Censorship by proxy

Based on Wikipedia: Censorship by proxy

In early 2025, ABC abruptly cancelled Jimmy Kimmel Live. The late-night institution had been on the air for over two decades. What happened? The Federal Communications Commission had been publicly pressuring The Walt Disney Company, ABC's parent, about the show's political content. No law was passed. No court order was issued. Disney simply decided it wasn't worth the fight.

This is censorship by proxy in action.

The Art of Getting Others to Do Your Dirty Work

Censorship by proxy is what happens when a government wants to silence speech but doesn't want to do it directly. Instead of banning a book or shutting down a website—actions that would trigger constitutional protections in most democracies—officials pressure private companies to do the suppressing for them. Internet service providers. Social media platforms. Payment processors. Publishers. Web hosting services. Any link in the chain between a speaker and an audience becomes a potential target.

The mechanism is elegant in its deniability. The government can claim its hands are clean. After all, it was a private company that made the decision, right?

Seth Kreimer, a law professor who coined the term in a 2006 paper, put it this way:

"Rather than attacking speakers or listeners directly, governments have sought to enlist private actors within a complex chain of digital connections as proxy censors to control the flow of information."

Think of it like hiring a hitman. You're still responsible for the outcome, even if your fingerprints aren't on the weapon.

A Taxonomy of Pressure

How do governments actually pull this off? The methods range from blunt to subtle.

The most direct approach is legal liability. Governments pass laws that make platforms responsible for content they host or broadcast. Suddenly, the safest business decision is to remove anything remotely controversial. Why risk a lawsuit?

Then there's informal pressure—what scholars sometimes call "jawboning." A senator writes a letter to Amazon suggesting that selling certain books makes them complicit in deaths. A regulator publicly wonders whether a social media platform is doing enough to combat misinformation. No explicit threat is made, but the implication hangs in the air: nice business you've got there. Shame if something happened to it.

Governments can threaten boycotts, regulatory action, or the loss of lucrative government contracts. They can make it difficult for disfavored organizations to access banking services or payment processing. They can demand that platforms reveal the identities of anonymous critics.

The pressure doesn't need to be public. A phone call from a White House staffer. A private meeting with a state regulator. The target gets the message without anyone having to put anything in writing.

This Is Not New

If you think censorship by proxy is a product of the internet age, think again. The technique is centuries old.

In 1662, the English Parliament passed the Licensing of the Press Act. Rather than hunting down every seditious pamphleteer, the government simply required printers to obtain licenses and held them legally accountable for whatever came off their presses. Suddenly, printers had a powerful incentive to self-censor. They became the government's unpaid enforcement arm.

The same logic appears throughout history. Hold the intermediary responsible, and you don't need to track down every troublesome speaker.

Kreimer identified McCarthyism as a particularly American example. During the Red Scare of the 1950s, the government couldn't legally prosecute most suspected communists and their sympathizers—free speech protections were too strong. So officials pressured private employers, Hollywood studios, publishers, and broadcasters to blacklist anyone with suspected leftist sympathies. The Constitution protected you from government prosecution. It didn't protect you from losing your job because the government made your employer nervous.

"The McCarthy era saw the rise of efforts by state and federal governments in the United States to persuade private parties to control speakers and publishers whom the accepted free speech jurisprudence placed beyond the reach of official prosecution."

The Digital Transformation

The internet didn't invent censorship by proxy, but it supercharged it.

In the pre-digital world, there were thousands of publishers, countless bookstores, endless ways for information to reach an audience. Taking down a book required physical effort—burning copies, shutting down printing presses, prosecuting booksellers one by one.

Today, a handful of platforms control most of the world's digital speech. A few social media companies. A few cloud hosting providers. A few payment processors. A few app stores. Pressure any one of these chokepoints, and you can silence voices at scale with unprecedented efficiency.

As law professors at Tel Aviv University observed in 2003, with remarkable prescience: the state had "allowed private nodes of control to emerge and develop in the information environment" and was now using them to "seize control."

They were writing about the early internet. They couldn't have imagined how concentrated those nodes of control would become.

The COVID-19 Flashpoint

The COVID-19 pandemic and the 2020 presidential election brought censorship by proxy into mainstream political debate in ways it hadn't been before.

Consider Senator Elizabeth Warren's 2021 letters to Amazon and other booksellers. Certain books were spreading COVID-19 misinformation, she argued, leading to "untold deaths." By selling these books, she suggested, retailers were "potentially unlawful" participants in that harm. She accused Amazon specifically of "potentially leading countless Americans to risk their health and the health of their neighbors based on misleading and inaccurate information."

Warren didn't have the power to ban these books. The First Amendment wouldn't allow it. But she could make Amazon uncomfortable about selling them.

Legal scholar Christopher Keleher saw a pattern:

"Although axiomatic that the government cannot silence speakers based on their viewpoint, using intermediaries to do so is equally impermissible. Yet defiance of the medical establishment prompted officials and their corporate proxies to collude."

Whether the senator's pressure campaign was appropriate public health advocacy or unconstitutional jawboning depends largely on your politics. What's notable is that the tactic—using official stature to pressure private companies—was openly employed and openly debated.

When Elon Bought Twitter

Elon Musk's 2022 acquisition of Twitter (now X) provided another flashpoint. As content moderation policies changed under new ownership, Representative Adam Schiff publicly pressured Musk to curtail what he called a rise in abusive speech on the platform.

Reason magazine called this censorship by proxy. A sitting congressman was using his office to pressure a private company about its editorial decisions. Whether Schiff's concerns about abusive speech were legitimate was almost beside the point. The mechanism itself was the problem.

This brings us to a fascinating wrinkle in the censorship-by-proxy debate: the tactic cuts across partisan lines.

A Bipartisan Temptation

One of the most important things to understand about censorship by proxy is that it's not a left-wing or right-wing phenomenon. It's a power phenomenon. Whoever holds government authority faces the temptation to use it.

During the Trump administration, critics accused officials of pressuring social media companies to be more permissive of conservative content. During the Biden administration, critics accused officials of pressuring those same companies to remove content questioning COVID-19 policies or election integrity.

The Buckeye Institute, filing a legal brief in the Murthy versus Missouri case, made the point explicitly:

"Censorship by proxy has a history of bipartisan abuse, with governments using it to silence people and organizations espousing politically unpopular ideas."

This is important because it suggests the problem isn't about which party is in power. It's about the structural relationship between government and the platforms that now mediate most public speech.

The Courts Weigh In

American courts have begun grappling with censorship by proxy, though the legal doctrine remains unsettled.

In 2024, the Supreme Court heard National Rifle Association of America versus Vullo. The case arose from meetings between the director of New York's Department of Financial Services and banks and insurance companies that did business with the NRA. The director allegedly pressured these financial institutions to sever their relationships with the gun rights organization.

The Court found that the NRA had plausibly alleged a First Amendment violation. When a government official uses their regulatory power to pressure private companies into punishing an organization for its speech, that's state action—even if the official never issued a formal order.

The case didn't definitively resolve the legal questions, but it established that jawboning can cross constitutional lines.

That same year, the Court heard Murthy versus Missouri, in which state attorneys general accused Biden administration officials of coercing social media platforms to remove content critical of government positions on COVID-19 and elections. Americans for Prosperity filed a brief with a memorable line:

"Censorship by proxy is still censorship."

The courts are still working out exactly where persuasion ends and coercion begins. When does a government official's strongly worded suggestion become an unconstitutional threat? The line isn't always clear.

Beyond America's Borders

The United States isn't unique in facing these questions, though its strong constitutional speech protections make the legal issues particularly sharp.

China represents the extreme case. There, the government can fine or simply shut down social media providers that don't comply with censorship directives. The "proxy" aspect is minimal—platforms either do as they're told or cease to exist. There's no pretense of independence.

The European Union presents a more complex picture. Under European law, hosting services enjoy protection from liability for content they host—but only if they agree to remove content at government request. This creates exactly the kind of structural incentive that censorship-by-proxy critics worry about. Platforms face a choice: accept conditional immunity that requires deference to government content demands, or face legal exposure.

European regulators would argue this framework appropriately balances free expression against other harms. Critics counter that it creates a system where governments can effectively dictate what platforms allow without ever technically ordering censorship.

The European Commission's 2025 fine against X (the platform formerly known as Twitter) illustrates the tensions. The €140 million penalty was ostensibly about transparency requirements—making blue checkmarks available to anyone, failing to maintain a proper advertising repository, not giving researchers sufficient data access. "Today's decision has nothing to do with content moderation," the Commission insisted.

But critics saw a pattern: European authorities repeatedly finding technical violations to pressure a platform they considered insufficiently cooperative on content issues. The fines might not be about speech directly, but they create leverage over a platform's speech decisions.

What Would the Opposite Look Like?

It's useful to consider what censorship by proxy is not.

It's not direct government censorship—the classic model where the state explicitly bans speech and punishes violations. In that scenario, constitutional protections clearly apply. Courts can strike down the law. Citizens know exactly what the government is doing.

It's also different from purely private content moderation. When Twitter or Facebook removes content based on their own policies, without government pressure, that's a private editorial decision. You might dislike it, but it doesn't implicate the First Amendment.

Censorship by proxy lives in the murky space between these clear cases. The government applies pressure but doesn't issue orders. The platform makes the decision but under duress. Responsibility becomes diffuse, accountability elusive.

This is why the phenomenon troubles civil libertarians across the political spectrum. It allows governments to achieve censorship outcomes while maintaining plausible deniability and evading constitutional constraints.

The Accountability Problem

In October 2025, Will Creeley of the Foundation for Individual Rights and Expression testified before the United States Senate Committee on Commerce, Science, and Transportation. His message was simple: "Federal officials must be meaningfully deterred from jawboning, and held accountable when they do."

But how?

The challenge is that censorship by proxy often happens in private. A phone call. A meeting. An implicit threat that's never quite explicit. Even when pressure campaigns are public—like Warren's letters to Amazon—they maintain enough ambiguity to avoid clear legal liability.

Congress has begun to respond. The House of Representatives passed the Protecting Speech from Government Interference Act, attempting to establish clearer boundaries around government pressure on platforms. Whether such legislation can effectively deter jawboning remains to be seen.

The New Civil Liberties Alliance took a different approach, suing the State Department in 2023 for allegedly developing tools to damage advertising revenues for media outlets—including the New York Post, Reason, RealClearPolitics, The Daily Wire, and The Federalist—deemed to spread views unfavorable to the administration. The theory: if you can't silence outlets directly, you can make it unprofitable for anyone to advertise with them.

The Stakes

Why does any of this matter?

Constitutional democracies developed free speech protections for good reasons. Open debate helps societies identify truth. Dissent checks government overreach. Minority viewpoints, today unpopular, might prove correct tomorrow.

These protections become meaningless if governments can simply route around them by pressuring private intermediaries. The formal right to speak means little if no platform will host you, no payment processor will take your money, and no bank will maintain your account—all because government officials made clear they'd prefer you silenced.

The concentration of digital infrastructure makes this especially concerning. When a handful of platforms control most online speech, pressure on those platforms affects everyone. There's no easy way to route around the chokepoints.

First Amendment scholar Evelyn Douek, commenting on the Kimmel cancellation, noted that government officials "need not act directly to censor speech." The Cato Institute agreed. The mechanism of indirect pressure can be as effective as direct prohibition—and harder to challenge in court.

At the same time, reasonable people disagree about where legitimate government advocacy ends and unconstitutional coercion begins. Should officials never express opinions about what platforms should do? That seems unrealistic. Governments have legitimate interests in combating genuine dangers—foreign propaganda, incitement to violence, public health misinformation.

The question is one of means, not ends. Officials can argue for their preferred policies. They can try to persuade. What they cannot do—or should not be able to do—is use the implicit threat of government power to compel private actors to suppress speech the government couldn't directly ban.

Drawing that line precisely is the work of courts, legislatures, and ongoing democratic debate. The fact that we're having that debate openly may be, in itself, a reason for cautious optimism. Censorship by proxy is harder to accomplish when everyone knows what it is and how it works.

But the temptation will always be there. Power seeks to protect itself from criticism. And finding someone else to do your censoring remains the oldest trick in the book.