Monuments to Tolerated Error
On the difference between opinion and judgment
In November 2024, California voters approved Proposition 36, increasing criminal penalties for theft and drug offenses. The measure passed with 70% support. It reversed much of Proposition 47, which California voters had approved ten years earlier with 60% support.
No committee deliberated. No amendments were offered. No one sat in a room and said, “What if we toughened some penalties but not others? What about the drug treatment provisions: can we keep those while addressing the theft problem?” The question was binary. Yes or no. The voters spoke. The nuance didn’t.
This is how we govern now. Through the clean binary of the referendum. Thumbs up. Thumbs down. I picture Commodus.
The legislature is too slow, too captured, too compromised. So we go around it. We take the question directly to the people.
It feels like more democracy, maybe. It is less republic.
The founders were specific about the distinction.
Madison, in Federalist 10, designed the republic explicitly to resist what he called the “violence of faction”: the tendency of passionate majorities to steamroll minorities and mistake collective feeling for collective wisdom. His solution was to refine popular will. Pass it “through the medium of a chosen body of citizens, whose wisdom may best discern the true interest of their country, and whose patriotism and love of justice will be least likely to sacrifice it to temporary or partial considerations.”
Madison worried about both directions. Minority factions could capture the machinery of government and obstruct from within (the filibuster, the committee bottleneck, the organized interest that blocks what most people want). We see this everywhere now: small, disciplined groups wielding procedural tools to veto what majorities support. The frustration with this obstruction is bubbling up, and it’s one reason people reach for the referendum. If the legislature won’t act, go around it.
But Madison understood something we keep forgetting: opinion and judgment are different things. Opinion is what I think right now, based on what I know, colored by what I feel. Judgment is what emerges when opinion meets counter-argument, competing evidence, and the discipline of having to govern the consequences. The answer to minority obstruction is to fix the deliberative process, not to abandon it.
The representative doesn’t just carry the people’s opinion to the capitol. The representative deliberates. Sits in committee with the opposition. Hears what the other district needs. Amends the proposal. Finds the version that addresses the problem without creating three new ones. The deliberation is the point. It’s where raw opinion gets refined into judgment through the friction of competing perspectives.
Burke made this even sharper. Speaking to his constituents in Bristol in 1774, he told them something no modern politician would dare: “Your representative owes you, not his industry only, but his judgment; and he betrays, instead of serving you, if he sacrifices it to your opinion.”
The representative who becomes a weathervane, who polls before every vote, who treats constituents’ preferences as binding instructions: this person has kept the office but abandoned the function.
The binary everywhere
Social media is a constant referendum. Every issue becomes a poll. Every position a signal. The algorithm rewards the stark binary, not the careful “it depends.” Twitter threads don’t have an amendment process. Instagram polls don’t allow counter-proposals. The format demands thumbs up or thumbs down, and we’ve internalized the format so deeply that we expect governance to work the same way.
When I look at my stats for these essays, that too is a referendum. I learn where people are willing to put their time. This essay likely won’t do well.
The revolt against institutional authority happens through tools of negation. The public can swarm, humiliate, and vote out. It cannot amend, deliberate, or build. Polling is negation dressed as participation. The information revolution gave everyone a voice, but the voice can only say yes or no. The kind of voice that says “what if we tried this instead” requires a room, a process, and time. We have less patience for all three.
Then the ballot initiatives. California has put hundreds on the ballot since establishing its initiative process in 1911. Each one reduces complex policy to a binary question. Do you want lower drug penalties? Yes or no. Do you want higher minimum wage? Yes or no. Do you want the gig economy regulated? Yes or no. Each question has a clean answer. But the clean answer has consequences the question doesn’t capture: trade-offs that only emerge in deliberation, second-order effects that nobody votes on because nobody is asked.
Prop 13 capped property taxes in 1978 and reshaped California governance for fifty years. Its consequences (funding crises in schools and local government, distorted housing markets, a shift of power from localities to Sacramento, creating a moat around certain families and not others) were not on the ballot. The question was: do you want lower property taxes? The answer was yes. The implications were a different question entirely, intense inequality, generational conflicts, and more. Nobody voted on those.
On the left, progressive ballot initiatives bypass legislatures to encode specific social visions directly into law. On the right, populist movements demand that representatives stop deliberating and start obeying. Both treat the representative as obstacle rather than tool. Both want the frictionless efficiency of direct will. The MAGA movement’s contempt for Republican moderates who “don’t fight” and the progressive left’s contempt for Democratic incrementalists who “don’t deliver” share the same structural complaint: stop deliberating and do what we want.
Same mechanism. Different content.
The school we’re closing
Tocqueville spent two years studying America and concluded that voluntary associations and local self-governance were the republic’s load-bearing walls. Citizens who join a school board, sit on a jury, negotiate with neighbors about the zoning variance: these people are practicing deliberation. They’re learning that their perspective is partial. That the other side has reasons. That the workable solution usually lives in the friction between competing interests. The township was the school of citizenship.
Take away the deliberative step, and you take away the school.
The farmer knows the land. The nurse knows the ward. The shopkeeper knows the block. A ballot initiative asks California voters to decide the fate of the state’s water policy, but the knowledge needed to make that decision well is distributed across thousands of farmers, engineers, ecologists, and municipal planners. None of it fits into a yes or no. It can only enter the system through a process where people speak, listen, amend, and compromise. Hayek called this the knowledge problem: the knowledge necessary for wise governance doesn’t live in any single person or faction. It emerges through the friction of deliberation.
Representative institutions exist because the process of debate and amendment generates insight that no poll, no referendum, no prediction market can access. Polling counts existing opinions. Deliberation produces new understanding.
Plebiscitary culture collapses a distinction the republic was built to maintain: the difference between a difference of opinion and a difference of principle. When every issue is a referendum, every disagreement becomes existential. There’s no space for “I see your reasoning but I think the trade-offs point differently.” There’s only yes or no. For or against. The capacity to treat disagreement as workable rather than threatening requires a process designed for that purpose. Referendums are designed for something else.
A society that only knows what it’s against finds plebiscitary culture suited to its needs. You can vote NO without offering an alternative. Negation scaled to governance.
I’m not arguing against citizen engagement. The founders built the republic on it. Washington and Franklin didn’t want a system designed for passive subjects. They designed one for active citizens who govern themselves through participation.
The question is what kind of participation.
Participation that submits to the discipline of deliberation (jury service, local governance, elected representation where the representative actually deliberates) builds the muscle of self-governance. It teaches us to hold our opinion alongside competing opinions and find the workable ground between them. This is hard. It’s slow. It produces outcomes that fully satisfy no one. It is also the only known process that converts raw popular will into governance that doesn’t consume itself.
Participation that bypasses deliberation (the poll, the referendum, the social media pile-on, the demand that representatives become mirrors rather than filters) feels more direct. More democratic. More responsive. But it replaces the process that generates judgment with the process that merely counts opinion. And opinion without deliberation is what Madison feared most: faction unchecked by the friction that forces it to reckon with what it doesn’t know.
The drift is real. We’re all doing it. We treat polls as mandates. We expect representatives to “fight” rather than deliberate. We reduce complex questions to binaries and call the result the voice of the people. I do it too. I catch myself wanting the clean answer, the decisive action, the leader who just does something. The patience required for deliberation feels like weakness. It isn’t. But it feels that way, and feelings are what plebiscitary culture runs on.
Back to California. In 2014, sixty percent of voters said yes to lower penalties. In 2024, seventy percent of voters said yes to higher ones. The same electorate, the same mechanism, opposite conclusions, ten years apart. No one sat in a room between those two votes and asked what changed. No one amended. No one deliberated. The people spoke twice, and the two answers canceled each other out.
Jefferson understood this better than we give him credit for.
He was the most democratic of the founders. The one who pushed hardest against Federalist caution, who trusted the people most. And in his First Inaugural, having won the most bitter election the young republic had ever seen, he said something that plebiscitary culture makes almost impossible:
“Every difference of opinion is not a difference of principle. We are all republicans: We are all federalists.”
Our disagreements about how to govern are not disagreements about whether to govern ourselves. The republic can contain differences of opinion because it has a process for working through them. That process is deliberation.
And then the less-remembered line: “If there be any among us who would wish to dissolve this Union, or to change its republican form, let them stand undisturbed as monuments of the safety with which error of opinion may be tolerated, where reason is left free to combat it.”
Monuments to the safety of tolerated error. The republic’s confidence in itself: we can afford to let wrong ideas stand because we have a process for testing them. Reason, given time and friction and the discipline of deliberation, will do its work.
Plebiscitary culture offers no such patience. It demands instant verdict. It builds guillotines of instant consensus where Jefferson would build monuments to tolerated error.
The people should govern. That was never the question. The question is whether we’re still willing to govern through the slow, maddening, indispensable process that converts what we think into something wiser than any of us thought alone.
The republic was designed to refine opinion into judgment. We are converting it back.
Sources and Inspiration
Democracy in America - Alexis de Tocqueville
A Conflict of Visions - Thomas Sowell
The Revolt of the Public - Martin Gurri