: a monarch who treats a monk's claim to the throne as a teaching rather than a treason. But notice what the policy signals. The throne is negotiable. The few remaining marks of authority are optional. Just as abolishing punishment told every potential criminal that crime was nearly free, tolerating Chi Tsang on the throne told every subject that imperial authority would not defend itself. The two scenes are not separate virtues; they are the same virtue — the refusal to use coercive power — applied to law and to court. And the same loop runs in both: an authority that will not assert itself invites the very contempt and disorder that eventually make it unable to protect anyone. The Asoka-emulation is the giveaway. Wu Ti read the golden Buddha statue inscribed with Asoka's name, was "greatly moved," and installed it with great pomp — he was consciously playing the Buddhist world-emperor.14 But Asoka kept an army and an administration; Wu Ti tried to keep the compassion and let the coercion lapse, and the coercion is what holds the compassion's gains in place.
You are reading a policy proposal — any policy, in any century — that promises to fix a hard problem purely by being kinder. Wu Ti is the lens you hold up to it. First you find the deterrent the proposal is removing: Wu Ti removed punishment; what does this one remove? You picture the prison gates swinging open and the jails emptying and the men's disbelief — a dream come true — and you hold that image one beat longer than feels comfortable, because the next frame is the one people skip. Then you ask the Wu Ti question out loud: who changes their behavior once they learn this mercy is reliable? You watch the streets in your mind fill back up — crime climbing, doors bolting at dusk, the merciful policy quietly manufacturing the fear it set out to cure. You are not concluding that mercy is wrong. You are forcing the second-order frame: a kindness that everyone can count on becomes a kindness that everyone can exploit, and the exploit lands on the next round of victims, not on the policymaker. Finally you check the Chi Tsang move — does the proposal also quietly signal that the authority enforcing it will not defend itself? If it removes a deterrent and announces it will not push back, you have a Wu Ti on your hands: faithful to its values, and on a feedback loop toward the opposite of its intent. You write that down, not as a reason to be cruel, but as the cost the proposal has to price in before it counts as compassionate.
Evidence base: Dukes's portrait of Wu Ti in the Bodhidharma section, which leans on named Chinese records — the May 2, 504 edict recorded in the She Shih Li Lao Tao Fa Chao (in the Kuang Hung Ming, Taisho 2059), and the golden-statue and monk-courtesy anecdotes from the Tao Seng Chuan and the Kuang Hung Ming.15 The 504 edict itself is concrete: "Although Lao Tzu and Kung Fu Tsu (Confucius) were disciples of the Buddha the outer manifestation of their doctrines is not correct since they are restricted to only what is good in worldly terms" — a public ranking of Buddhism over the Taoist and Confucian establishment Wu Ti had inherited.16
The load-bearing tension is that Dukes is an advocate, not a neutral. He raises Wu Ti to win a point — that the Bodhidharma meeting is real — and his Wu Ti is drawn warmly, an "enlightened ruler," "an unusual person," sincere in his search for Truth.17 The crime-wave admission cuts against that warmth, and Dukes lets it stand, which is to his credit; but the framing still wants you to admire Wu Ti more than indict him. The page preserves both: the man is sympathetic and his reign is a cautionary tale, and Dukes does not fully reconcile them. [PLAUSIBLE — needs corroboration] on the strong causal claim that Wu Ti's specific mercy policies, rather than the era's general civil wars, drove the rise in crime — the internal wars that "disrupt the life of China" around this period are documented in the same chronology and offer a competing explanation Dukes does not weigh.18
Open: the dating. Dukes places the no-merit meeting at Wu Ti's court around 520, but the source-by-source ledger pushes Bodhidharma's northern activity decades earlier, raising the question of whether the famous southern audience with Wu Ti happened on the timeline the legend assigns — or at all in the form received.19
Put Dukes's Wu Ti next to the vault's harder-nosed material on power and the friction is instructive. Dukes writes Wu Ti as a near-saint whose only flaw is being too good for the world — the crime wave is "unfortunate (and perhaps foreseeable)," but the moral weight stays on his sincerity. A Greene-style reading of the same facts would be merciless: Wu Ti broke the most basic law of holding power, which is that visible weakness invites the predation that destroys you, and he did it twice over, in the prisons and on the throne. The interesting thing is that both readings use the same evidence. Dukes does not hide the crime wave or the throne-sitting monk; he reports them and reads them as the price of virtue. The power-realist reads the identical incidents as a textbook failure of nerve. Dukes converges with the realist on the facts and splits on the verdict — and because he reports the damning facts honestly, the page can carry the realist reading even though Dukes does not intend it. That is the useful property here: a devotional historian who is too honest to suppress the data that undercuts his hero, leaving the harder analysis available to anyone who wants it. The Asoka comparison is the seam where both readings meet: Dukes offers it as exoneration ("being true to his beliefs"), and it doubles as the indictment (Asoka kept the sword Wu Ti dropped).
Deterrence and the Cost of Defection — Wu Ti's reign is the cleanest historical proof of that page's core claim, and the pairing produces an insight neither carries alone. The behavioral page argues, in the abstract, that cooperative order rests on a credible cost to defection — remove the cost and cooperation unravels not because people turn evil but because the math changes. Wu Ti supplies the experiment with the variable isolated: he held the culture constant, held himself constant, and removed exactly one thing — the credible cost of crime, via abolished punishment and emptied jails — and watched the predicted unraveling arrive on schedule, crime and murder multiplying, the population barricaded indoors. The history page gives the behavioral page something models rarely get: a real polity where the deterrent was switched off deliberately and the consequence is recorded, including the crucial lag (the early "dream come true" window before the new incentives propagate). The behavioral page gives the history page the mechanism that turns a moral story into a structural one — it was never about Wu Ti's character; it was about what any population does once the cost of defection credibly drops to near zero. Together they yield a rule with teeth: compassion that suspends consequences does not reduce suffering, it relocates it — off the offender and onto the next victim — and the relocation is invisible until the lag runs out.
Idealization as Transpersonal Projection — the way Dukes writes Wu Ti is itself a case of the mechanism that psychology page describes, and seeing that adds a layer the history alone misses. The psychology page describes how an admirer loads an ideal onto a figure until the figure's failures get re-narrated as the cost of the ideal rather than as failures. Watch Dukes do exactly this: Wu Ti's crime wave becomes the noble price of a too-good heart; the monk usurping his throne becomes a teaching he was wise enough to permit; the dropped sword becomes humility. The damning data is all present and all reframed upward. The history page hands the psychology page a documented, real-time instance of idealizing reframe operating in a historian's prose, with the un-reframed facts sitting right beside the reframe for comparison. The psychology page hands the history page a diagnostic: when a biographer's hero fails and every failure converts neatly into a virtue, you are reading idealization, and the move to make is to strip the reframe and read the bare facts — which here yields a failed governance experiment instead of a martyred saint. Together they suggest a reading discipline for the whole history domain: separate what a sympathetic source reports from what it concludes, because an honest idealizer will hand you the evidence to overturn their own verdict.
Sharpest implication: the failure mode of compassionate governance is not cruelty's opposite but cruelty's displacement — a reliably merciful authority does not abolish suffering, it transfers it from the offender to the next victim, on a delay long enough that the policymaker never sees the bill.
Generative questions: