elias — Essays & projectshttps://elias-essays.pages.dev/en/Essays by elias on Québec and Canadian politics, political philosophy, faith, culture and technology. Written in Chinese, with English and French editions.enSun, 11 Oct 2026 12:00:00 GMTMiddlemenhttps://elias-essays.pages.dev/en/blog/middlemen/https://elias-essays.pages.dev/en/blog/middlemen/In religion, the middleman does not protect me from God; he protects God from me.I. Thirty minutes

In June 2025, lying in bed watching a video, I cried for about half an hour. Since then one desire has stayed with me: to kneel all the way down and surrender completely to the Lord.

The desire is real, but my life does not match it. In Guangzhou I was part of a church community for a while; I have not gone back in a long time. I want to be baptized, and I have never taken the step. I have been to the Oratoire Saint-Joseph in Montréal, and that is all.

Someone once asked me: the baptism you want, is it the kind with a community and a godfather as witness, or something that belongs only to you and God? Mostly the second, I said. Deep down, I added, I lean toward mysticism; I trust what I feel inside.

This essay tries to work out where that answer comes from, and whether it holds.

II. A conviction older than faith

My politics have changed four times: Maoist, utopian believer in a digital planned economy, liberal, and now something close to a libertarian. It looks like a wide arc. Underneath, one line has never moved: not to be dominated.

That line predates every position. The refusal of central control and the pursuit of individual freedom came first; the economics followed. Looking back, much of what drew me to Maoism was its anti-authority side. The positions kept changing; what they served did not.

When I turned toward religion, the line came with me. I am not wary of God. I am wary of whoever stands between God and me.

III. Mysticism: religion’s refusal to be ruled

Mysticism holds the place in religion that the refusal of domination holds in politics. Its claim fits in one sentence: a person can know God directly, without an intermediary. Sacraments, hierarchy and doctrine are not denied, but they move to second place. The final evidence is the soul’s direct encounter.

The institutional Church has always understood what that implies, which is why it has treated its mystics with equal parts reverence and suspicion. Eckhart said that God’s ground is my ground. In 1329, more than a year after his death, Pope John XXII’s bull In agro dominico condemned more than twenty propositions drawn from his works. Madame Guyon wrote a small book on inner prayer. Arrested at the end of 1695, she was held at Vincennes, in a convent at Vaugirard and in the Bastille, and was not released until 1703.

None of these people denied God or the Church. Their offense was going around it. If a person can receive God directly in prayer, a priest’s direction stops being a required passage, and the Church’s authority shrinks from a monopoly to a reference. For any center, that is dangerous.

There is a more radical version. With the help of AI, I translated A. E. Waite’s The Hidden Church of the Holy Graal (1909). Waite argued that behind the Grail romances lies a “hidden church,” deeper than the visible one, whose mystery is not handed down through the official succession. The visible Church keeps the outside; the real sanctuary is elsewhere.

That is nearly the theoretical version of my own position. I trust those thirty minutes more than any parish register. I do not need a godfather to vouch for me or a parish to stamp me. What happens between God and me is no one else’s to certify.

IV. The objection: in politics, I defend the middlemen

The strongest objection to this position comes from me.

In politics, what I defend is precisely the middlemen. I admire the Middle Ages before absolutism, when church, towns, guilds and families checked one another and together checked the center. Montesquieu wrote that “intermediate, subordinate and dependent powers” are what make a monarchy a monarchy rather than a despotism. On a common reading, Tocqueville saw what followed: once those bodies are destroyed, the individual stands alone before the state, which is exactly the condition despotism needs.

So my political conviction is that something must stand between a person and the highest power. In religion, I want the opposite: to tear down every mediation and set the soul alone before the Most High.

One could say there is no contradiction, since the state can be a tyrant and God cannot. But the problem is not God; it is the interpreter. Without a witness, there is only one interpreter of what God wants from me, and that is me. My wants, my fears and my mood that day can all dress up as God. In religion, the middleman does not protect me from God; he protects God from me.

History has run the experiment. The Reformation removed the Roman middleman in the name of direct access to God, and in much of Europe the result was not freedom but state churches in princes’ hands: cuius regio, eius religio. In England the king became supreme head of the Church. Newman, after leaving it, could see in it nothing but “a mere national institution.” Once the church that crossed borders was gone, religion fell to the prince.

The second objection comes from the mystics themselves: almost none of them walked alone. In December 1577, John of the Cross was locked in a cell in the Carmelite priory of Toledo by brothers of his own order, and held there for more than eight months. In that cell, with almost no light, he composed the first thirty-one stanzas of the Spiritual Canticle. After he escaped, he stayed in the order. Teresa of Ávila wrote down her experience because her confessors told her to.

John also described the dark night: a time comes when every feeling of having been touched disappears, and only a faith without sensation remains. If my spirituality rests on feeling alone, nothing will be left when that night comes. That is why the tradition insists on spiritual directors and community: feelings leave; witnesses stay.

The third case is Evelyn Underhill, two of whose books I translated with the help of AI. In 1907 she had a conversion experience and wanted to become a Catholic, then lingered on the threshold for more than a decade. She went back to the Church of England, and in 1921 asked Baron Friedrich von Hügel to be her director. The author of Mysticism ended by accepting guidance. Von Hügel admired her books but thought she leaned too heavily on mysticism. To him her faith looked like a broad and rather intellectual theism, and he steered her toward one with Christ at its center.

On my side there seems to be only Simone Weil. She loved Catholicism deeply, was never baptized, and stayed at the threshold all her life. But in her letters to Father Perrin she explains that she does not enter the Church because she does not believe God wants her to, for now. Even her refusal was made in the name of obedience, not autonomy.

One detail I cannot get around. Later I found the video again, watched it, and cried again. What moved me was seeing and hearing “them,” real people, living out their faith together. I chose a road for one; what made me weep was the sight of many.

V. Only the truly free dare to walk in

I used to think that not being dominated meant staying out: joining nothing, being registered nowhere, giving no one standing to judge me. The mystics offer another reading.

An institution dominates a person only through what that person needs from it: recognition, identity, security. Eckhart, on trial, declared that he would submit to the Church’s judgment; his ground was never in the courtroom. John of the Cross stayed in the order that had imprisoned him, because the order never owned him. Because their freedom came from elsewhere, they could enter an institution without belonging to it.

So standing outside is not proof of freedom. More likely it shows a freedom not yet finished: I am still afraid of being shaped by others, so I will only submit in an inner space where I alone govern. But a surrender whose terms I set is no surrender. I have written elsewhere that heart means giving yourself entirely to something that needs no justification. A gift that keeps a last share of sovereignty is not yet a gift.

Ignatius of Loyola has a prayer that begins: “Take, Lord, receive all my liberty.” I have always valued freedom above everything else. That is exactly why it is the most precious thing I have to offer. Mysticism should not be my way of keeping it, but the place where I learn to give it away.

I have not walked in yet; practical things still come first. But I no longer take standing outside as proof that I am free.

Sources

Further reading: Meister Eckhart, German Sermon 5b (“God’s ground is my ground”); Teresa of Ávila, The Book of Her Life, prologue (written at her confessors’ command); Ignatius of Loyola, Spiritual Exercises §234 (the Suscipe); the Peace of Augsburg (1555) and the English Act of Supremacy (1534).

]]>Sun, 11 Oct 2026 12:00:00 GMTWhere is America’s Rubicon?https://elias-essays.pages.dev/en/blog/americas-rubicon/https://elias-essays.pages.dev/en/blog/americas-rubicon/The real question may not be “Will the president cross the river?” but “Is the river still there?”A shallow river

In January 49 BC, Julius Caesar stood with the Thirteenth Legion on the bank of a small river called the Rubicon.

The river lies near present-day Rimini, in northeastern Italy. It is neither wide nor deep. What made it matter was not the water but the law: the Rubicon marked the boundary between the province of Cisalpine Gaul and Italy proper. Under Roman rules, a provincial governor’s military command was valid only inside his province. Leading an army across that line into Italy was treason against the Republic.

So what stood in Caesar’s way was never the water. It was an agreement. Romans shared the belief that this line could not be crossed with an army. The agreement had no walls and no garrison; its only guarantee was that generals were willing to respect it, and that soldiers would refuse to follow anyone who did not.

Why did Caesar cross? Later writers often tell the scene as an ambitious man’s gamble. But the ancient sources point to a more practical motive as well: fear.

Caesar had fought in Gaul for nearly ten years, and his enemies in Rome had been waiting almost as long. The Senate demanded that he give up his command and return to Rome as a private citizen. But a private citizen had no immunity from prosecution. According to Suetonius, Cato the Younger swore more than once that he would haul Caesar into court the moment he dismissed his army. A year and a half later, at Pharsalus, looking out over the fallen Pompeian soldiers, Caesar said:

This is what they wanted. After achievements so great, I, Gaius Caesar, would have been condemned had I not sought help from the army.

It was, of course, self-justification. But it lays bare the most dangerous logic of a republic on the verge of collapse: when losing power means losing your freedom, those who hold power will no longer give it up willingly.

More than two thousand years later, I often think about this river. Not because I believe any American president is Caesar—historical analogy is not evidence, and America is not Rome. But because it raises a question that still holds: a border with no walls—what keeps it in place?

Put another way: where is America’s Rubicon? And is it still there?

Before the crossing, the river had already run shallow

Caesar was not the first to lead an army into Rome. Long before him, the river had been tested many times.

From the time of Marius onward, Roman legions were increasingly recruited from the landless poor. Whether a veteran received a plot of land to retire on often depended on whether his general could win it for him in Rome. The soldier’s livelihood became bound to the general’s political fortunes, and his loyalty slowly shifted from the Republic to the general.

In 88 BC, after his political enemies stripped him of command in the war in the East, Sulla marched on Rome. It was the first time a Roman army had attacked its own capital. According to Appian, nearly all of Sulla’s officers abandoned him; only one quaestor stayed. They knew the line could not be crossed. But the soldiers followed Sulla.

Over the next forty years, Marius and Cinna retook Rome by force; Sulla fought his way back a second time and published his proscription lists, putting bounties on his enemies’ heads and seizing their property. The price of political defeat went from losing office to losing one’s life and estate. By the time Caesar stood on the bank of the Rubicon, Romans had seen the line crossed too many times.

This time, the most important figure to leave Caesar was Labienus, his ablest lieutenant in Gaul, who went over to Pompey. The legions stayed. Caesar told his soldiers that the Senate had trampled on the rights of the tribunes of the plebs, and that he was crossing to defend them—in other words, the man crossing the line claimed to be protecting the Republic.

In hindsight, Caesar’s crossing was less a sudden coup than a confirmation: confirmation that the river had grown too shallow to stop anyone. Sulla’s officers could still refuse; Caesar’s soldiers no longer hesitated. Every crossing that went unpunished made the next one easier; every political struggle settled by trials and violence made the next person in power more afraid to let go.

A republic rarely loses its borders in a single flood. More often, the river runs shallow a little at a time.

Americans wrote the river into an oath

America’s founders had read their Roman history. What they feared most was exactly this: a standing army becoming one man’s private force. So the Constitution made the president commander in chief, but gave Congress the power to raise armies, fund them and declare war. The military was to obey elected civilians, but not any one particular person.

The line is written most clearly in an oath. When an American military officer is commissioned, the oath has a single object of loyalty: to “support and defend the Constitution of the United States against all enemies, foreign and domestic” and to “bear true faith and allegiance to the same.” The president does not appear anywhere in it. Neither does anyone’s name.

The enlistment oath adds a promise to obey orders, but its order and its qualifications are deliberate: first, defend the Constitution; only then obey the orders of the president and of superior officers—and that obedience must be “according to regulations and the Uniform Code of Military Justice.” Under that code, the duty to obey extends only to lawful orders.

These words look unremarkable. Their weight shows only by contrast. Soldiers of the Weimar Republic once swore loyalty to the Reich constitution. In August 1934, on the day Hindenburg died, they were required to swear a new oath: “unconditional obedience” to Adolf Hitler himself. Between a constitution and a man, there was only the text of an oath.

So as I understand it, America’s Rubicon is not first of all on a map. It is in one word of that oath: “Constitution.” That word replaced loyalty to a person with loyalty to a set of rules.

But this line has to be seen from both sides. Civilian control of the military is itself a republican principle. A president may replace generals, change military policy and direct the armed forces to carry out lawful missions; officers may not refuse to obey simply because they dislike the president. An army that decides for itself whom to obey has also crossed the Rubicon—just in the other direction.

The real line is narrow: obey lawful orders; refuse manifestly unlawful ones. What holds that line is not any general’s politics. It is thousands upon thousands of officers, prosecutors, judges and civil servants who, when they are asked to choose, still know where their loyalty lies.

2021: the people on the riverbank

The closest America has come to this river was after the 2020 election.

Trump lost the election and refused to accept it. His team filed more than sixty lawsuits across the states and lost nearly all of them, many before judges appointed by Republicans. Then the pressure turned to particular people. Georgia’s secretary of state, Brad Raffensperger, got a call from the president asking him to “find” 11,780 votes—exactly one more than Biden’s margin. He refused.

On January 6, 2021, the pressure fell on Vice President Mike Pence. Trump wanted him to refuse to count certain states’ electoral votes while presiding over the congressional certification. In a letter released that day, Pence wrote that “my oath to support and defend the Constitution constrains me from claiming unilateral authority to determine which electoral votes should be counted and which should not.” Hours later, some of the crowd that stormed the Capitol were chanting for Pence to be hanged. That night, Congress reconvened, and in the early hours of the next morning it completed the certification.

Six days later, every member of the Joint Chiefs of Staff signed a message to the entire force: “Any act to disrupt the Constitutional process is not only against our traditions, values, and oath; it is against the law.” The message also reminded the troops that on January 20, President-elect Biden would be sworn in as their 46th commander in chief.

Most of these people were not Trump’s political enemies. Raffensperger and Pence are Republicans; Pence had served Trump for four years. They did not need to oppose any of his policies. They only needed to refuse, within their own authority, to do what the law did not allow. That is exactly what Sulla’s officers had done: they were not necessarily against Sulla; they simply knew the line could not be crossed.

That time, the river held.

But the lesson of Rome is that holding once does not mean the water will stay as deep. What happens after a crossing attempt matters just as much.

After the 2024 election, the Justice Department, following its long-standing policy against prosecuting a sitting president, dropped the federal charges against Trump over his efforts to overturn the election. On January 20, 2025, the day he returned to the White House, Trump pardoned some 1,500 defendants connected to the Capitol attack and commuted the sentences of fourteen more. Those granted clemency included people convicted of assaulting police officers or of conspiring to stop the transfer of power.

The pardon power is expressly granted to the president by the Constitution, and supporters can reasonably argue that some defendants were sentenced too harshly. But the message these pardons sent went far beyond individual cases: those who tried to stop a lawful transfer of power can be forgiven for their political loyalty. For the next person standing on the riverbank, the cost of crossing just got lower.

Someone is testing the depth

Nothing that has happened since 2025 is a Caesar-style crossing. The president has not ordered troops to surround Congress, nor has he announced that elections are cancelled. That is precisely why these events deserve a closer look. Taken one by one, each can be given a legal justification. Taken together, they look like someone probing the river, spot by spot, to see how deep it is.

The first spot is the military and the cities. In June 2025, over the objections of California’s governor, Trump federalized the California National Guard and sent it to Los Angeles, followed by Marines. Similar deployments or attempts followed in Washington, Memphis, Portland and Chicago. On September 30, hundreds of generals and admirals were summoned on short notice to Quantico, Virginia. Trump told them that America was facing an “invasion from within,” and that some “dangerous cities” should be used as training grounds for the military. Reporters in the room described the generals as largely silent. In January 2026, as protests against federal immigration enforcement spread in Minnesota, Trump threatened to invoke the Insurrection Act and send in regular troops, then said the next day there was no reason to do so “right now.”

The question is not whether the military may ever appear on American soil. In disasters, in riots, when federal law truly cannot be enforced, the law allows it. The question is this: when a president describes his domestic political opponents as enemies who simply don’t wear uniforms, he is quietly moving the line in soldiers’ minds. The “domestic enemies” their oath tells them to defend against—are they people who would subvert the Constitution, or people who oppose the president?

The second spot is the military’s lawyers. On February 21, 2025, Trump removed the chairman of the Joint Chiefs of Staff; the same day, Defense Secretary Pete Hegseth dismissed the Judge Advocates General of the Army, Navy and Air Force. These three were the military’s most senior legal advisers, and part of their job is precisely to tell commanders whether an order is lawful. Hegseth later explained that he did not want military lawyers to be “roadblocks to orders that are given by a commander in chief.”

The president has the authority to replace his generals; that in itself is not illegal. But when Sulla marched on Rome, it was his officers who walked away. Today, the people whose job is to tell officers which orders are unlawful were the first to be replaced.

The third spot is the oath itself. In November 2025, six Democratic members of Congress with military or intelligence backgrounds released a video reminding service members that they can refuse illegal orders. That sentence does little more than restate the oath and the military code. Critics can argue that the video never said which orders were illegal, and that its timing and intent were political—that is a fair debate. But Trump’s response was to call it “seditious behavior, punishable by death.” In January 2026, Hegseth issued a formal letter of censure to one of them, Senator Mark Kelly, and opened proceedings to reduce the retired Navy captain’s rank and pension.

I consider this one of the most alarming moments of the past two years. It was not an order to the military to do something illegal. It made reminding the military not to do anything illegal into a dangerous act. If even restating the oath carries a price, what will the next officer who receives an unlawful order think? To make a river useless, you do not have to cross it. You only have to make people afraid to say where it is.

The fourth spot is the prosecutors. In September 2025, Trump publicly named his targets on social media and demanded that the attorney general go after former FBI director James Comey, New York attorney general Letitia James and others. The U.S. attorney for the Eastern District of Virginia resigned under pressure, and a White House aide with no prosecutorial experience was installed in his place. Within weeks she had obtained indictments against both.

Here I have to be honest about an objection. Trump’s supporters will say: in 2023, Trump himself was criminally indicted four times—wasn’t that political prosecution? I do not think the two can simply be equated. Those cases were brought by a special counsel and by local prosecutors, went through grand juries, and in one of them a trial jury returned a guilty verdict. In this case, the president first named his targets in public, then replaced a prosecutor who would not bring charges.

Yet the objection points to a real danger, and it is exactly the lesson Caesar left behind: once both sides believe that losing power means being dragged into court by the other, no one will be willing to hand over power peacefully. So whether a prosecution crosses the line depends not on who the defendant is, but on the process itself. Is the decision to prosecute driven by evidence and independent prosecutorial judgment, or by the president’s wishes?

But the river is still there

Over these two years, many people have also taken their stand on the riverbank—and not all of them are Trump’s opponents.

The state governments of Illinois, California and Oregon took the federal government to court. Here, federalism is not an abstract principle from a textbook but a real levee: when the president wants to turn a state’s militia into a federal army, the governor can say no, and go to a judge.

On December 23, 2025, the Supreme Court turned down the administration’s emergency application in the Chicago case. Its unsigned opinion hinged on a single phrase. The statute the president had invoked lets him federalize the Guard only if he is “unable with the regular forces” to execute federal law, and the Court read “regular forces” as most likely meaning the military itself. Because the military may lawfully enforce the law only in “exceptional” circumstances, the statute is confined to those circumstances as well. That judgment came from a court with a conservative majority. Before it, four lower-court judges had already ruled against the deployment, finding that conditions on the ground did not justify sending in the Guard. Two of the four had been appointed by Democratic presidents and two by Republicans; one of the appellate judges was Trump’s own appointee. The appeals panel put it bluntly: “political opposition is not rebellion.”

The cases against Comey and James were dismissed by a federal judge in November 2025, on the ground that the interim prosecutor’s appointment was itself unlawful. The Justice Department then took the case against James back to a grand jury twice, and was turned down both times. A grand jury is made up of sixteen to twenty-three ordinary citizens, and its refusal to indict at a prosecutor’s request is extremely rare in American practice. The Justice Department is still appealing; the Fourth Circuit heard arguments in mid-September and, as of this writing, has not ruled.

Mark Kelly’s case was halted in February 2026 by a federal judge appointed by George W. Bush, who found that the Pentagon had violated Kelly’s freedom of speech and threatened the constitutional rights of millions of military retirees. Hegseth said the ruling would be “immediately appealed.”

The military’s top brass have not stayed silent to the end, either. In late August 2026, replying to a letter from Senator Elissa Slotkin, Joint Chiefs Chairman Gen. Dan Caine wrote that the military had no plans to send federal troops or federalized National Guard members to polling places, and no plans to seize ballots or voting machines. He added one more sentence: “I have neither received nor anticipate receiving any unlawful order” concerning the military’s role in the midterm elections.

Reading that sentence, I had mixed feelings. On one hand, here was a general hand-picked by Trump writing the words “unlawful order” in a public letter—he knows where the line is. On the other hand, the country with the world’s oldest written constitution now needs its highest-ranking officer to promise, in black and white, that troops will not show up where its citizens vote.

The river is still there. But it is there not because it is simply part of the landscape, but because someone has always been guarding it.

The Romans later forgot where the river was

Back to the question I started with: where is America’s Rubicon?

My answer is that it is not in the White House, nor on Capitol Hill, nor in the result of any one election. It is in one word of the oath—“Constitution”—and in every particular person who is asked to make a choice: an officer who receives an order, a prosecutor told to bring charges, a secretary of state asked to “find” votes, an ordinary citizen sitting on a grand jury.

So the real question may not be “Will the president cross the river?” but “Is the river still there?”

A crossing is a dramatic moment; everyone can see it, and it can be resisted. The more dangerous possibility is a different one: no one crosses, and the river simply dries up. The people whose job is to say “this is unlawful” are replaced. Those who remind soldiers of their oath are called seditious. Who gets prosecuted depends on whose name the president posts. Those who tried to stop a transfer of power are pardoned. Every step has a legal basis; every step can be explained away as an exception. Until one day a manifestly unlawful order really is given, and the person who receives it looks around and finds that no one can say any longer where the line is.

This is not only a metaphor. The location of the Rubicon really was forgotten.

After the Republic fell, the little river lost its legal meaning, and with it, slowly, any reason to be remembered. Its course shifted, place names changed, and several small rivers in northeastern Italy were each held to be the one; the dispute went on for centuries. In 1933, Mussolini’s government officially designated one of them, the Fiumicino, as the Rubicon, and the town on its bank was renamed to match. Later scholars found some evidence for the choice, but the debate has never been fully settled.

Whenever I think about this, I feel an irony I can’t quite put into words: a dictator decided for the Romans where their Rubicon was.

And in Rome itself, nearly all the names survived. Augustus claimed he had handed the Republic back to the Senate and the people of Rome. The Senate still met; consuls were still elected. Only everyone knew that the real decisions were made elsewhere. The names were all still there. The river was gone.

I am not American. But I have always thought that what is most remarkable about the United States is not its wealth or its power, but that it turned “no one is above the law” into a system that has worked for more than two centuries. That system has no walls, just as the Rubicon had none. How deep it runs depends on whether those who hold power truly believe in it—not on the day they take their oath, but on the day believing in it comes at a price.

As long as enough people remember where the river is, it remains a border. By the time those in power have to tell everyone where the river is, the republic is already gone.

Notes: Caesar’s words at Pharsalus and Cato’s threat to prosecute him are in Suetonius, Life of the Deified Julius, 30; the desertion of Sulla’s officers is in Appian, Civil Wars, 1.57; Caesar’s speech to his soldiers about the tribunes is in Caesar, The Civil War, 1.7; Augustus’s claim to have handed back the Republic is in the Res Gestae Divi Augusti, 34. Caesar’s words are translated by the author directly from Suetonius’s Latin; the other translations are also the author’s own.

]]>Sat, 10 Oct 2026 12:00:00 GMTEncounters in the Wildhttps://elias-essays.pages.dev/en/blog/encounters-in-the-wild/https://elias-essays.pages.dev/en/blog/encounters-in-the-wild/We may be the last generation who could still stumble on great creations “in the wild” while our taste was still taking shape.When I came out of the cinema after the Akira re-release, the thing I couldn’t get out of my head was, of all things, the AI short dramas I’d been scrolling through these past few days.

The two have nothing to do with each other. But colliding like that, they gave me an absurd kind of vertigo.

Celluloid

On the big screen, Neo-Tokyo explodes. The hand-drawn detail is packed tight, even a little morbid, and what comes at you is nothing but sheer physical effort and will.

Akira, Ghost in the Shell, Evangelion… These were monsters that only Japan’s bubble economy at its most feverish, and the warmth still lingering after it burst, could afford to raise. Hundreds, even thousands of people hunched over celluloid sheets day and night, grinding away, just to chisel the hallucinations in their heads onto film.

Katsuhiro Otomo turned his own manga into the animated film Akira. It came out in July 1988, just as the bubble was racing toward its peak. The production committee that put up the money included Kodansha, Toho, and Bandai, and even a trading conglomerate like Sumitomo Corporation had a stake. The Ghost in the Shell film and the Neon Genesis Evangelion TV series both came out in 1995, by which time the bubble had been gone for a few years.

According to Japanese sources, Akira alone took three years. About 1,300 people worked on it, drawing roughly 150,000 cels and using 327 colours. The dialogue was recorded first, and the animators drew the mouth movements to match the voice actors’ performances. The usual practice in Japanese animation at the time was the reverse: draw first, dub later.

Almost forty years on, that density still pins you to your seat.

The banquet that never ends

And on the other screen? AI short dramas pour out by the thousand every day, and streams generated in real time can even keep going for as long as the viewer doesn’t switch them off.

According to DataEye, a data company that tracks short dramas, more than 220,000 new AI dramas and animated comic dramas went live on Douyin alone in the first half of 2026, an average of over 1,200 a day. A March report in Science and Technology Daily said AI comic dramas now cost as little as 500 yuan per minute to make, and a small team of ten or so people can turn out fifty episodes in twenty days. Akira took 1,300 people three years.

The technology really is astonishing, and the cost has dropped almost to zero. But the logic coming off all of it is the same: no resistance at all, no human warmth, only an endless supply of sensory bait, cobbled together to capture neurotransmitters. In the future, when a child opens their eyes, this is probably what they’ll face: a banquet that never ends and never runs dry. You don’t order, you don’t wait, and the dishes keep coming after you’re full.

The yardstick

What really sends a chill down my spine isn’t the technology itself. It’s the shift in the “aesthetic anchor.”

When my generation watches AI junk, we’re actually carrying a yardstick inside us. We were knocked flat by genuinely good things in films, books, and old video games. Even if we didn’t know what “serious art” meant at the time, our taste buds had long since been trained to be picky. So today, when the algorithm dumps its garbage on us, you can taste the industrial saccharine in a second.

But what about the next generation? The baseline of their palate may well be built on industrial saccharine from the start.

A shifted anchor means the yardstick has been swapped. When we measure AI short dramas against Akira, what we get is cheapness. A kid who grows up using AI short dramas as their yardstick will measure Akira and probably get slow, old, and hard to follow. Both will be sure they measured right.

No one taught us this yardstick. It built up over all the times we stumbled onto something.

Encounters in the wild

As a kid, when you stumbled on a masterpiece on TV or on a disc, you weren’t after anything. You didn’t know what awards it had won or where it ranked in film history. You were just passing by, and it hit you. This kind of “encounter in the wild” has a magic to it. Without a sound, it becomes part of the base layer of who you are.

The reason you could stumble on things is that back then you didn’t get to choose. You watched whatever the station put on, and if you didn’t want to, the most you could do was change the channel. With discs, too, you watched whatever was lying around. There were plenty of things you’d never have picked yourself, like an old film you couldn’t really follow from beginning to end, but it happened to be on, so you kept watching. Some you forgot as soon as they were over. Some stayed with you for years. That’s the good thing about stumbling on something: it has nothing to do with the taste you already had. It breaks in from outside.

The algorithm works the other way around. Its whole skill is making sure you never come across anything that isn’t to your taste. Linger a few extra seconds on something and it gives you more of it; swipe past something and it shows you less. It won’t stop you from going to look for Akira. It just makes sure you never stumble on it. Even if Akira does show up in your feed, it will most likely be one of those “Akira in a few minutes” recap videos: the plot gets explained, and the two hours get saved.

But if you miss that window, the good stuff turns into “classics that need to be excavated.”

Watching them as an adult is a make-up class. You can still understand them that way, you may even pick up more of the craft, but what you catch up on rarely makes it into that base layer. What you stumbled on in your teens and what you worked through later, one film at a time off a list, don’t carry the same weight.

Akira won’t disappear, of course. It will lie safe and sound in film archives and on film buffs’ watchlists. But if a child is fed seamlessly by the algorithm from an early age, and gets used to a twist every three seconds and stimulation that never once lets up, what reason would they have as an adult to spend two hours sitting through the slow first twenty minutes of an old film and getting used to the way stories breathed decades ago? Finding it isn’t the problem. They simply don’t need it.

There’s actually no shortage of time. According to the China Online Audiovisual Development Research Report (2026), people spend an average of 129 minutes a day on micro-dramas, longer than Akira’s 124-minute running time. The two-plus hours a day are there to be had. They’ve just been cut into little pieces.

Nostalgia

Every generation of elders thinks the popular culture of the young is going downhill, and I know this can easily turn into clichéd nostalgia.

I should lay out that objection in full first. People have worried in the same way about novels, television, comics, and video games. Japan’s 1955 campaign to drive out “bad books” is one example. According to Japanese Wikipedia, at the campaign’s height a federation of mothers’ groups held a “rally to drive out bad books,” where members wearing aprons put some 60,000 magazines and manga to the flames. It had started with nothing more than a slogan: that May, a group of mothers in Akasaka, Tokyo, launched “Don’t look, don’t buy, don’t read.” Osamu Tezuka later recalled that some people had even called him “an enemy of children.”

Thirty-odd years later, Akira grew out of Japanese manga and animation. The thing I now use as my yardstick is the very kind of thing adults were throwing into the fire back then.

Looking back, most of these fears came to nothing, or at least things never got as bad as people said at the time. The new things people looked down on often went on to produce masterpieces of their own. Maybe the next generation will also stumble on something that’s theirs in ways I can’t imagine, maybe even in the vertical short dramas I turn my nose up at today.

But my worry isn’t a generation gap. It’s that a certain “sense of touch” is wasting away.

Even when novels, television, and comics were being attacked, they were finished works, made and then left sitting there. Once a book is printed or a film is cut, it’s fixed. It won’t change a word or a shot depending on who’s watching. It doesn’t know who you are; you have to walk over to it yourself. Behind the short dramas you scroll past today is a system that measures you in real time. It records the second you swipe away and the frame you stop on, and it adjusts the next one to match. All it wants is for you not to leave.

Drain the things that once stung us and shook us out of a person’s growing up, and what becomes of the way they see the world?

Resistance

Earlier I said AI short dramas have “no resistance at all.” The more I think about that word, the more it seems to matter.

It was resistance that pressed Akira into that density. Every cel had to be drawn by someone, stroke by stroke, then painted, sheet by sheet. A mistake meant starting over, and even the mouth movements had to chase the pre-recorded voices. Staff, money, and time were all limited, and the deadline was right there. Under those conditions, every shot that made it into the film was there because someone had decided it was worth the effort. That density is what you get when decisions like that pile up one after another. AI short dramas don’t have this layer: generating one more segment costs almost nothing, so keeping it doesn’t need a reason.

There is resistance on the viewing end too. A good work doesn’t try to please you right away. It breathes at its own pace and leaves a lot unexplained, and the best parts are often hidden in places that aren’t much fun to watch: a long stretch of silence, a scene you didn’t get the first time. You have to go along with it for a while, put up with a spell of confusion and impatience, before you find out where it’s taking you.

The “sense of touch” I’m talking about is probably honed against the resistance at both ends. Touch is, at bottom, a feeling of resistance: when you press your hand on a table, you can feel the table because the table pushes back against your hand.

AI short dramas take away the resistance at both ends. The people making them don’t have to choose, the people watching don’t have to put up with anything, there’s a twist every three seconds, and nowhere does anything push back. If someone has only ever touched things like this since childhood, the good things will all still be there, but their hand may no longer be able to feel them.

After the show

Preaching like “This is a classic, you should watch it” has never worked. We fell into those works back then because they had a savage vitality of their own. But when the cleverest algorithms in the world are all fighting over one child’s attention, shoving the most mediocre, cheapest dopamine into their eyes at the lowest possible cost, that vitality may never even get the chance to cross their path.

A re-release can’t manufacture an encounter in the wild either. Anyone who buys a ticket for a 1988 animated film probably already knows what it is. But for those two hours, at least, the resistance came back. Once the lights went down, the film moved forward at its own pace. It wasn’t going to cut to a flashier shot because somebody’s attention drifted, and there was nowhere to swipe away to.

When I came out of the cinema, the neon on the street was bright and my phone was still buzzing with notifications. It just suddenly hit me that we may be the last generation who could still stumble on those great creations “in the wild” while our taste was still taking shape.

Sources

Cited at second hand: Yoshiyuki Nagaoka, Manga wa naze kisei sareru no ka (Why is manga regulated?), Heibonsha Shinsho, 2010, pp. 97–98; Osamu Tezuka, Garasu no chikyū o sukue (Save the glass Earth), Kōbunsha, 1996, p. 14. Page references follow the Japanese Wikipedia article 悪書追放運動 listed above; the books themselves were not checked.

]]>Sat, 10 Oct 2026 12:00:00 GMTFreedom in the Pluralhttps://elias-essays.pages.dev/en/blog/freedom-in-the-plural/https://elias-essays.pages.dev/en/blog/freedom-in-the-plural/When medieval people spoke of freedom, they used the plural.The Shoemakers’ Guild

A while ago I thought through a question like this.

In a medieval town, the shoemakers’ guild rules that only its members may make and sell shoes. To join, you first spend many years as an apprentice. Once you are in, everything follows the guild’s rules: prices, working hours, which religious holidays you may not work, how a pair of shoes ought to be made. Then an outsider comes to town. He is willing to sell his shoes for less, but he won’t join the guild. Does the guild have the right to stop him from opening a shop?

My answer is yes. On this one, I side with the community’s right to govern itself.

By my usual standards, this answer doesn’t sit right. I believe in market competition, I oppose tariffs and subsidies, and I have always thought that a company that lobbies politicians to push out its rivals has nothing to do with a free market. The outsider sells cheaper, and the people buying shoes come out ahead. A guild that blocks him looks like it is just protecting its own livelihood. By that logic, I should be on the outsider’s side.

I still chose the guild. The reason I gave myself was that even a polycentric authority like this one is freer than the authority of modern capitalism and the state, which flattens everything.

Thinking it over, what makes me wary has never been “restriction” as such. It is who is doing the restricting, and whether there is anywhere else to go outside it. The biggest difference between a government and a company is that if a company does a bad job, I can go to another one. The guild only governs the shoemakers of one town. Beyond the walls there are other towns with other rules. Inside the walls, the guild isn’t the only authority either: there are church courts and the king’s courts. A centralized state has no “outside.”

This explanation doesn’t clear up the whole contradiction. The outsider really does lose out. He wants to compete on price, and a group of people uses its rules to keep him out. That conflicts with the individual freedom I have always cared about. But if I have to put them in order, I think a community’s freedom from outside interference ranks above an individual’s freedom from control by the community.

A Contract

My earlier idea of the state was simpler. It came from social contract theory. The state is a contract among citizens, and taxes, military service, compulsory vaccination, the ban on drugs and limits on high-risk investment should all count as terms of that contract. Anyone who stays out of the contract is, in turn, shut out of some of the state’s services.

That idea doesn’t hold up, because order spills over. A man who doesn’t pay taxes is still protected by the police and the army. There is no clean way to shut him out. I accept that.

But admitting the spillover did not push me toward central government. I still think the spontaneous division of interests among small communities before absolutism, among the Church, the townspeople and the guilds, was better than today’s central government. Social contract theory asks whether I consented to each obligation. What I care about now is something else: whether there is one power that can swallow all the others.

Some will say the modern state also grew out of centuries of war, annexation and compromise, so why shouldn’t it count as a spontaneous order? I won’t argue with it over origins. Setting up any order has a cost, and changing an order that already exists, on the grounds that it is unjust, has a cost too. Even if a central state really did grow up organically in this way, I would still oppose it, because that degree of concentration is illegitimate in itself.

A Year and a Day

When medieval people spoke of freedom, they often used the plural: in Latin, libertates. A town’s charter, a guild’s statutes, a church’s exemption: each of these was called a “liberty.”

In the 1215 Magna Carta, clause 1 provides “that the English Church is to be free, and to have its full rights and its liberties intact.” Clause 13 provides that “the city of London is to have all its ancient liberties and free customs, both on land and water.” In both clauses the liberties have an owner: this church, that city.

In 1273 the small town of Burgdorf, now in Switzerland, received a charter from the local count and countess. Article 51 says that anyone who comes to Burgdorf, whatever their rank or sex, who has acquired citizenship there and has lived there openly for a year and a day without anyone raising a lawful objection, can no longer be challenged by anyone. A serf who fled into the town and held out for that year and a day could usually no longer be claimed back by his old lord. In the nineteenth century, rules like this were summed up in one phrase: “City air makes you free” (Stadtluft macht frei). The phrase is a later summary. The medieval charters were not so elegant; they only set out the time and the procedure.

The same person often lived under several systems of law at once. Cases about marriage and wills went to the church courts; trading in town meant following urban law and merchant law; lords and vassals dealt with one another under feudal law, and peasants on a manor under manorial law; the king, meanwhile, had a law of his own. The legal historian Harold J. Berman works through each of these in Law and Revolution (1983). As he sees it, one important reason Westerners came to believe that law stands above those who govern is that none of these systems could swallow the others; they had to live side by side and contend.

Outside the Gate at Canossa

Most of these liberties grew out of standoffs. The most famous standoff was the Investiture Controversy.

In 1075 Pope Gregory VII’s Dictatus Papae set out twenty-seven propositions. One says that the pope may depose emperors; another, that the pope himself may be judged by no one. The German king Henry IV refused to give up the power to appoint bishops. The following year the pope excommunicated him and declared his subjects released from their oaths of loyalty to him. In January 1077 he came to the castle of Canossa in northern Italy and stood outside the gate for three days, barefoot and dressed in wool, before he was granted absolution.

The story is often told up to this point, as if the pope had won. But the two sides kept fighting for more than forty years. The Concordat of Worms in 1122 was a compromise. The emperor gave up investing bishops with ring and staff, and bishops were to be elected by the Church according to canon law. The man elected still had to receive his secular fiefs and rights from the emperor.

In 1877 Lord Acton gave a lecture titled “The History of Freedom in Christianity.” For him, the investiture quarrel was only the opening round of a much longer struggle between the Church and the kings. He said: “To that conflict of four hundred years we owe the rise of civil liberty. If the Church had continued to buttress the thrones of the kings whom it anointed, or if the struggle had terminated speedily in an undivided victory, all Europe would have sunk down under a Byzantine or Muscovite despotism.” He goes on to say that what both sides wanted was absolute authority, and that liberty was only the means by which they “called the nations to their aid.”

This is the point I agree with most. The pope wanted to be a man no one could judge. The emperor wanted to hold the Church in his own hand. Neither of them managed it. What Europe got was something neither side wanted: a crack between two powers.

I also thought about a harder case. Two communities clash over the same water source. Who decides? Conflict can’t be avoided. Even if there is a higher authority, someone’s interests will always suffer. I believe the dynamic balance between communities will slowly form a new order, but that doesn’t count as an answer, because I don’t have one. All I know is that if I accept a power that can make the final ruling, I have to accept a central authority. At this point I had to laugh at myself: when did my thinking become so “retro”?

I believe universal law has another lawgiver, above every human power. Only God is sovereign. When any limited person or institution claims to be the final judge, that is usurpation, and this includes the 1075 line “that he himself may be judged by no one.” So I don’t think a modern state guided by rationalism and utilitarianism can be sovereign for long, and I don’t accept a single human religious authority either. On earth, each community interprets universal law for itself, and a shared law is slowly worked out through historical custom and arbitration between communities. Whatever conflicts remain unresolved are left hanging, with no final verdict. There is tension in this mix. It is a relative, dynamic balance.

Only Individuals and the State

Modern “sovereignty” means setting up a final judge on earth. In 1576 Jean Bodin, in his Six Books of the Commonwealth, gave sovereignty its first definition: sovereignty is the absolute and perpetual power of a commonwealth.

In 1651, in chapter 29 of Leviathan, Hobbes lists the infirmities that weaken a commonwealth. He includes “the immoderate greatness of a town” and “the great number of corporations; which are as it were many lesser commonwealths in the bowels of a greater, like worms in the entrails of a natural man.” What I see as plural liberties, he saw as worms.

The French Revolution turned this way of thinking into law. The Allarde Decree of March 1791 declared that anyone was free to take up whatever trade, profession or craft he saw fit, as long as he paid the business tax. The outsider shoemaker had won. Three months later, the Le Chapelier Law, passed on June 14, laid down that people of the same trade, when assembled, could not appoint a president, secretaries or trustees, nor draft regulations concerning their “alleged common interests.” Isaac Le Chapelier, who put the law forward, said that there were no longer any guilds in the state, only the particular interest of each individual and the general interest; no one was allowed to instill in citizens an intermediate interest.

The outsider could now open a shop as he pleased, but the town’s shoemakers could not even have an organization to talk things over. On one side, the individual; on the other, the state; everything in between torn down. That is what I mean by “flattening.”

In The Old Regime and the Revolution (1856), Tocqueville says that administrative centralization was a product of the old regime, and the only part of the old regime’s political constitution to survive the Revolution. On this, the kings and the revolutionaries were on the same side.

In 1931 Pope Pius XI recorded the result in Quadragesimo Anno. He blamed individualism: “following upon the overthrow and near extinction of that rich social life which was once highly developed through associations of various kinds, there remain virtually only individuals and the State.”

What I Don’t Defend

I don’t want to go back to the Middle Ages. That world had serfdom, and Burgdorf’s “year and a day” was precious precisely because there were still serfs outside the walls. From 1209 to 1229 the Albigensian Crusade, launched by Pope Innocent III, pushed into southern France against the Cathars. In 1233 Gregory IX issued the bulls that set up the papal Inquisition, which in practice often became a tool in the pope’s hands. These darkest pages happen to be tied to the concentration of Church power. I don’t use that to let the era off the hook.

Small communities can oppress people too. Guilds shut others out, and families and villages can press down on a person until he can hardly breathe. Here is where I draw the line: a community may condemn someone, refuse fellowship with him, even drive him out, but it may not use prison and violence to force him to be virtuous. Morality is for restraining oneself. Human law has no standing to judge morals and souls.

The body and the conscience are two hard limits. For a while I wasn’t sure where the rights of the person come from. Now I believe they come from theological natural law, and also from the same principle that stands against absolute sovereignty: no state can ultimately own the whole of society, and no community can ultimately own a person’s body. Every religion has mystical currents that speak directly with the divine and resist authority. They have long been suppressed by secular powers and by dogmatism within their own faiths, and they have never died out. Peter and the apostles said, “We must obey God rather than men” (Acts 5:29). But conscience only gives a person the right to refuse to obey. It does not give anyone the right to rule others on the strength of his own certainty.

One more: a community may not deliberately raise its members so that they grow up unable to leave. That goes against God’s will. When I say the community’s freedom comes first, I mean within these limits.

Hierarchy, too, I only half defend. Suppose a city’s judicial offices can only be held by three old families, and a more capable man of ordinary birth is shut out because of it. I would call that a pity; and if those three families fail in their duty, other bodies will respond. As I see it, hierarchy is unjust only when it turns into domination that is closed, absolute, and unable to provoke any pushback. What I am least sure about is this: what if no one responds? The absence of resistance doesn’t show that an order is legitimate. Domination can shape people’s minds so that the tension never even shows. That is possible, and it is real. For people who don’t even have the capacity to push back, I don’t yet have a good answer.

Layer Within Layer

As for today, I don’t want to restore any medieval institution. What I want to keep is its structure. In 1603, in his Politica, Johannes Althusius described society as associations nested one inside another: the family, the voluntary association, the local community, the province, the state, each made up of the ones before it. Sovereignty belongs to the people as organized through these associations, and it cannot be given away.

That is how I understand “the people” too. It is a compound made up of real communities: families, churches, towns, regions and professional associations. It is not the arithmetic sum of equal individuals. A majority cannot rule a community on a core matter like education just because it has the numbers. There is no such thing as a neutral education; the state is only replacing the traditions of families and communities with its own ideology. A child has to grow up inside a concrete way of life before he can become a person.

The evil of power, I think, comes mainly from its reaching into too many areas of life; how many people it governs matters less. By that reasoning, small communities that govern their members’ lives very deeply would seem to call for wariness too. Yet they are indispensable. Only when such communities already exist does a large association that handles just a few matters have anywhere local to leave education, religion, economic organization and most of the administration of justice. You can’t pick one over the other.

Paragraph 79 of Quadragesimo Anno says that to hand over to “a greater and higher association what lesser and subordinate organizations can do” is an injustice and a disturbance of right order, and that “every social activity ought of its very nature to furnish help to the members of the body social, and never destroy and absorb them.” This is what is usually called the principle of subsidiarity.

This will have a cost. So that different communities keep the ability to judge for themselves, to make mistakes and to live in their own way, I am willing to accept a higher crime rate, lower efficiency and greater material cost. But that is a choice I make for myself. Each person can only choose sacrifices for himself. No one can choose them for others in the name of tradition or the common good.

The Next Town

In 1791 France answered the question on the outsider shoemaker’s behalf: let him open his shop, and forbid the town’s shoemakers from ever again forming any association.

I still prefer the old answer. The guild can keep him out. He can pack up his tools and go to the next town, where there is another guild with another set of rules, and maybe they are short a shoemaker.

What I want is a world where there is still a next town.

Sources

Further reading: Robert Nisbet, The Quest for Community (1953); Jacob T. Levy, Rationalism, Pluralism, and Freedom (2015).

]]>Sat, 10 Oct 2026 12:00:00 GMTThe Golden Age of Piracyhttps://elias-essays.pages.dev/en/blog/golden-age-of-piracy/https://elias-essays.pages.dev/en/blog/golden-age-of-piracy/The higher the tower grows, the more leaks down from it.Ocarina of Time

Late this July there was a piece of gaming news: an independent developer used OpenAI’s Codex to port The Legend of Zelda: Ocarina of Time natively to the iPhone and iPad. The original on the N64 ran at 240p and about 20 frames per second; the phone version is widescreen and runs at 60. The source code he used came from the player community, and anyone who wants to play has to supply their own ROM.

Chinese-language communities have this kind of thing too. I’ve seen someone port Diablo II to iOS, and I’ve seen people piece games together with AI.

There’s a Chinese project on GitHub called sts2-ios that ports Slay the Spire 2 natively to the iPhone. The most important file in the repository is written for AI to read. It’s called AGENTS.md, and the README introduces it as “a deterministic runbook that goes from environment checks, filling in config, one-command build, all the way to signing and installing. Follow it and the game lands on your iPhone.” The part of the work that takes knowing how to compile and sign is handed off to coding assistants like Claude Code and Cursor.

People tend to assume there are giant corporations standing behind AI: they hold the compute, the capital and the data, they’ll control everything, and in the end they’ll build a cyberpunk society. That picture isn’t wrong, but only half of it has been painted.

I’ve come to feel that AI is a golden age for pirates, and a golden age for fan creators too. The really interesting things are happening at the giants’ feet. What I want to say comes down, more or less, to the picture below.

Moats

The tower is real. The compute, capital and data it takes to train frontier models are concentrated in a few companies.

But what gets made in the tower is, by nature, impossible to keep in.

On February 24, 2023, Meta released LLaMA. The code was public, but the weights were handed out only to researchers who applied, case by case. Within a week, someone had posted a torrent of it on 4chan.

The next few weeks were later set down as a timeline in an internal Google memo. On March 12, someone got LLaMA running on a Raspberry Pi, so slowly it was barely usable. Six days later, someone shrank it with 4-bit quantization, and it ran fast enough on a MacBook’s CPU, no graphics card needed. The next day, several universities working together trained a 13-billion-parameter model at a training cost of $300, and the memo said it achieved “parity” with Google’s own Bard in evaluations.

The memo was titled “We Have No Moat, And Neither Does OpenAI.” Its author was a researcher at Google, and it leaked in May 2023. He wrote that Google and OpenAI had been fighting each other, and “while we’ve been squabbling, a third faction has been quietly eating our lunch.” He meant open source. “Many of the new ideas are from ordinary people,” he went on. “The barrier to entry for training and experimentation has dropped from the total output of a major research organization to one person, an evening, and a beefy laptop.”

The way it leaked was fitting too: an anonymous person posted it on a public Discord server. The document admitting that things couldn’t be kept in couldn’t be kept in either.

On January 20, 2025, DeepSeek released R1, announcing “performance on par with OpenAI-o1.” The model weights were open-sourced at the same time under the MIT license. “Distill & commercialize freely!” the announcement said, and from then on, training other models on R1’s outputs, by fine-tuning or distillation, was explicitly permitted. Two years earlier, the weights of a good model still had to be approved one request at a time.

The higher the tower grows, the more leaks down from it. The giants have to compete, and competition forces them to cut prices, open-source their models and open up their APIs. Once capability is sold by the token, some people will share their quota with others, and some will build front ends for it and write the manuals.

More importantly, what the giants are selling happens to be exactly the kind of tool that makes individuals stronger. The biggest use of a model that can write code is letting one person finish what used to take a team. I use one for my own coding too. The developer who ported Ocarina of Time did it with an OpenAI product. What it takes to build models can be monopolized. The people who use them can’t be.

Pirates

In 1981, the British Phonographic Industry launched a campaign with the slogan “Home Taping Is Killing Music.” Its logo was a skull and crossbones made from a cassette tape.

Music didn’t die.

Today’s pirates look different. The sts2-ios repository doesn’t include the game itself, and its README says, “Please support the official release: buy the game on Steam first.” The Ocarina of Time port also makes players bring their own ROM. What they change is where a game can be played and how; sts2-ios also added 4x speed and a one-tap room restart to the phone version.

One section of the Google memo has a very candid subheading: “Individuals are not constrained by licenses to the same degree as corporations.” Things a company has to hold meetings about and check with legal, one person just does in an evening.

Some people see pirates as parasites feeding off the giants: the giants create the capability and the pirates steal a little of it. What I see is the reverse. Once capability flows out of the tower, someone has to catch it, take it apart, carry it somewhere new, and write it up as a tutorial others can follow. Only then does it really become everyone’s. That’s the road Ocarina of Time took to the phone. First, a small team spent 21 months reverse-engineering and decompiling the original N64 release. That step made Ship of Harkinian, a source port, possible. Finally, one person used AI to get it onto the phone. Each layer picked up where the one before left off, and none of them was arranged by Nintendo.

Fan works

Gaming history has plenty of famous titles that started out as fan works. Counter-Strike began as a mod for Half-Life, and Valve released it officially the next year. The entire MOBA genre, for its part, was shaped by one fan-made map for Warcraft III. That map was DotA, and a modder who went by Eul put out its first version in 2003. Two big games in the genre trace straight back to it. Dota 2 came from Valve, which had bought the rights to DotA in 2009; League of Legends came from Riot Games, which brought in one of DotA’s early maintainers, and much of what he had learned on the map went into the new game’s foundations.

Making fan works used to mean first knowing how to draw, write or program. Plenty of people were fans. Very few could make anything.

AI has taken down most of that barrier. Another Chinese project on GitHub, diablo-web, is a browser game its author made with AI as a tribute to Diablo II. You open a browser and play, and it has random dungeons, skill trees, set items and quests. He started with Gemini, later switched to Claude Code, and says he worked “entirely in Vibe Coding mode, without looking at a single line of code.”

The Tavern is another example. SillyTavern is an open-source front end; connect it to a large language model and you can role-play with AI and write stories together. Chinese players call it “the Tavern.” In the Tavern, a character is written up as a “character card,” a world’s setting as a “world book,” and how the model should write and play its part as a “preset.” In March 2025, someone on LINUX DO posted a thread titled “For people who love mods, the Tavern is just too much fun.” That’s spot on. A character card is a mod for stories: if you love a character, you can get them talking and have them keep writing the story with you.

Getting in the door is easier now, but going further in still takes craft. Writing a good card means working out how the model reads prompts; making a plugin means knowing how to code. For a character they love, plenty of people are willing to learn these things.

Leinao

People who love the same thing end up together sooner or later. A lot of Chinese Tavern players gather in a place called Leinao (类脑, ΟΔΥΣΣΕΙΑ).

Leinao is a Chinese-language Discord server where members share character cards, presets, world books, plugins and tutorials. In October 2024, “A Beginner’s Guide to SillyTavern” on LINUX DO called it “the biggest Chinese community for character cards and jailbreaking that I know of” and said most of the resources in the post came from there. Jailbreaking means finding ways around the restrictions placed on a model. Leinao also has a lot of adult content. Pirates were never saints.

It grew fast. In March 2025, someone on LINUX DO posted its invite link, and the preview showed more than 59,000 members; in February 2026, the preview in another post showed nearly 244,000. In under a year, it more than quadrupled. The description in the invite preview reads: “Dedicated to the endless pursuit of knowledge and technology in artificial intelligence, setting out on an ever more glorious journey. At the dim edge of contemplation, the singularity arises of itself.” Its doors open and close from time to time: it was shut for a while this February, and by the end of March people could get in again.

There are a lot of people, but the rules are clear. To download resources, you first read the rules, then answer a quiz on them; only once you pass do the downloads unlock. One rule reads: “Do not mention anything paid, compensated or commercial.” Most of the cards there are given away free by the people who made them. One Tavern tutorial calls the people who take cards elsewhere and resell them “profiteers,” tells newcomers to watch out for anyone selling “registration on your behalf” or “ready-made accounts,” and urges people, whenever they use someone’s card, to go to the community and thank its author.

Lücheng (旅程, “Journey”) started as a branch of Leinao dedicated to new cards and is now independent. In March 2025, someone posted its invite link, and the preview showed more than 27,000 people.

Public-good stations

All the posts I quoted above are on LINUX DO. There are plenty of posts there recommending and discussing Leinao, and quite a few asking for invites. People there call each other lao (佬), the lao in dalao, which means roughly “big shot”; the February 2026 post asking for an invite opens its title with “Dear laos.”

LINUX DO is a Chinese tech community built on Discourse. Its motto is “Sincere, friendly, united, professional: building together a community you and I can be proud of.” After registering, a newcomer is at trust level TL0 and has to enter 5 topics, read 30 posts and spend a total of 10 minutes reading before reaching TL1. Going higher takes weeks or months of steady participation, and TL3 also requires not having been suspended in the past six months; the top level, TL4, is reached “only by manual promotion from staff.” The community has its own points, called LDC. Logging in each day, getting likes and having an answer accepted all earn a little. What you save up can be spent at the community’s corner shop, and some public-good stations (gongyi zhan) also take it in exchange for API quota.

The most interesting part is the public-good stations. These are sites that members set up themselves to make AI model APIs available to others. A newcomers’ guide filed under the “Collaborative Documentation” category lists six of them. The table has a column called “station owner,” and one of the owners is listed as “any 大善人,” roughly “any, the great benefactor.” Each station sets its own bar: some can be used at trust level 0, some require trust level 1 plus an invite code, some want you to fill out a questionnaire first. One station’s entry requirements say “unavailable.” The guide makes a point of noting that the table lists “only some of the public-good stations.” The guide itself is pieced together too: several sections note their sources, quoting posts written by other members.

Even the beginner’s post on the Tavern took a moment to add: “Don’t use public-good station APIs for R18.” When something belongs to everyone, people know where the line is.

The public-good stations have no headquarters. Whoever wants to offer API access sets up a station and sets its own bar; when one station stops, the others keep running; people remind each other how the APIs should be used. No one is arranging things from above, and it hasn’t fallen into chaos.

The rooftop

I can’t say whether the worst case will come. If one day a few companies lock away all the best models, cut off API access whenever they please and raise prices whenever they please, these places will be the foundation of a decentralized resistance.

Everything that would be useful then, these places already have. Open-source weights downloaded onto countless hard drives can’t be taken back by anyone. There are people who can move software onto other devices and write manuals for AI to follow; there are people willing to share their API access with strangers. The communities know how to set entry bars, how to build trust bit by bit, and how to keep anyone from taking what belongs to everyone and selling it. These skills were learned for playing games and raising characters, and they work just as well somewhere else.

When that time comes, resistance will probably look very ordinary: one weekend, a few people install a model on their own machines and write a manual so the next person can do it too.

Resistance doesn’t need to start from scratch. These places are getting practice every day.

In that picture, a web of points of light starts from a rooftop piled with old equipment, links up with many rooftops across the city, and reaches all the way to the foot of the tower. The tower reads “Safety,” “Productivity,” “Stability.” On the rooftop, a few people are gathered around laptops writing code, and a dog is asleep in the corner.

The manual for AI in sts2-ios is still up on GitHub, under the MIT license, the same license DeepSeek used when it released R1. Anyone can take it, hand it to their own AI, and follow it step by step.

Sources

Quotations from Chinese-language sources (the sts2-ios and diablo-web READMEs, the LINUX DO posts and newcomers’ guide, the Leinao invite blurb and the Erocraft tutorial) are the author’s own translations.

The illustration was generated with ChatGPT.

]]>Sat, 10 Oct 2026 12:00:00 GMTNothing Is Freehttps://elias-essays.pages.dev/en/blog/nothing-is-free/https://elias-essays.pages.dev/en/blog/nothing-is-free/What public services, subsidies and the absence of competition really cost in QuébecThe project of the century

On September 2, 2025, Québec officially buried Northvolt. Christine Fréchette, then economy minister, announced the end of all public funding for the battery plant planned in the Montérégie (The Canadian Press, September 2025). The $270 million invested in the Swedish parent company had already been lost when it went bankrupt in March.

The Caisse de dépôt et placement du Québec, which invests, among other things, contributions to the Québec Pension Plan, lost another $200 million (Benefits and Pensions Monitor, September 2025). Nobody received a bill. No Quebecer found a line reading “Northvolt” on their notice of assessment.

The money left the way it came: under the vague name of “government money.” This essay starts there.

In Québec, people talk about $9.65-a-day daycare, “free” health care, electricity “among the cheapest on the continent,” “strategic” investments. Each time, the word hides a bill. Someone always pays it: the taxpayer, the patient who waits, the retiree whose savings were invested, or the business that never got started because the money had already gone elsewhere.

This essay does not, first of all, ask for less government. It asks for the bill. A people that cannot see the price of what it buys is not choosing; it is consuming whatever it is served.

The invisible bill

In 2024, the three levels of government collected $237.8 billion in taxes and contributions in Québec. That is 38.6% of GDP, or roughly $26,000 per resident, infants included. No other province carries a tax burden this heavy; among 32 advanced OECD economies, Québec ranks 11th (CFFP, 2026).

Almost no one sees this number. Wage earners receive a paycheque that has already been cut: income tax is withheld at source, along with contributions to the Québec Pension Plan, the Québec Parental Insurance Plan and Employment Insurance. They never write a cheque to the government. They look at their take-home pay and end up believing it is their salary.

The rest is paid elsewhere, a little at a time. At the till, where GST and QST add almost 15%. At the pump. In the rent, which absorbs property tax. In the raises the employer doesn’t give, because it already pays payroll taxes. Each levy is small and scattered; together they are enormous.

Yet the most hidden price is not the amount. It is what the system does to work. For a couple with two children, the effective marginal tax rate reached 90.5% in 2025 at certain income levels: of each extra dollar earned, the family kept less than ten cents. With the abolition of the tax shield announced for 2026, that peak rises to 97.5% (CFFP highlights).

This is not a tax on the rich. These are families of modest or middle income who lose almost all the fruit of an extra hour, a promotion, a second job. They never see it on a bill. They only feel it as a kind of fatigue: you work more, and nothing changes.

What is seen and what is not seen

A French economist explained it all in 1850, better than any auditor general’s report. In What Is Seen and What Is Not Seen, Frédéric Bastiat introduces a citizen, Jacques Bonhomme, whose son has broken a pane of glass. The glazier earns six francs, and the onlookers console him: at least it gives someone work.

What is not seen are the worn-out shoes Jacques Bonhomme will not replace, since his six francs went into the window. Henry Hazlitt, who popularized the fable a century later, drew the lesson: the crowd forgets the merchant who would have made the sale, because it never catches sight of him. He does not protest either; he has no idea what he missed. The whole politics of subsidies fits in that scene.

We see the plant being announced, the minister in a hard hat, the “jobs created.” We do not see the businesses that received nothing, the tax that did not go down, the nurse who was not hired. The gain has a face and a press conference; the loss has neither.

On June 10, 2026, Québec’s auditor general reviewed 29 files in the battery sector: 11 companies, about $2.2 billion in authorized assistance. She described a “poorly planned” approach, without objectives or timelines. Four of these companies, authorized to receive $760 million, ended up under creditor protection (Noovo, June 2026).

Losses were estimated at $375 million as of March, and could rise. Three companies authorized to receive $1 billion saw the cost of their projects more than double. After the report, the premier herself acknowledged that it is not necessarily the government’s job to inject funds directly into start-ups.

Lion Electric had received up to $177 million in support from Québec. In May 2025, François Legault acknowledged the province would lose about $140 million; the leader of the opposition accused the government of handing Lion a monopoly on the electrification of school buses (Noovo, May 2025).

That companies go bankrupt is no scandal. It is the normal price of innovation, and the market pays it every day. The scandal lies elsewhere: the venture capitalist chooses to risk his own money; the taxpayer was never asked.

The price of monopoly

When money no longer sets the price, the price does not disappear. It changes currency. We pay it in hours, in waiting, in choices we no longer have.

Care is “free” at the point of service. In 2025–2026, a patient on a stretcher in the emergency room waited on average 16 hours 46 minutes before being admitted or sent home. At Lakeshore General Hospital it was nearly 31 hours, and close to half of the visits involved people over 75 (Noovo, April 2026).

Those hours are a price, paid in lost wages, in pain, in worry. Back in 2005, in Chaoulli, a case that concerned Québec itself, Chief Justice McLachlin and Justice Major put it in their reasons: “Access to a waiting list is not access to health care” (2005 SCC 35, para. 123).

Daycare follows the same logic. At $9.65 a day, a subsidized spot is a bargain, for those who get one. As of May 31, 2025, 30,688 children were still waiting for a spot, subsidized or not, while some families pay more than $60 a day elsewhere (Narcity, March 2026).

The low price did not abolish scarcity; it turned it into a lottery. The winners pay $9.65. The losers pay for their own daycare, partly refunded through a tax credit, and, through their taxes, for their neighbour’s spot.

Then there is the SAQ, the provincial liquor monopoly, which describes itself as owned by its “nearly 9 million shareholders,” Quebecers. Dividends and taxes included, it sends nearly $2.1 billion a year straight to the Québec treasury (SAQ). But a shareholder who can neither sell his shares nor shop elsewhere is not a shareholder: he is a captive customer.

The SAQ’s dividend is a tax paid at the counter, without a single line on the receipt naming it. And there is something more serious here than price. Albert Hirschman distinguished two ways of responding to a disappointing service: exit, going to a competitor, and voice, complaining.

Monopoly abolishes exit. Only voice remains, and for most people voice comes down to a vote every four years. What competition offers the ordinary consumer is not perfection. It is the door.

Why we don’t see the bill

We don’t see the bill because everything pushes toward hiding it. It is not a conspiracy. It is a slope, and every actor has an interest in going down it.

The economist Mancur Olson showed it in 1965: a small group that gains a lot always organizes better than a crowd that loses a little. A company hoping for $200 million hires lobbyists. Nine million taxpayers who each lose about twenty dollars don’t even leave the house.

As early as 1903, the Italian Amilcare Puviani described “fiscal illusion”: the state prefers levies the taxpayer does not feel. Withholding at source is its masterpiece. Milton Friedman, who helped design it in the United States during the war, told Reason in 1995 that it was “a great mistake for peacetime” and that he wished “there were some way of abolishing withholding now” (Reason, 1995).

Vocabulary does the rest. “Québec invests $500 million,” “the government offers,” “state assistance”: these sentences have a subject who gives and no subject who pays. At the Conservative Party conference on October 14, 1983, Margaret Thatcher put it bluntly: “There is no such thing as public money; there is only taxpayers’ money” (Margaret Thatcher Foundation). One can loathe Thatcher and still admit that, on this point, grammar is on her side.

The word “free” is the most effective of all. It turns a collective purchase into a gift. And one does not haggle over a gift, nor ask whether it was worth the price.

Some will object that Quebecers get back a good share of what they pay, through benefits and credits. The tax research chair at the Université de Sherbrooke even notes that once benefits are counted, the picture is more nuanced: applied to Québec households, the tax-and-transfer systems of four other provinces would actually be heavier.

That is true, and it is precisely the problem. A dollar that leaves as tax and comes back as a benefit makes no one richer. It passes through an administration that counts it, attaches conditions to it, then claws it back as soon as you earn a little more. This back-and-forth is the direct cause of the 90% marginal rates described above: the state returns your money on condition that you don’t work too much.

The short memory of the twentieth century

Sympathy for socialism is first a matter of memory, and the memory of the twentieth century is fading. In the fall of 2022, a Léger poll commissioned by the Fraser Institute found that 42% of Canadians favour socialism. The figure rises to 50% among 18-to-24-year-olds and falls to 38% among those over 55 (Toronto Sun, February 2023).

One may distrust the sponsor. But among the results the Institute highlights, the most revealing concerns the definition: only 25% of respondents understand socialism as the state taking control of companies; 65% see it simply as more public services.

And when asked who will pay, 72% support a wealth tax on the richest 1%, but only 31% a general income tax increase, and 16% a GST increase. The study’s authors, Jason Clemens and Steven Globerman, draw their own conclusion: Canadians want more services, provided someone else, the wealthy, pays for them (Wealth Professional, February 2023). That is the reading of a free-market think tank; if it holds, this kind of socialism is nothing but a bill you cannot see.

Nobody in Québec is calling for collective farms. But the starting intuition is the same: that a price can be abolished by political decision. The twentieth century ran that experiment to the end.

In the Soviet Union, cheap bread was paid for in hours spent in line. In China, the Great Leap Forward caused a famine that killed tens of millions, according to most historians. In Cambodia, the Khmer Rouge abolished money and the market; they caused the deaths of at least one in five inhabitants.

The difference with today’s Québec is immense, but it is a difference of degree. There as here, when the price is abolished by decree, the queue replaces it. Sixteen hours on a stretcher are not a Soviet bread line; they obey the same law all the same.

Montréal is home to people who know all this without ever having opened an economics textbook. In 1979 and 1980, Canada took in some 60,000 refugees from Southeast Asia, many of them Vietnamese boat people driven out by a regime that had just confiscated their businesses. Families who came from Poland, Romania, Cuba, China and, more recently, Venezuela carry the same memory.

That memory is rarely invited into Québec’s public debate. It is suspected of exaggerating, or of being “right-wing.” Yet it is the only direct experience Québec has of what waits at the bottom of the slope.

Young people who dream of more services are not naive by nature. They have simply never seen the bill arrive. An honest debate shows it to them in advance, instead of letting them learn it the way others learned it elsewhere.

The bill first

A grown-up democratic debate begins with the bill. On October 5, Québec changed governments. The question of the bill remains wide open.

I am not proposing a program. I am asking for something simpler and harder: to see the price. In practice, it could look like this:

  • a notice of assessment that says, in dollars, where its recipient’s taxes went, like the annual tax summary the United Kingdom has provided to its taxpayers since 2014;
  • a daycare receipt that shows the real cost of the spot, next to the $9.65 paid by the parent;
  • a statement telling patients what their emergency room visit cost the public;
  • a subsidy announcement that states its cost per taxpayer and the risk of loss, like any prospectus.

None of this takes away a single service. It only takes away the illusion.

Some will say that showing the price is laying the groundwork for cuts. If a service cannot survive the sight of its price, it was not worth it. If it is worth it, Quebecers will pay for it knowingly, and that, at last, will be a real choice.

Northvolt’s $270 million had no face. Neither did those who paid it. Consent without knowledge is not consent; a society that wants to be one of solidarity must first agree to look at what its solidarity costs.

Sources

Further reading: Frédéric Bastiat, What Is Seen and What Is Not Seen (1850); Amilcare Puviani, Teoria della illusione finanziaria (1903); Henry Hazlitt, Economics in One Lesson (1946); Mancur Olson, The Logic of Collective Action (1965); Albert O. Hirschman, Exit, Voice, and Loyalty (1970).

]]>Sat, 10 Oct 2026 12:00:00 GMTShould the law punish incivility, or ban the freedom that might lead to it?https://elias-essays.pages.dev/en/blog/punish-incivility-not-freedom/https://elias-essays.pages.dev/en/blog/punish-incivility-not-freedom/The law should punish incivility, not confiscate everyone’s freedom in advance.A big green bottle

I come from East Asia, and back home my beer was Tsingtao. On a summer evening you can carry a big green bottle down the street, drinking as you walk and watching people go by, and nobody thinks anything of it. In Qingdao, draught beer is even sold in plastic bags: you carry it by hand with a straw stuck in the top, and it is part of the city’s scenery.

When I got to Montréal, I quickly fell for Québec craft beer: sours, IPAs, stouts, every microbrewery with a temperament of its own, the kind of place you keep going back to. But when I wanted to take a few cans to the park, or drink one while walking along the river on a summer night and looking at the city lights, I found out that here it is illegal. Montréal’s by-law P-1 (article 3) lets you drink in a park only during an outdoor meal, and only in a section where the City has put picnic tables.

This puzzled me. In 1898, Canada held a national referendum on prohibition. Every other province voted in favour; only in Québec did more than eighty percent vote against. How did the place that once rejected prohibition most firmly end up requiring you to lay out a sandwich before you can drink a can of beer in the park?

Banning harm, or banning freedom

The law can be written in two ways here. One way targets harm that has actually happened: late-night noise, fights, thrown bottles, public urination. Each of these is already covered by its own rule. The other way targets a freedom that “might lead to harm” and takes it away entirely. The ban on public drinking is the second kind: what it punishes is not waking the neighbours, but the bare fact of holding a can of beer.

In On Liberty, John Stuart Mill drew a line: the state may coerce someone for one reason only, “to prevent harm to others.” He happened to use drinking as an example: “No person ought to be punished simply for being drunk; but a soldier or a policeman should be punished for being drunk on duty.” Someone once convicted of violence while drunk, he argued, could be placed under “a special legal restriction, personal to himself.” Mill does not deny the risk. He only insists that restrictions fall on the situations and the people that actually create it.

Is punishment after the fact enough?

The other side’s case is not weak. Some harms cannot be undone: broken glass cuts a child, a drunken argument can turn into a stabbing, and a ticket afterward heals no one. Enforcement after the fact is also hard. The police cannot be there every time someone gets loud, while “no drinking” is clear and easy to enforce. And risks add up: one person with one can is harmless, but hundreds drinking late into the night in the same park can get out of hand. On the night Québec’s curfew was lifted in May 2021, Victoria Park in Québec City filled with people. By the end of the night a man had been stabbed, a woman had a head injury, and the police cleared the park.

So the claim that all preventive restrictions are illegitimate does not hold up. The real question is what kind of risk justifies what degree of prevention.

Grading the risk

I would rather sort risks into three tiers.

The first tier is serious, irreversible, falls on bystanders, and is not unlikely. Here a ban in advance is legitimate. Drunk driving is the classic case; no one proposes waiting until someone is killed before punishing it.

The second tier is moderate and reversible, but concentrated at particular times and places. The right tool here is a limit on time and place. Japan has long allowed drinking in the street. Tokyo’s Shibuya ward at first banned street drinking around the station only at Halloween and New Year, and only in October 2024 extended the ban year-round, from 6 p.m. to 5 a.m., with no penalty for violators. After the Victoria Park incident, Québec City likewise only moved its cutoff for drinking in parks from 11 p.m. to 8 p.m.

The third tier is minor and reversible, borne mainly by the drinker, or simply “unsightly.” Here the law should punish only disturbance that actually happens. Someone sitting quietly on a bench with a can of craft beer clearly belongs in this tier, yet the current rules treat it as if it were the first.

Who gets the ticket

A broad ban turns almost everyone into a potential offender, so who gets fined comes down to the discretion of whoever enforces it. Picnickers drinking red wine with a baguette and cheese are rarely bothered. Céline Bellot’s team at the Université de Montréal analysed years of municipal court tickets and found that the tickets given to homeless people were heavily concentrated in minor offences such as public intoxication and loitering. The ban also carries a hidden price threshold: people who can afford a bar terrace can raise a glass outdoors; people holding a beer from the dépanneur cannot.

There is another way, and it has been tried. In 2023, Toronto ran a pilot allowing personal drinking in 27 parks. The evaluation found that the disturbances and litter people had feared largely did not materialize, and alcohol-related emergency visits did not go up. In 2024, city council made the program permanent, with at least one park in every ward, while banning drinking within two metres of playgrounds and wading pools. Openness is the default; restrictions target specific points of risk.

Back to Québec

Québec went its own way. By 1919 the other provinces had all gone completely dry, most of them during the First World War. Québec’s own ban took effect that May, but a referendum a few weeks earlier had exempted beer, cider and wine, so only spirits were outlawed. About nine in ten Québec municipalities still banned sales locally, yet the province became the only place in Canada or the United States that never went completely dry. In 1921 it replaced prohibition with regulation, handing the liquor trade to a state commission. That generation acknowledged the harms of heavy drinking, but did not take away everyone’s glass because of them.

I am not asking the law to go easy on drunks. If I shout in the park or smash a bottle, fine me by the book. I only hope that, before it comes to that, I can sit quietly by the river, open a can of Québec craft beer, and, the way I used to carry a big green bottle back home, watch the city at night for a while.

The law should punish incivility, not confiscate everyone’s freedom in advance.

Sources

Further reading: John Stuart Mill, On Liberty (1859), chapter 1 (the harm principle) and chapters 4 and 5.

]]>Sat, 10 Oct 2026 12:00:00 GMTThe Empty Shellhttps://elias-essays.pages.dev/en/blog/the-empty-shell/https://elias-essays.pages.dev/en/blog/the-empty-shell/Honesty means not deceiving; heart means staking your whole self.Niu Lai

In the summer of 2026, a Chinese animated film called Niu Lai (牛来, after its calf hero) opened in theatres. No marketing, no trailer. In its first ten days it took in a little over 7,700 yuan and was seen by fewer than 300 people in the entire country.

It is astonishingly crude. Stiff 3D models, mechanical movements, and an opening shot of a lark circling the sky for five minutes. The director, Xin Yumeng, is 34. He trained in landscape architecture, never studied animation, and made the film with software meant for drawing garden plans.

The film was made almost entirely by two people: him and his mother, Sun Lifang. For five years the son modelled, animated, edited and voiced; the mother wrote the script, voiced every female character, and taught herself composition to write the whole score and the closing song. The production company used to be a renovation firm. The distributor, a friend, urged him not to release it in theatres. He answered that it had taken too much to make.

The story is simple. A newborn calf cannot stand. Exhausted, he falls asleep and dreams. In the dream the wolves come, and his mother sacrifices herself to lead them away from the herd; grown up, the calf in turn gives his life for his own. Then he wakes, stands, and walks, step by step, toward the herd.

At first his parents thought he was mad. Then they told him what so many Chinese parents tell their children: do what you can.

It was the closing song that made me think of Shostakovich’s Second Waltz. Written and sung by the mother, it has the lilt of a Russian folk tune; many Chinese of her generation grew up on Soviet songs. The waltz is ballroom music, glittering and turning, but something heavy lies under the melody, like sadness behind a smile. The song is like that.

The film fails from start to finish. But a mother spending five years helping her son finish a dream that everyone laughed at: that is real. In the film the mother dies for the herd; outside it, she writes songs for her son. Both are real.

That is what I want to write about. For lack of a better word, I will call it heart.

What I mean by heart

Heart is not honesty. A person can never tell a lie and still care about nothing. Honesty means not deceiving; heart means staking your whole self.

A mother’s love for her child is heart in its purest form. It has no audience: when no one is watching, she loves just the same. It expects nothing back: she does not love her child because he will one day repay her. And it needs no reason: ask a mother why she loves her child and she cannot answer, nor does she need to. Heart always points toward something that no longer needs to be justified.

In human history, it was religion that supplied this something: an absolute, larger and truer than oneself, to which a person could give himself entirely. That is why so many of those who took up arms in God’s name, from the Crusaders to the Taiping, were sincere. They wanted more than bread; they wanted a true world.

Some will say the fanatic is sincere too. He is not. A person with heart is honest with himself: he doubts, he suffers, he lies awake asking whether he was wrong. The fanatic no longer doubts; he has handed his thinking over to a slogan and no longer faces himself. Fanaticism is not too much heart. It is another kind of falseness.

A morality without roots

Our society does not lack moral vocabulary. Equality, inclusion, diversity, progress, rights: there has never been more of it. But there is nothing underneath.

At the opening of After Virtue, Alasdair MacIntyre imagines a catastrophe that destroys the natural sciences. The survivors find only fragments of formulas and terms, which they go on reciting without knowing what whole they once belonged to. Our moral language, he says, is in that state. We still say “good,” “duty,” “dignity,” but the world that gave those words their meaning, the one that answered why we live and what a human being should become, is gone.

The American philosopher Elton Trueblood called this a “cut-flower civilization”: in the vase, the flower lasts a while, but it has no roots. Tom Holland, in Dominion, puts it more bluntly: the best of our secular values grew out of Christianity before the root was cut.

The words remain, the roots are gone, and only two things are left: posturing and transaction. Saying the right thing buys safety and approval; everything is negotiable, nothing is beyond exchange. People follow this morality not out of conviction but out of fear of being shut out.

In The Power of the Powerless, Václav Havel described a greengrocer who places a sign in his shop window: “Workers of the world, unite!” He never gives a thought to what it says, and the passers-by do not even read it. The sign is only a token of loyalty: belief is optional, compliance is not, and so everyone puts one up. The whole system rested on this shared untruth. Today no one forces us to put up signs. We do it ourselves. The slogan has changed; the window is the same.

Deeper still: in secular progressive morality, everything must be justified and everything can be overturned. What is right today will be wrong tomorrow. It calls this openness. In fact it means that nothing in this morality deserves to be loved unconditionally. A mother need not justify her love; a morality in which everything demands justification cannot be loved that way, and so it cannot give anyone heart.

The universal and the particular

What kind of community can hold heart? My test is this: it rests on a universal conviction that anyone can join fully, provided they truly believe it.

The French Republic says: you are a citizen. It does not ask who your ancestors were or where you were born; it asks whether you make its principles your own. Rousseau devoted a chapter of The Social Contract to “civil religion,” and France built it a temple: across the front of the Panthéon run the words Aux grands hommes la patrie reconnaissante, “to its great men, a grateful nation.” For Renan, a nation is “a daily plebiscite”: it rests on consent renewed every day, not on blood.

The United States is the other example. Chesterton called America the only nation in the world founded on a creed, “a nation with the soul of a church.” “All men are created equal” is an article of faith: believe it and you are American. American exceptionalism, at bottom, is this: a nation that believes it carries a mission for all of humanity.

Both traditions are secular, but they kept the skeleton of religion: an absolute conviction, a call addressed to everyone, something worth dying for. That is why they can ask for heart.

Against them stands particular identity: language, ancestry, memory, “us.” A particular identity does not ask you to believe anything, only to belong. And belonging requires no heart. It is inherited.

The universal is accused of arrogance and fanaticism, of waging war in the name of humanity. But a conviction, however dangerous, is still a conviction; a shell, however harmless, is still a shell. The deeper sickness is the second.

Québec: the container without the content

I live in Québec. By this test, it is the clearest example of an empty shell I know.

It was not always empty. Before the Quiet Revolution of the 1960s, Québec was the most Catholic society in North America. Language and faith went together; people used to say that the language was the guardian of the faith. The formula is Henri Bourassa’s: the nationalist leader argued it in his 1910 speech at Notre-Dame in Montréal, and in 1918 gave a lecture under that very title. French was defended because it carried a faith, a whole understanding of why we live.

And “catholic” means universal. Pre-1960 Québec was a particular people within a universal Church: its language and history were the particular half, its faith the universal half.

The Quiet Revolution threw out the universal half and kept the particular one. The churches emptied, then became condos, libraries, concert halls. The language stayed, defended more fiercely than ever. But it went from container to end in itself: people guard the bottle with everything they have, while what it held was poured out long ago.

The Québec sociologist Fernand Dumont spent his life asking what, once a shared faith is taken away, could still hold an “us” together. Québec has still not answered.

Québec will say: we have laïcité. In 2019, Bill 21 wrote the secularism of the state into law. But the law only says what Québec refuses: religious symbols on certain public employees, the return of the old Church. Something founded on a refusal is a defence, not a conviction. Laïcité says what Québec is not; it cannot say what Québec is.

Québec will also say: we have civic nationalism; French is a door open to all, and anyone who learns it and takes part in common life becomes Québécois. But a door is worth what lies behind it. When the Republic says “you are a citizen,” behind it stands a whole conception of the human being. When Québec says “you speak French,” what stands behind it? French.

Québec’s intellectuals have long condemned Canadian multiculturalism as hollow relativism that welcomes everything and is therefore nothing. That sentence can be handed back to them word for word.

Postnational Canada, and elsewhere

If Québec is a shell, Canadian multiculturalism is emptier still. Québec at least defends one thing with everything it has. Canada has given up even that, and calls the surrender a virtue.

In 2015, Justin Trudeau told The New York Times Magazine that Canada has no core identity and no mainstream, that Canadians share values instead, such as openness, respect and compassion, and that this makes Canada the first postnational state. He said it with pride. Openness, respect, compassion: these words are not wrong, they are empty. They ask no one to give himself. A community that welcomes everything ends up demanding nothing, and so it ends up being nothing.

Canadian conservatism, too, long defined itself by not being American. In 1965, in Lament for a Nation, George Grant was already mourning that tradition’s collapse. A country founded on what it is not has, like Québec’s laïcité, a defence and no calling.

When nothing is believed in common, only interests remain. Confederation becomes a bargaining session between Ottawa and the provinces; the West becomes a resource reserve to be drilled. Immigration, which ought to be an invitation into a shared life, has become in Canada a way to supply labour: to hold down the cost of work, the elites open the door to the whole world.

The numbers are there. At the 2021 census, Canada had fewer than one million non-permanent residents (temporary workers, international students, asylum claimants and others), 2.5 percent of the population. By the fall of 2024, Statistics Canada estimated more than three million, over 7 percent. In the low-wage stream of the Temporary Foreign Worker Program, Ottawa approved about 28,000 positions in 2019 and about 84,000 in 2023, three times as many, with cooks and cashiers near the top of the list. Immigration Minister Marc Miller admitted that Canadian employers had become “addicted” to temporary foreign workers. In October 2024, Justin Trudeau conceded that his government “didn’t get the balance quite right,” cut the 2025 target for permanent residents from 500,000 to 395,000, and set a goal of bringing temporary residents back down to 5 percent of the population.

When a country takes someone in, is it inviting him to become something, or does it only need him to do something? Canada’s answer is the second.

The United States and France once had something, but they are losing it. Trump-era America is retreating from a creed you join by believing to an identity of birth and blood; it is taking Québec’s road. France as well: when the republican ideal shrinks to administrative rules on secularism, the shell is not far off.

The laughter

What happened to Niu Lai after its release says more than the film itself.

It became a sensation, passing 63 million yuan by mid-September, but as a joke. People did not go to watch it; they went to be seen at it, to shout “Mama!” along with the calf and film it all on their phones. The title was read as “the bull market is coming,” and stock traders bought tickets for luck. Brands selling bubble tea, milk and sauces rushed out deliberately crude copycat ads. Rumours piled up: money laundering, subsidy fraud; AI was used to fake interviews and clips. Abroad, a pirated copy was screened to paying audiences in San Francisco. Someone even launched a cryptocurrency in its name.

Five years of a mother’s and a son’s heart became, in this world’s hands, three things: a meme, clicks, and a business.

Most of the people laughing meant no harm. That is exactly the problem. A hollow society can only receive heart through irony. It does not know where to put it, so it laughs; and when it has finished laughing, it sells it. Irony is the mother tongue of the empty shell.

Some were moved, of course, and called it the most sincere film of the year. But they often had to laugh first before they could say it. Today, admitting that something sincere has moved you requires first proving that you find it ridiculous. Otherwise you look naive.

What remains

Shostakovich spent almost his whole life under a regime that demanded a performance from him. In 1936 an editorial in Pravda denounced his opera as “muddle instead of music”; in 1948 he was condemned again. He made his self-criticisms, issued statements, read speeches others had written for him. He said everything he was supposed to say.

But his music did not lie. The Second Waltz is, on the surface, a light-hearted ball; underneath, a sorrow that cannot be spoken. He hid his heart where the regime could not find it, or pretended not to.

Heart does not change regimes. Shostakovich did not change the Soviet Union, and Niu Lai will change no one. But regimes pass, slogans change, the right words go out of fashion, one joke replaces another. A melody that hides its sorrow under glitter, a song a mother wrote for her son: these remain.

At the end of Niu Lai, the calf wakes from his dream, stands up, and walks alone toward the herd.

This is what we can do: in an empty shell, begin by not being one.

Sources

]]>Sat, 10 Oct 2026 12:00:00 GMTThe Stationer Statehttps://elias-essays.pages.dev/en/blog/the-stationer-state/https://elias-essays.pages.dev/en/blog/the-stationer-state/School supplies, supply management, and the bill the central neighbourhoods never seeA hundred and twenty-seven dollars

Every July, Retraite Québec pays parents a supplement for the purchase of school supplies: $127 per child aged 4 to 16 for 2026–2027, regardless of income (Narcity, July 2026). Parents spend it as they see fit: a new backpack, the pencils the child likes, or cheap notebooks and the rest on groceries.

On August 26, 2026, Québec solidaire proposed replacing that cheque. The Centre d’acquisitions gouvernementales, the government’s central purchasing agency, would buy supplies in bulk, and the state would then distribute them to all parents, whatever their income. Ruba Ghazal estimated the savings at $60 per child, plus the time parents would save at back-to-school (Noovo, August 2026).

She admitted the measure would also help well-off families who don’t need it, but insisted on avoiding any stigma at school. The argument is telling: so that no one stands out, everyone gets the same box.

Two years earlier, in August 2024, the same party demanded the opposite: doubling the cheque from $121 to $242, at a cost of $140 million. Ruba Ghazal said at the time that a full backpack is not a luxury (Noovo, August 2024). In two years, cash for parents went from necessity to a problem to be fixed.

This reversal is no campaign detail. It betrays a conviction: the state knows better than parents what belongs in their children’s backpacks. The same conviction underpins supply management, the fear of failure, and the blindness of the neighbourhoods that vote for it to what their comfort costs others.

The state pencil

A cheque is always worth at least as much as a parcel of the same value. It is one of the simplest results in economics: with $127, parents can buy exactly what the state would have bought, or something better suited to what they know about their child. With a parcel, they get what an official planned for an average child who does not exist.

And supply lists are not the same from one school to the next, or even from one class to the next. Many ask parents to reuse last year’s supplies if they are still in good shape. A central buyer must therefore choose: standardize the lists and take that choice away from teachers, or handle thousands of different lists and lose the advantage of volume. Either way, some of the notebooks will end up at the back of a closet.

Then everything has to be stored, shipped and handed out. In 2024, Québec solidaire itself counted 1.65 million children covered by the supplement. At the current amount, that is about $210 million a year flowing through a single window, instead of being spent in hundreds of stationery stores, pharmacies and dollar stores that compete with one another every August.

That window already has a record. In November 2024, the auditor general found that many of the group purchases the Centre d’acquisitions gouvernementales had planned never took place; running behind schedule, it signed contracts without tenders to keep services from stopping. As for the $407.9 million in savings it credits itself with, she called its method of calculation “sometimes questionable” (Noovo, November 2024).

This is the agency that would be handed back-to-school for all of Québec, with a fixed deadline and no margin for error. A delay would no longer mean a renewed contract; it would mean children without notebooks on the first day of school. One can guess what kind of contract gets signed under those conditions.

I am not accusing anyone of corruption. I am describing a structure. When a single buyer awards hundreds of millions in contracts every year, suppliers no longer compete for parents; they compete for a handful of decision-makers. Québec has seen this film: the Charbonneau Commission documented collusion in public construction contracts, and in December 2025 the Gallant report found that the SAAQ’s leadership had concealed for years the cost explosion of SAAQclic, now estimated at no less than $1.1 billion (Noovo, December 2025).

The market is not virtuous. It is simply harder to corrupt, because purchasing power is scattered across millions of hands. Centralizing purchases creates a target.

Virtue paid for at the grocery store

In a university seminar, supply management has everything going for it. It protects the family farm from American giants, stabilizes prices, keeps the countryside alive. One leaves convinced of having stood up for the weak. Yet the question rarely asked there is the simplest one: who pays?

The mechanism is well known. Quotas limit the production of milk, eggs and poultry; prohibitive tariffs block imports beyond small quotas. Canadian prices therefore stay above market prices. The difference does not vanish: it is paid at the checkout.

The first to pay is the consumer, and the poorest first of all. In June 2026, the Montreal Economic Institute, a pro-market think tank, put the cost at $224 per person per year: 1.25% of the income of the poorest households, against 0.33% for the richest. It draws on a 2015 study in Canadian Public Policy: the lower the income, the larger the share of the household budget this cost takes. By the institute’s calculations, the policy keeps 41,279 Canadian households below the low-income threshold (MEI, 2026).

The second to pay is the young farmer. In Québec, the right to produce one kilogram of butterfat per day has been capped at $24,000 since 2015 (UPA Montérégie). A 100-kilogram quota therefore costs $2.4 million before buying a single cow. Supply management does not protect “farmers”; it protects those already inside, against those who would like to get in.

The third to pay is the rest of the economy. In June 2025, Parliament passed Bill C-202, introduced by the Bloc Québécois, which forbids any trade concession on these tariffs (Parliament of Canada). Just as Canada was renegotiating with the United States, the C.D. Howe Institute observed that the law ties the hands of our negotiators (C.D. Howe, July 2025). Aluminum, lumber and autos now negotiate with one arm tied behind their back.

This is not solidarity. It is a private rent guaranteed by the state, paid for in part by the single mother buying milk. The seminar sees the farmer it protects; it does not see the shopper paying at the checkout.

A world without bankruptcy

Behind the box of pencils lies a vision of a world where nothing fails. No stationery store closes, no farm disappears, no public investment is lost, no recession interrupts the promises. It is a comfortable world. It is not ours.

In a market economy, failure is information. The store that sells pencils that break loses its customers to the one across the street, and money flows to where it is better used. Joseph Schumpeter called it creative destruction: the disappearance of those who serve badly is the price of the arrival of those who serve better.

A single supplier chosen by tender has no such problem. Its pencils can break; it keeps the contract until the next call for tenders, and the child keeps the broken pencils. The dissatisfied parent no longer has a store across the street. He has a complaint form.

Investment obeys the same law. A venture capitalist knows that most of his bets will fail and that a single success will pay for the rest. The state cannot admit that it is betting. Every project is presented as a sure success, until the day bankruptcy comes anyway and no one is responsible. Denying failure does not prevent it; it only prevents it from teaching anything.

Then there are cycles. Universal promises are made in times of growth and paid for in times of recession. In the fall of 2025, Québec was already forecasting a $12.4 billion deficit for 2025–2026, or 1.9% of GDP (CFFP, December 2025). That is the context in which the campaign paraded its new universal programs. A family that reasoned this way would be called reckless; a party that does it calls itself generous.

Finally, everything has a cost. This is not ideology; it is arithmetic. Every notebook handed out for free was made, shipped, stored and paid for by someone. Refusing that arithmetic is not compassion. It is the luxury of those who never see the bill arrive at their own door.

The bill the central neighbourhoods never see

On October 5, Québec solidaire won nine ridings, mostly in Montréal. There its ideas seem self-evident, because people live surrounded by services whose payer they never see. They get around on foot, by bike, by metro; the metro seems as natural as tap water.

It is not. Public transit in the metropolitan region is funded by riders, by municipalities, and by motorists across the region, through fuel and vehicle registration taxes. Since January 1, 2025, every owner of a passenger vehicle in the Communauté métropolitaine de Montréal has paid a $150 registration tax instead of $59, to cover $361 million of a $561 million transit deficit (Noovo, May 2024).

The person paying that tax is often someone living in an outlying borough, in Laval or on the South Shore, who drives because no metro runs near their home. They fund the line that someone without a car takes downtown. The leader of the opposition at city hall, Aref Salem of Ensemble Montréal, said at the time that this “draconian increase punishes drivers who have no other choice when it comes to public transit” (Noovo, May 2024).

Québec solidaire’s response to the increase is telling. Its MNA Alexandre Leduc dubbed it the “Guilbault tax,” after the transport minister, and called for Québec to pay more instead (Noovo, May 2024). In other words: not the driver from Laval, but the taxpayer from all of Québec. In the same report, the minister, Geneviève Guilbault, put it this way: “All of Québec’s regions cannot systematically absorb half-billion-dollar deficits for public transit in Greater Montréal.”

The arts follow the same logic. Since the March 2025 budget, the Conseil des arts et des lettres du Québec has $200 million a year in program funding for three years; the programs of the SODEC, the provincial agency for cultural businesses, get more than $125 million for 2025–2026 (Québec budget plan, March 2025). Add to that the Canada Council for the Arts and the Conseil des arts de Montréal.

I am not saying artists live in luxury. Most earn little, and they said so in the streets in 2025. I am saying that a jury’s judgment has replaced the public’s. A work few people want to see can be funded; a work many people want to see must first convince a committee. That is not a crime. It is a privilege, paid for in part by people who will never set foot in those theatres.

The same goes for subsidized daycare, free schooling, health care and family allowances. These gains do not hold because the central neighbourhoods produce them. They hold because the whole economy, its mines, forests, factories, truckers and offices, generates enough surplus to fund them without much thought. And by not thinking about it, people end up mistaking a transfer for a right.

Generosity with other people’s money

Back to the box of pencils. Québec solidaire’s promise is not absurd because it is generous. It is absurd because it confuses generosity with control. A cheque tells parents: here is some help, you know better than we do what your child needs. A box tells them: here is what you will get.

Supply management, the box of pencils and the services of the central neighbourhoods have one thing in common. Each is a benefit visible to whoever receives it and invisible to whoever pays for it: the single mother at the grocery store, the parent who would have chosen differently, the driver from Laval, the taxpayer in Rimouski.

In 2024, Québec solidaire wanted to double the cheque; in 2026, it wanted to replace it with a parcel. Between the two proposals, only one thing stayed the same: neither began by asking who would pay.

A party that wants to govern for everyone should start by counting those who pay. True generosity is done with one’s own money. Generosity with other people’s money goes by another name: power.

Sources

Further reading: Ryan Cardwell et al., “Milked and Feathered: The Regressive Welfare Effects of Canada’s Supply Management Regime,” Canadian Public Policy, vol. 41, no. 1, 2015; Joseph Schumpeter, Capitalism, Socialism and Democracy (1942); Commission of Inquiry on the Awarding and Management of Public Contracts in the Construction Industry (Charbonneau Commission), final report, 2015; Frédéric Bastiat, “What Is Seen and What Is Not Seen” (1850). Quotations from French sources are the author’s translations.

Want to argue back? Here is your ammunition:

  • Drivers get their own free ride: nobody ever mails them the bill for traffic jams and exhaust.
  • The French-language market is only so big; without grants, much of this work would never exist. That is a market failure, not a privilege.
  • Supply management costs the treasury nothing and keeps farms and prices steady. American farmers can only dream.
  • Group purchasing gets volume discounts; hand out cheques, and who knows whether the $127 ends up as pencils or beer.
]]>Sat, 10 Oct 2026 12:00:00 GMTWhat do I believe?https://elias-essays.pages.dev/en/blog/what-i-believe/https://elias-essays.pages.dev/en/blog/what-i-believe/I believe human beings have enormous creative power, but in the end we are not God.A while ago I came across a political test called 12 Axes on X and took it out of curiosity.

The results were interesting. On the economic and technology dimensions I ended up near the outer edge of the chart: over 90% for private ownership, more than 80% on free markets, free trade and individual liberty, and a full 92.5% on technological acceleration. On the social and ethical dimensions, though, I landed on the traditionalist side (traditional morality took two thirds), and on the religion axis, secular and faith split almost exactly down the middle.

In today’s highly polarized politics, an algorithm would probably have a hard time giving this result a ready-made camp label. Neoliberals would find my attachment to morality and tradition too conservative. Old-school conservatives wouldn’t like how radical I am on global free trade and new technology. And committed secular progressives would probably understand even less why someone who writes code and builds products every day, and believes in market competition, holds religion in such deep reverence.

To me, though, these twelve dimensions are not contradictions stitched together. They point to one coherent philosophy of life: I believe human beings have enormous creative power, but in the end we are not God.

The limits of power and the cost of monopoly

Not long ago I had to correct a company’s registration information in Quebec. It was a simple change, yet the government system required me to print a paper form, send it by registered mail from the post office, and then spend weeks checking again and again whether it had actually arrived. The registered-mail tracking showed it had been signed for, but when I called the office, they still told me they had no record of it.

All the time, effort and cost of fixing the error fell on the ordinary applicant.

I don’t deny that the public sector has legitimate functions. Courts, property rights, contract enforcement and public safety are the minimum a civilized society needs to function. But administrative power has something no private company will ever have: it is compulsory and exclusive. If a courier company or a software vendor gives me terrible service, I can replace it the next day. If a government department is rigid and has made red tape its reason for existing, I have no alternative at all.

Max Weber defined the state by its monopoly on the legitimate use of force. Precisely because it holds that monopoly, its power, more than that of any other institution, needs to be kept in a cage.

Most regulations start from goals that sound beyond reproach: protecting jobs, preserving fairness, reducing market risk. But good intentions never guarantee results. Every new approval requirement quietly raises the bar for ordinary people trying to get something done, and every administrative licence eliminates potential competitors on some group’s behalf. What calls for the most vigilance is the bureaucracy’s built-in tendency to grow. When an intervention fails, the failure is seldom treated as evidence that power overstepped. Instead it becomes the excuse to create a new department and pass more regulations to “fix the previous problem.”

Government should stay within its limits. It should mind its own business and do well the few things it really ought to do.

Wealth comes from trial and error, not privilege

I support the market economy not because business people are naturally virtuous, but because human society is too complex for any central committee to be smart enough to hold all the information needed to run an entire economy. This is the knowledge problem that Hayek identified.

How many people a company should hire, what a product is worth, whether a new technology is worth investing in: these judgments require countless people to keep trying and failing in response to price, profit and loss signals. The great achievement of the price mechanism is that it pools the intuitions scattered across millions of minds and makes blind decisions bear their own costs.

But protecting the market does not mean shielding vested interests.

A company that wins with better technology and lower costs deserves everyone’s respect. If instead it lobbies politicians to push out rivals, or relies on government subsidies to keep an inefficient business alive, then however private it is on paper, at bottom it is the well-connected dividing up the spoils.

Money and trade follow the same logic. A person’s savings are the fruit of a lifetime of work and should not be left to a single central bank to dilute at will behind closed doors. Competition among private currencies, crypto-assets and the existing payment systems should have room to exist legally.

The same goes for tariffs and trade protection. Politicians like to package them as “protecting domestic industry,” but in the end the tariff bill is paid by consumers and downstream businesses across society. Forcing everyone to put up with poorer, pricier goods for years will not, by itself, make a lagging industry competitive. A country’s prosperity depends, in the end, on how much value it creates, not on how high its politicians build tariff walls.

Democracy means checks on power, not winner-take-all

Watching elections in Quebec, what struck me most is how often the share of votes and the share of seats come apart. A party that wins hundreds of thousands of votes and has broad public support can end up without a single seat in the legislature, while a governing party that barely reaches 40% of the popular vote can hold an outright majority.

This is not healthy. If a system distorts voters’ real political choices, it is unfair, even when the party I support happens to benefit in the short term.

Winning an election is not a blank cheque to do as one pleases. Independent judges, news media willing to ask hard questions, the opposition’s right to question the government, and local self-government are not obstacles to administrative efficiency. They are the brakes that stop the tyranny of the majority.

I also support pushing power down to the local level as far as possible. People who live in Montreal or Quebec City usually understand the problems in their own neighbourhoods better than federal bureaucrats in Ottawa. If Quebecers one day decide their political future through a clear and transparent democratic process, that choice should be respected. But the same principle can’t be brought out only when it works in one’s own favour. The First Nations who live on the same land must also have a real voice, a right to self-determination, on major matters that concern their lands and their future.

The real test of democracy has never been cheering when you are on top. It is choosing to hold the line when your principles may cut against your own interests.

A community thrives by attracting people, not by building walls

In a North America where English is overwhelmingly dominant, the cultural anxiety of a French-speaking society is not made up.

Living in Montreal, I feel first-hand the urgency with which Quebec wants to protect its language and traditions. When immigrants come to a new country, making the effort to learn the local language, understand its institutions and norms, and take on responsibilities in their community is the good faith that joining a society calls for.

But how a culture is protected matters a great deal. A thriving French-speaking society should draw people to speak French through excellent films, competitive businesses and open universities, not through ever stricter administrative laws that take away young people’s option of pursuing higher education in English. Sustaining your own existence by steadily cutting back other people’s choices tends to buy only an illusory sense of security.

A mature citizen can speak fluent French and obey Quebec’s laws while keeping the memory of their mother tongue and their family traditions. A living community rests on shared responsibilities and hopes for the future, not on checking people’s bloodlines or requiring everyone to think in lockstep.

Mortals must not overstep: reverence for morality and a ceiling on power

While many people my age have drifted into outright nihilism or consumerism, I have, over the past few years, become more and more interested in the Catholic tradition.

This doesn’t mean I want to impose any religious doctrine on anyone, and I have honest respect and tolerance for other faiths too. But I have come to see more and more clearly that human beings can’t live by calculation alone; we also need reverence. The human world can’t consist only of power and efficiency. It needs a moral ceiling that stands above any earthly government.

The most painful lesson in the whole history of civilization is that people keep imagining they can play God with their limited reason.

Ever since the ancient Tower of Babel, whenever people have set their hearts on building “a tower with its top in the heavens” to “make a name for ourselves” (Gen 11:4), whenever they have tried to build an all-knowing, all-powerful perfect order on earth, it has ended, almost without exception, in a hell on earth soaked in blood. Hölderlin gave the same warning in Hyperion: those who set out to make the state a heaven end up making it a hell. Which of the twentieth century’s utopian fevers, pursued in the name of “scientific socialism” or a “paradise on earth,” did not start with a perfect design for human nature and end with secret police and concentration camps?

Human beings are finite creatures, born with blind spots and weaknesses. Once we admit that, we have to admit that every earthly power has limits it cannot cross.

The Book of Ecclesiastes has a very sober line: “for God is in heaven, and you upon earth; therefore let your words be few” (Eccl 5:2). The Book of Micah also reminds us of what is required of us: “to do justice, and to love kindness, and to walk humbly with your God” (Mic 6:8). In secular politics, this is the strongest line of defence against totalitarianism: no leader, party or state machine can be allowed to take God’s place. Mortals must not overstep; there is no all-knowing, all-powerful saviour on earth. Once we let the state reach into people’s consciences and souls, totalitarianism is already knocking at the door.

This reverence for moral authority also sets my bottom line on specific social issues.

On abortion, I know my conservative position conflicts with the absolute bodily autonomy that many people champion. But precisely because life is sacred and not an appendage of power, I cannot see an unborn fetus as merely tissue to be disposed of at will. The law should place strict limits on abortion, with clear exceptions, of course, for the heartbreaking extremes of rape, incest and a direct threat to the mother’s life. This is a heavy ethical choice. We have to face the moral tension honestly instead of pretending that the cost to life doesn’t exist.

The same logic applies to the right to bear arms. Public safety is vital, but preventing crime cannot rest on presuming every law-abiding citizen to be a potential criminal. Owning a weapon is a free person’s exercise of the right to self-defence, and it comes with serious adult responsibility. A free society is built on trust in the self-discipline of law-abiding citizens, not on disarming everyone and keeping them like overgrown infants under the state’s lifelong guardianship.

Strength is armour for defence, not a drug for expansion

On foreign policy and defence, I subscribe to the philosophy of armed neutrality.

A country must have credible armed forces to defend its territory and the safety of its people. Pacifism without the means to defend yourself is a fantasy that leaves you at others’ mercy. In the face of naked aggression and the atrocities of despots, it is both legitimate and necessary for the international community to give those who resist weapons and economic aid.

But the military is a shield, not a bargaining chip for politicians’ grand imperial dreams. A country should do everything it can to stay out of foreign conflicts that have nothing to do with its own survival. It must not, for the sake of vague geopolitical ambitions, throw young people’s lives and taxpayers’ hard-earned money into the quagmire of wars whose purpose no one can explain and whose cost no one can estimate.

The dignity of sovereignty lies in being able to decide your own fate, not in becoming the vassal of some hegemonic system.

Technology should free individuals, not enslave ordinary people

If anything in my views looks especially radical, it is my optimism about technology.

I use artificial intelligence myself to write code and build products. In the past, developing a finished piece of software meant putting together a full team with a product manager, front-end and back-end developers, and testers. Today an independent developer sitting in a home office in Montreal can, with AI as leverage, turn out on their own what a small start-up used to produce.

This tears down the barriers of power and organization completely. Technology is handing the creative capacity once monopolized by big companies, academic credentials and huge amounts of capital back to ordinary individuals.

Disruptive technology does bring painful transitions, of course. It wipes out old jobs and may even widen information inequality. But the right response to that shock is to draw clear lines of conduct through specific laws and institutions, not to stop the gears from turning out of fear of the unknown.

On the environment and energy, I particularly dislike the “degrowth” environmental narrative that asks humanity to tighten its belt and go back to material scarcity. Whale-oil lamps weren’t phased out because a law banned them. Cheaper kerosene replaced them, and electric light later replaced kerosene. More abundant nuclear power, more advanced materials and more efficient computing are the only answer to resource constraints and climate problems.

The real mission of technology is to expand material abundance and freedom of choice, not to leave people suspicious of one another in scarcity.

Conclusion

Looking back at the 12 Axes chart, I’m actually quite at peace with the “doesn’t fit in” verdict it gave me.

In this black-and-white era, people are too used to lining up in preset camps. Those who talk about markets and technology often look down on tradition and faith, while those who cherish family and morality are often afraid of freedom and progress.

But the two never had to be at odds.

I want government to show humility and restraint, and to take its hands off the market and people’s private lives. I value traditional moral anchors and reverence for the sacred, because I know that reason with no bottom line only leads people to arrogance and destruction. And I embrace technological acceleration, because I believe it can give ordinary people more tools to change their lives.

A healthy society has no need for the state to lay out the same life template for everyone, and it certainly cannot let any earthly power try to build a so-called “kingdom of heaven on earth.” All it needs is to uphold just laws, draw clear lines around power, and then step back, handing the right to shape their own life, faith and future, whole and intact, back to every humble and free person.

Note: Bible quotations are from the Revised Standard Version.

]]>Sat, 10 Oct 2026 12:00:00 GMTEvery riding has a member, so why is the National Assembly still skewed?https://elias-essays.pages.dev/en/blog/quebec-election-2026-representation/https://elias-essays.pages.dev/en/blog/quebec-election-2026-representation/The Parti Québécois won nearly half the seats with 28% of the vote. The CAQ won 13% and no seats at all.What I keep coming back to in this Quebec election is that the votes and the seats don’t match.

In a single riding, first-past-the-post (FPTP), where the candidate with the most votes wins, sounds simple enough: whoever comes first goes to the National Assembly to represent the area. But add up the results of all 127 ridings and the Assembly no longer looks like the way Quebecers voted.

Here are Élections Québec’s counts as of 2:32 p.m. on October 7 (results are not yet official; seats are attributed to the candidate leading in each riding):

Party Share of valid votes Seats Share of seats
Parti Québécois (PQ) 28.02% 59 46.46%
Quebec Liberal Party (PLQ) 24.23% 40 31.50%
Conservative Party of Quebec (PCQ) 20.66% 19 14.96%
Coalition Avenir Québec (CAQ) 13.35% 0 0.00%
Québec solidaire (QS) 12.42% 9 7.09%
Other parties and independents 1.32% 0 0.00%

Calculated from data published by Élections Québec. Vote shares use all valid votes as the denominator; seat shares use 127 seats. Totals may not add up exactly because of rounding.

With 28% of the popular vote, the PQ won nearly 46.5% of the seats, five short of the 64 it needs for a majority on its own. The CAQ, the outgoing government, got 572,000 votes (13.35%) and not a single seat. No party has a majority, but that does not make the rules fair.

Even if you don’t support the CAQ, it is hard to call this result reasonable. More than half a million people voted, and nobody in the Assembly speaks for them. Whether a set of rules is fair should not depend on whether this time’s winner is the party you like.

Now look riding by riding. In 100 of the 127 ridings, the winner got less than half the vote. Taken together, the 127 winners received only 41.73% of all votes cast. The other 58.27%, close to 2.5 million votes, did not translate into a single seat.

Supporters of FPTP often say that once elected, a member represents every voter in the riding, not just one party. That is true. What it leaves out is that on the question of who forms government and who holds power, the rules filter out the choices of more than half the voters. The claim that each area elects the person who best represents it no longer holds up when several parties compete. The current rules only decide who scooped up the most chips in the scramble. Whether the Assembly reflects the range of political opinion across Quebec is not something they deal with.

People often say FPTP is the natural foundation of parliamentary democracy. Yet among developed democracies, few still elect their national legislature by this rule alone, and several countries that share the Commonwealth tradition have moved away from it. Australia dropped FPTP in 1918, switching to preferential (ranked) voting for the House of Representatives, and the Senate moved to the single transferable vote (STV) in 1948. New Zealand, after seeing similar gaps between votes and seats, voted in a referendum in the 1990s to abolish the FPTP system it had inherited from Britain and adopted mixed-member proportional representation (MMP). When the United Kingdom set up the Scottish Parliament and the Welsh Assembly in 1999, it also chose a mixed system.

France, the country across the Atlantic whose people Quebecers call their “cousins” (nos cousins), used proportional representation under the Fourth Republic. The National Assembly of the Fifth Republic is elected in two rounds: to win in the first round, a candidate needs more than half the valid votes, and those votes must add up to at least a quarter of registered voters; otherwise voters go back to the polls, and in that second round the candidate with the most votes wins, majority or not. European and regional elections use proportional representation. Quebec’s civil law comes from the French tradition, and Quebecers have long defended their distinctiveness (société distincte) in language, culture and law. So what moral reason does Quebec have to keep its democracy tied to the Westminster rules that British colonial authorities brought over in the 18th century?

Closing the gap means bringing in some degree of proportional representation (PR). Riding representation and proportional representation can coexist. An MMP system keeps riding members, and with them the local connection, while adding compensatory party seats that bring each party’s total close to its share of the vote.

Reform does come with technical hurdles and costs: how large the share of compensatory seats should be, how to handle vast, thinly populated regions, how to keep party lists from being drawn up behind closed doors, and the adjustment to coalition governments becoming the norm. These concerns are reasonable. But these costs weigh less than the legitimacy problem created when nearly 60% of votes win no seats. Public opinion in a modern society is deeply divided. The Assembly’s job is to show those divisions as they are and look for compromise from there. A majority manufactured by the voting rules cannot do that.

On page 15 of its 2026 platform, the PQ did commit in black and white to electoral reform, with a proportional element. But it has just turned 28% of the vote into nearly 46.5% of the seats. How much incentive does a party that has just won big under the current rules have to take apart the ladder it climbed?

In 2019, the CAQ also tabled an electoral reform bill, Bill 39. It was adopted in principle in 2020, then shelved once the party’s hold on power was secure.

Unless citizens and civil society keep up the pressure, the PQ’s promise could well end up as one more scrap of paper. To judge whether the PQ means it, look at two things: whether, once in office, it puts a timeline into law, and whether its proposal includes real compensatory seats or just tinkering.

]]>Wed, 07 Oct 2026 12:00:00 GMTWhat is left of the Parti Québécois’s independence ideal?https://elias-essays.pages.dev/en/blog/quebec-sovereignty-without-a-future/https://elias-essays.pages.dev/en/blog/quebec-sovereignty-without-a-future/Where the PQ’s case for a new country falls short: its vision, its record on governance, and its positions on Indigenous nations, immigration and language.The Parti Québécois has won the election again. But I increasingly feel that the kind of Quebec independence it stands for has lost the pull it once had.

I think Quebecers have the right to decide their own future, and that is not what this piece is about. The question went to voters in 1980 and again in 1995, and in its 1998 reference opinion the Supreme Court of Canada recognized that a clear majority on a clear question would give rise to a duty to negotiate.

A party can return to power, hold another referendum and keep rallying its supporters around the same flag. But to found a country worth founding, sovereignists still have to explain what, apart from leaving Canada, they intend to build together.

The sovereignist case, at its strongest

Most of what worries sovereignists has a real basis.

French speakers make up about 2 percent of North America’s population. The 2021 census found that the share of Quebecers who mainly speak French at home fell from 79 percent in 2016 to 77.5 percent, and the shift was sharper in Montreal.

The constitutional wound is real too. The 1982 Constitution was adopted without the consent of Quebec’s National Assembly, and Quebec has still not signed it. The Meech Lake Accord (1990) and the Charlottetown Accord (1992) both tried to repair the breach, and both failed. In 1995 the No side won by 50.58 percent to 49.42. Many sovereignists take all this as proof that the federation cannot accommodate Quebec as an equal.

Seen this way, much of what today’s PQ proposes, such as lower immigration, a stronger Bill 101 and a wary eye on Ottawa, has roots. It grows out of a minority’s long-standing fear about whether it can survive in North America, and the history and numbers above show that the fear is well founded.

A business plan that is all defence

In 2020, according to The Canadian Press, Paul St-Pierre Plamondon (PSPP) called himself an “entrepreneur en pays,” a founder whose startup is a country, and added that the joke said something about the mood of the moment. Likening the founding of a country to launching a business is an appealing image. But a startup rarely survives by guarding against the incumbent platform and listing its rivals’ faults. Investors and customers want to know what it can offer that nobody else can.

Today’s PQ has become very good at explaining what it wants to guard against: interference from Ottawa, the influence of English, and the strain immigration puts on public services. Each of these deserves debate. Put together, though, they still do not describe a new society that would be livelier, more open or more appealing than staying in Canada. That is not enough to persuade millions of undecided voters to give up the room Canada offers and take on the risk of building a country.

The Quiet Revolution, more than half a century ago, was a high point of modern Quebec history, and the people who led it were willing to build. Jean Lesage, René Lévesque and Georges-Émile Lapalme campaigned on Maîtres chez nous (“masters in our own house”), and then they got to work.

They brought the private power companies into Hydro-Québec, which put French-speaking engineers and technicians at the centre of a modern industry for the first time. They founded the Caisse de dépôt et placement du Québec to give the province its own pool of capital. They created the CEGEP system in 1967 and the Université du Québec network in 1968, ending the Church’s hold on education and giving young francophones from every class and region a way up.

That generation protected French by making it the language of engineering, finance and the university.

Neither the CAQ, just voted out, nor the returning PQ has shown anything like that ambition. Many of the institutions that broke new ground back then have since become shelters for vested interests and bureaucratic habit.

The Quebec model is showing its age

Statistics Canada data put Quebec’s labour productivity below Ontario’s and the Canadian average in every year since 1997: in 2024 an hour of work in Quebec produced about $59 of output, against about $62 in Ontario (in 2017 dollars). The per capita GDP gap with Ontario shrank from 12.2 percent in 2015 to 8.6 percent in 2023, but HEC Montréal’s productivity centre puts that down mostly to Ontario’s own weak performance, not to any leap forward by Quebec.

To pay for its large public services, Quebec carries one of the highest personal tax burdens in North America. Even so, its tax base and output are not big enough, and it still relies on more than $13 billion a year in federal equalization payments to balance the books. With independence, that money would be gone. How to close the gap, with productivity lagging and taxes already high, is the first question the PQ’s plan for independence has to answer.

In 2025, Quebec’s auditor general looked at CASA, the IT project behind SAAQclic, and found it had cost nearly $500 million more than planned, with the total expected to pass $1 billion. Accountability has been slow to follow. Back in 2005, former PQ premier Lucien Bouchard argued in the manifesto Pour un Québec lucide that whatever constitutional path Quebec chose, it would not prosper unless it dealt with productivity, debt and the efficiency of government.

The Quebec model has its achievements. Affordable, universal child care and relatively low income inequality are real successes, and the child care program was created by a PQ government in 1997. But a system that has trouble holding its own institutions to account, and whose productivity keeps lagging, will not become a more prosperous country by raising a new flag. If the PQ cannot offer an economic plan that improves on the status quo, its “business plan” for a country will look thin to people adding up their monthly bills.

Referendum night

If I had to pick one moment that marks the turn, it would be October 30, 1995.

On the night of the defeat, Jacques Parizeau blamed the result on “money and some ethnic votes.” In the same speech he used “we” to mean the francophones who had voted Yes, and suggested that a bigger share of the francophone vote would be enough next time.

A movement that said it was founding a country for everyone who lived there responded to defeat by sorting people by origin into “us” and those in the way. The people who voted No lived in Quebec too, paid the same taxes and carried the same civic duties. Putting them in a separate category amounted to taking back the promise of a country built together.

Many sovereignists later distanced themselves from those words, and Parizeau announced his resignation the next day.

Indigenous nations: the PQ used to do better

On Indigenous nations, today’s PQ is especially disappointing, because its own record used to be better.

In 1985, under René Lévesque’s PQ government, the National Assembly passed a motion recognizing the Indigenous peoples of Quebec as “nations.” In 2002 another PQ premier, Bernard Landry, signed the Paix des Braves with the Grand Council of the Crees, an agreement that dealt with the two sides as nation to nation.

Compare that with September 30, 2026, when PSPP said that under the interim constitution planned for the transition to independence, First Nations and Inuit would be consulted but would not have a veto. His reasoning was that an interim constitution and a permanent one have different approval thresholds. That distinction can be argued on procedural grounds. But a movement whose central cause is national self-determination will have a hard time explaining why other nations, whose ties to the land go back much further, should only be “consulted” when a new country is founded.

Indigenous peoples are nations, as the PQ’s own government recognized in 1985. If one side has already set the framework for the new country on its own, it is hard to see how consultation would be more than a formality. The PQ platform does promise support for Indigenous language and cultural education, which is to its credit. But cultural support is no substitute for making decisions together. A new country that began by repeating the pattern of the majority deciding for the minority would end up on the very side Quebecers have long opposed.

Québec solidaire, which is also part of the sovereignist camp, has tried something different. In its plans for a constituent process, it has at least explored writing a constitution on an equal footing with First Nations and Inuit. How that would work is still disputed, but it shows that the sovereignty movement can take a more open approach on this question.

Where newcomers fit

On immigration, the PQ’s story is just as defensive.

Housing shortages, a health system with limited capacity and the resources needed for francization are all legitimate reasons to talk about immigration levels, and wanting to change them is not xenophobic in itself. The trouble is that in the PQ’s immigration plan, newcomers come up mainly as a strain on capacity and a risk to French. The party explains in detail how it would set limits and test people, but says little about how they would take part as equals in designing the new country’s institutions, or share in its opportunities. If a common future asks those who arrive later to quietly fit into an identity that has already been decided for them, it will be hard to get people of every origin behind the project of building a country.

Language and the CEGEPs

Protecting French in North America, so that people can work, study and live in French with dignity, is a legitimate goal, and I don’t dispute it. Bill 101 sent the children of immigrants to French-language schools and changed the linguistic make-up of Montreal, and that contribution should be acknowledged.

The PQ’s plan to extend Bill 101 to the CEGEPs has its own logic. Many young people from francophone and immigrant families switch to the English system at the CEGEP level, and their working and social lives then tilt toward English as well. Bill 96 already caps enrolment at English CEGEPs; the PQ says that is not enough.

I still think it is the wrong direction. The previous government sharply raised tuition for students from other provinces and clawed back most of the tuition paid by international students, and the English universities took the brunt. McGill estimated in 2023 that the two measures, along with the expected drop in enrolment, would cost it $42 million to $94 million a year; Concordia attributes its run of deficits mainly to the new funding formula for international students and to immigration restrictions. On page 40 of its 2026 platform, PSPP’s PQ promises to extend Bill 101 to the CEGEPs, as well as to secondary-level vocational training and adult education. That would mean young people, just as they become adults and choose a CEGEP or a trade, would lose the option of studying in English.

Fluent English matters today for anyone who wants to take part in research, the global exchange of ideas or international trade. Figures cited in a House of Commons science committee report show English’s share of scholarly publishing worldwide rising from 64 percent in 1995 to more than 90 percent in 2019. A French-speaking society that is confident in its own culture can treat its young people’s bilingualism, or multilingualism, as an advantage in the wider world. The voluntary agreements on French that the Quebec government signed with the English universities in June 2026 show that French can be promoted without bans. Restricting choice by decree costs more, and is less likely to win young people over.

Conclusion

The openness and drive the PQ once had are getting harder to find in today’s party. A movement that can say clearly what it wants to hold on to, but not what it wants to create, is running out of ideals, even if the swing of the electoral pendulum has put it back in office.

Quebecers’ worry about their survival is real. But people of every background who live in Quebec should not have to pay for a national project built mainly on defence. If the PQ wants people to believe in independence, it will have to do what the Quiet Revolution generation did and say clearly what it intends to build.

]]>Wed, 07 Oct 2026 12:00:00 GMT