There is a courtroom in Washington, D.C., where the most consequential decisions about American privacy are made — and where, by design, only one side is ever heard. Its proceedings are classified. Its opinions are secret. The people whose lives it touches almost never learn that it has ruled on them. It hears arguments only from the government. There is no defendant in the room, no opposing counsel, no adversary to test the government's claims. And in the decades of its existence, it has approved the government's requests to spy on Americans and others at a rate that beggars belief: tens of thousands of applications granted, a literal handful denied. This is the United States Foreign Intelligence Surveillance Court — the FISC, or FISA Court — and it represents one of the most profound departures from the Founders' vision of justice in the entire American legal system.
The framers of the Constitution fought a revolution, in significant part, over the government's power to search and seize without meaningful check. They wrote the Fourth Amendment specifically to abolish the instruments of arbitrary search that the British Crown had used against them. And yet, two centuries later, the United States has constructed a secret court whose entire function is to authorize searches and surveillance in proceedings that violate nearly every principle the Fourth Amendment was written to protect: openness, adversarial testing, particularity, and the interposition of a genuinely neutral magistrate between the government and the governed. This is the story of how that court came to be, how it operates, how spectacularly it has failed to protect the rights it was nominally created to safeguard, and why its very existence should trouble anyone who takes the Founders' warnings seriously.
The Writs of Assistance and the Birth of the Fourth Amendment
To understand why a secret surveillance court is such a betrayal of the founding vision, you have to go back to Boston in 1761 — fifteen years before the Declaration of Independence — and to a legal instrument called the writ of assistance. A writ of assistance was a general warrant. It authorized royal customs officials to search anywhere, at any time, for smuggled goods, without having to specify what they were looking for or where they expected to find it. It was, in effect, a blank check for search. Once issued, it remained valid for the life of the reigning monarch, and it could be used against any home, any warehouse, any person, on nothing more than the official's discretion.
In February 1761, a Boston lawyer named James Otis Jr. argued against the writs of assistance before the Superior Court of Massachusetts. He lost the case. But his argument — that the writs were "the worst instrument of arbitrary power, the most destructive of English liberty and the fundamental principles of law, that ever was found in an English law book" — electrified the colonies. A young John Adams sat in the courtroom that day and watched Otis speak. Decades later, Adams wrote that "then and there the child Independence was born." The general warrant, the power to search without particularized cause or limit, was among the original grievances that drove the American Revolution.
When the founding generation came to write the Fourth Amendment, they wrote it as a direct repudiation of the writs of assistance. The text is precise, and every clause is a lesson learned from British abuse: "The right of the people to be secure in their persons, houses, papers, and effects, against unreasonable searches and seizures, shall not be violated, and no Warrants shall issue, but upon probable cause, supported by Oath or affirmation, and particularly describing the place to be searched, and the persons or things to be seized."
"A man's house is his castle; and whilst he is quiet, he is as well guarded as a prince in his castle." — James Otis, arguing against the writs of assistance, 1761
Notice what the amendment demands. Probable cause — not mere suspicion. An oath — a sworn, accountable representation. And, above all, particularity: the warrant must describe specifically the place to be searched and the things to be seized. The whole point was to abolish the general warrant, to forbid the government from conducting open-ended searches based on official discretion. The framers wanted every search to be justified, specific, and accountable. They wanted a neutral magistrate to stand between the government's desire to search and the citizen's right to be left alone, and they wanted that magistrate to say no when the government could not meet the standard.
Implicit in this design is something the framers took for granted: that the process would be, in the ordinary course, testable. The reasonableness of a search could be challenged in open court. Evidence obtained in violation of the Fourth Amendment could be suppressed. The citizen would eventually have the opportunity to contest the government's action before a genuinely neutral tribunal, in an adversarial proceeding. The Fourth Amendment was not merely a set of standards; it presupposed a system of accountability in which those standards could be enforced. The FISA Court dismantles that presupposition at its foundation.
A Court Born of Scandal
The FISA Court was not created out of malice. It was, in a sense, a reform — a response to genuine abuses. In the 1970s, the Church Committee, a Senate investigation led by Senator Frank Church of Idaho, exposed a vast apparatus of illegal government surveillance. For decades, the FBI, CIA, and NSA had spied on Americans without warrants, without oversight, and often for nakedly political purposes. The FBI's COINTELPRO program had surveilled and harassed civil rights leaders, antiwar activists, and political dissidents. The government had wiretapped Martin Luther King Jr. and attempted to blackmail him. The intelligence agencies had operated, in the domain of domestic surveillance, entirely outside the law.
Congress responded in 1978 with the Foreign Intelligence Surveillance Act, which created the FISA Court to provide judicial oversight of foreign-intelligence surveillance. The idea was reasonable on its face: rather than letting the executive branch spy on its own say-so, the government would have to go before a federal judge and obtain a warrant, just as in an ordinary criminal case. The FISC would consist of federal district judges, appointed to the court by the Chief Justice of the United States, who would review the government's applications and ensure they met the statutory standard. Oversight where there had been none — a genuine improvement.
But the design contained a fatal flaw, one that has metastasized over the decades. Because the surveillance at issue involved classified national-security matters, the proceedings would be secret. And because the targets of surveillance obviously could not be notified in advance — that would defeat the purpose — the proceedings would be ex parte: the government would appear alone, with no opposing party, no defense counsel, no adversary of any kind. The court would hear only the government's version of the facts and the government's argument for why surveillance was justified. A court designed to check the government was structured so that the government would never face a challenger before it.
The Numbers That Define a Rubber Stamp
The consequences of that structure are written plainly in the court's own statistics, and they are staggering. According to a fact sheet published by the Center for Strategic and International Studies in early 2026, over the course of its existence the FISA Court had rejected just 11 applications while approving 33,942 — modifying another 504. Eleven denials out of nearly thirty-four thousand applications. That is an approval rate exceeding 99.9 percent.
The pattern holds year after year. As Lawfare reported, in 2013 the U.S. government made 1,588 requests to the FISC for authority to conduct electronic surveillance. The number denied: zero. Thirty-four were modified. In the 2023–2024 reporting period, according to figures compiled from the court's own disclosures, the FISC received 637 applications for electronic surveillance or physical search orders and denied 29 in full or in part while substantially modifying another 135 — a somewhat higher rejection rate than in earlier years, but still overwhelmingly a court that says yes. As one scholarly analysis in the Washington University law repository summarized the critique, the court "has approved more than 99% of the government's 33,000 ex parte surveillance requests, a fact cited as extraordinary evidence that the court is nothing more than a rubber stamp."
Defenders of the court object to the "rubber stamp" characterization, and their arguments deserve a fair hearing. They point out, correctly, that the raw approval rate is misleading in one respect: the government engages in an iterative, back-and-forth process with the court's legal staff before formally submitting an application, and applications that would not survive scrutiny are often revised or withdrawn before they are ever formally denied. The high approval rate, on this view, reflects not rubber-stamping but a filtering process that weeds out weak applications before the formal decision. The FISC, defenders argue, exercises "extensive and serious judicial oversight."
There is something to this. But it does not answer the fundamental objection, and in some ways it deepens it. An iterative process in which the government and the court's staff work together to craft an application that the court will approve is not adversarial oversight — it is collaboration. The whole premise of the Fourth Amendment's warrant requirement, and of the adversarial system generally, is that the government's claims must be tested by someone with an interest in defeating them. A process in which the only participants are the government and a court helping the government succeed is missing the essential ingredient. There is no one in the room whose job is to say, "That is not true," or "You have not met your burden," or "The person you want to surveil is innocent." And as the Carter Page affair would demonstrate with devastating clarity, the absence of that adversary is not a technicality. It is the difference between oversight and theater.
Carter Page and the Collapse of the Myth
For years, defenders of the FISA Court could argue that its secrecy, however uncomfortable, was benign — that the judges were diligent, the government was honest, and the lack of an adversary did not matter because the process worked. Then came the case of Carter Page, and the myth collapsed under the weight of the government's own inspector general.
Carter Page was a foreign-policy adviser to the 2016 Trump presidential campaign. In the course of the FBI's "Crossfire Hurricane" investigation into possible links between the campaign and Russia, the FBI sought and obtained FISA warrants to surveil Page — one initial application and three renewals, spanning 2016 and 2017. In December 2019, the Justice Department's Inspector General, Michael Horowitz, published an exhaustive review of how those warrants had been obtained. What he found should have ended any illusion that the FISA Court's secret, one-sided process could be trusted to protect the rights of surveillance targets.
Horowitz identified, in his own words, "significant inaccuracies and omissions" in the applications — 7 in the first application and a total of 17 by the time of the final renewal in 2017. Seventeen significant errors or omissions in the government's sworn representations to a federal court. As the Heritage Foundation summarized the findings, these were errors "any one of which is inexcusable." The applications relied heavily on the so-called Steele dossier, whose reliability the FBI had reason to doubt and whose problems the FBI failed to disclose to the court. Exculpatory information — information that would have undercut the case for surveilling Page — was omitted. In one especially egregious instance, an FBI lawyer altered an email to make it say the opposite of what it originally said, in order to support the application. The lawyer, Kevin Clinesmith, later pleaded guilty to a criminal charge arising from the falsification.
Seventeen significant errors and omissions in four sworn applications to surveil an American — including a document altered to say the opposite of the truth. And the secret court, hearing only the government's side, approved every one.
Here is the point that cuts to the heart of the matter. The FISA Court approved all four applications. It did not catch the seventeen errors. It did not catch the doctored email. It did not catch the omitted exculpatory evidence or the concealed doubts about the dossier. And it could not have, because the process was designed to prevent it from catching such things. There was no adversary in the room to challenge the government's representations, no defense counsel to demand the underlying evidence, no one with any interest in or ability to test whether the sworn statements were true. The court took the government's word, because the government's word was all it was ever given. The one-sided structure that defenders call "serious judicial oversight" produced, in the highest-profile surveillance case of the era, a series of warrants built on a foundation of error and, in at least one instance, deliberate falsification.
Horowitz's subsequent work made clear the Page case was not an aberration. In follow-up reviews of a broader sample of FISA applications, his office found errors and failures to follow the FBI's own verification procedures — the "Woods Procedures" — in essentially every application it examined. The problem was not one rogue lawyer or one botched case. The problem was systemic: a secret, non-adversarial process that placed enormous trust in the government's honesty and had no mechanism to verify it. When the inspector general finally looked, he found that the trust had been misplaced.
Section 702 and the Dragnet
The Carter Page case involved traditional FISA — individualized applications to surveil a specific person. But the modern surveillance state has moved far beyond individualized warrants into the realm of programmatic, bulk collection, and here the FISA Court’s role becomes even more troubling. The key authority is Section 702 of the FISA Amendments Act, first enacted in 2008 and repeatedly reauthorized since. Section 702 permits the government to collect the communications of non-U.S. persons located abroad without individualized warrants — but in doing so, it inevitably sweeps up the communications of Americans who are in contact with those foreign targets.
Under Section 702, the FISA Court does not review individual surveillance targets at all. Instead, it approves broad annual “certifications” and the general procedures the government will use to target foreigners and to handle the American communications it collects incidentally. The court signs off on the framework; the executive branch then applies it to specific targets on its own authority, without returning to the court for each one. This is a profound shift away from the particularity the Fourth Amendment demands. The framers required that warrants “particularly describe the place to be searched, and the persons or things to be seized.” Section 702 authorizes surveillance by category and procedure, not by particularized warrant — the very definition of the general search the Fourth Amendment was written to forbid.
The volume is enormous. The government has acknowledged that Section 702 collection sweeps up hundreds of millions of communications annually. And once the communications of Americans are in the government’s databases — collected “incidentally” to the targeting of foreigners — the FBI has historically been able to search those databases for the communications of specific Americans without a warrant. These are the so-called “backdoor searches”: the government collects Americans’ communications without a warrant under the theory that it was really targeting foreigners, and then queries that trove for the American’s communications, again without a warrant. Civil-liberties advocates have argued for years that this practice drives a truck through the Fourth Amendment — using the foreign-intelligence justification to accomplish, against Americans, exactly the kind of warrantless search the amendment prohibits.
Government compliance reviews and FISA Court opinions that have been declassified reveal that the FBI conducted hundreds of thousands of improper queries of Section 702 data — searches that violated even the government’s own rules. In some declassified opinions, the FISA Court itself documented tens of thousands of queries that failed to meet the applicable standards, including searches run on people connected to political campaigns, protests, and other constitutionally protected activity. The abuses were not the work of a foreign adversary breaking into the system. They were the routine operation of the system itself, by the agency entrusted to run it.
What Snowden Revealed
For most of its existence, the FISA Court operated in such deep secrecy that the public had almost no ability to evaluate what it was authorizing. That changed in June 2013, when Edward Snowden, a contractor for the National Security Agency, disclosed a trove of classified documents revealing the true scope of American surveillance. Among the most consequential disclosures was a secret FISA Court order directing the telecommunications company Verizon to hand over, on an ongoing daily basis, the call records of millions of American customers — the metadata of essentially every phone call, foreign and domestic.
The revelation stunned the public and even many members of Congress, who had voted to reauthorize the underlying statutes without understanding how the FISA Court had interpreted them. The court had secretly construed Section 215 of the Patriot Act — a provision allowing the government to obtain business records “relevant” to a terrorism investigation — to authorize the bulk collection of the phone records of the entire nation. “Relevant to an investigation” had been stretched, in secret, to mean “all records of all Americans.” This was lawmaking in the dark: a secret court, interpreting a public statute in a classified opinion, arriving at a reading that the public would never have accepted and that Congress itself had not clearly intended. A federal appeals court would later rule that the bulk metadata program had in fact exceeded what the statute authorized — but only after years of operation, and only after Snowden’s disclosures made public challenge possible.
A secret court, reading a public law in a classified opinion, decided that “records relevant to an investigation” meant the phone records of every American. No one outside the government knew until a whistleblower told them.
The Snowden episode crystallized the deepest danger of the FISA regime: a system of secret law. In the Anglo-American tradition, law is public. Citizens are entitled to know the rules that bind them and the powers the government claims over them. A body of secret judicial opinions, interpreting statutes in ways the public cannot see and cannot contest, is fundamentally incompatible with the rule of law as the founding generation understood it. The framers fought a revolution against arbitrary power exercised without accountability. A court that makes law in secret, authorizing surveillance the public will never learn of, is a form of exactly that arbitrary power — dressed in judicial robes, but operating beyond the reach of the public check the Constitution presupposes.
Everything the Founders Feared, Rebuilt
Step back and consider what the FISA Court actually is, measured against the principles of the Fourth Amendment. The framers demanded openness; the FISC operates in total secrecy. The framers demanded particularity; the FISC, especially under programmatic authorities like Section 702, has authorized surveillance of a breadth the framers could not have imagined, sweeping up the communications of vast numbers of people, including Americans, without individualized warrants. The framers demanded a neutral magistrate standing between the government and the citizen; the FISC hears only from the government, in proceedings the citizen will almost never learn about. The framers built the Fourth Amendment to abolish the general warrant; the FISA regime has, in the eyes of many critics, effectively resurrected it in digital form.
The comparison to the writs of assistance is not rhetorical excess. The writs were objectionable to the founding generation for specific reasons: they were general rather than particular, they rested on official discretion rather than tested probable cause, and they left the citizen with no meaningful way to challenge the search. The modern surveillance state, authorized by a secret court that approves virtually everything the government asks, reproduces each of these features. It is not identical — the technology and the legal forms are different — but the essential character is the same: a government power to search and surveil that is not meaningfully checked by an independent, adversarial process, and that the target cannot contest because the target does not know.
Even the ordinary criminal warrant system has drifted toward the same pathology. As one analysis noted, in 2016 only two out of 3,168 federal Title III wiretap applications were rejected — an approval rate of 99.9994 percent. The FISA Court's rubber-stamp tendency is the acute form of a chronic disease: the transformation of the warrant requirement from a genuine check into a formality, a box to be checked rather than a barrier to be cleared. The magistrate who was supposed to say no has, across the system, very nearly stopped saying no at all.
The Ghost of COINTELPRO
There is a bitter irony at the center of this story. The FISA Court was created in 1978 as a remedy for the surveillance abuses the Church Committee had exposed — the warrantless spying on civil rights leaders, antiwar activists, and political dissidents that the FBI had carried out under COINTELPRO and related programs. The court was supposed to ensure that such abuses could never happen again, that surveillance would henceforth require the approval of a neutral judge. Yet the declassified record of the FISA era reveals surveillance touching precisely the kinds of constitutionally protected activity that COINTELPRO targeted: queries of surveillance databases run against people connected to political campaigns, against participants in protests, against individuals whose only apparent offense was their political activity.
The lesson is sobering. The abuses the FISA Court was built to prevent did not stop; they were routed through the court, given a judicial imprimatur, and continued under color of law. This is worse, in a sense, than lawless spying. Lawless spying, once exposed, is plainly illegal and can be condemned as such. Surveillance blessed by a secret court carries the appearance of legality — it has been “approved,” after all — even when it reaches the same targets and serves the same ends. The court that was meant to be a barrier became, too often, a laundering mechanism, converting what would otherwise be recognized as abuse into the ordinary, approved operation of the national-security state.
The founding generation understood that surveillance is not politically neutral. The writs of assistance were not used against everyone equally; they were tools wielded by those in power against those who challenged them. The Church Committee found the same pattern two centuries later: the surveillance apparatus turned, again and again, against dissidents, critics, and political opponents. There is no reason to believe human nature has changed. A surveillance power that operates in secret, approved by a court that hears only the government and denies almost nothing, will inevitably tend toward the same abuse — not necessarily through malice, but through the simple absence of the checks that keep power honest. The framers did not trust in the good intentions of officials. They built structural checks precisely because they knew good intentions were not enough. The FISA regime removed the checks and left only the intentions.
The Accountability That Never Comes
Perhaps the most corrosive feature of the entire system is the near-total absence of consequences when abuses are discovered. The Carter Page applications contained seventeen significant errors and a deliberately falsified email. What happened? One FBI lawyer pleaded guilty to a single count and received probation — no prison time. The broader institutional failures that produced seventeen errors across four sworn applications resulted in policy reviews, revised procedures, and promises to do better. No warrants that had been wrongly obtained were meaningfully unwound; the surveillance had already occurred, the information already gathered. The target, Carter Page, was surveilled on the strength of a flawed and partly falsified record, and no court order could give him back the privacy that had been taken.
This is the pattern throughout the FISA era. Abuses come to light — usually years later, usually through an inspector general or a whistleblower rather than through the court itself — and the response is procedural rather than consequential. New rules are written. Training is enhanced. Compliance is promised. And the machine rolls on, its fundamental structure unchanged, until the next round of abuses is discovered and the cycle repeats. Because the targets of surveillance almost never learn they were surveilled, they almost never have standing to sue; the Supreme Court has repeatedly turned away challenges to surveillance programs on the ground that the plaintiffs could not prove they had been surveilled — a fact the government keeps secret. The secrecy that defines the system also insulates it from the accountability that might reform it. You cannot challenge what you cannot see, and you are not permitted to see it.
The Reforms That Have Not Been Enough
The problems with the FISA Court have not gone entirely unaddressed. After the Edward Snowden disclosures of 2013 revealed the scope of the surveillance the court had authorized, Congress and reformers proposed changes. The USA FREEDOM Act of 2015 made some modifications, including creating a panel of outside "amici curiae" — friends of the court — who could be appointed to provide an independent perspective in certain significant cases. This was an attempt to inject some adversarial element into the one-sided process. Reform bills like the "Ensuring Adversarial Process in the FISA Court Act" have sought to go further.
But these reforms have been partial and, in the view of civil-liberties advocates, inadequate. The amicus provision applies only in a narrow category of cases and gives the outside advocate limited access and no real client. The court remains fundamentally secret and fundamentally non-adversarial. The Electronic Privacy Information Center, which has long tracked the court, acknowledges that the FISC "has at times rejected government activities because they were overbroad or because of persistent violations" — but the operative phrase is "at times." The rejections are the rare exceptions that prove the rule of near-total approval. The Carter Page abuses occurred after the post-Snowden reforms were already in place. The structural problem — a court that hears only the government, in secret, and approves almost everything — remains largely intact.
The deeper difficulty is that the entire model may be irreconcilable with the Fourth Amendment's design. The framers did not envision a category of searches so secret that they could never be tested in open, adversarial proceedings. They built a system premised on eventual accountability — on the citizen's ability to challenge the government's action before a neutral tribunal. A permanent secret court, authorizing surveillance that its targets will never learn of and can never contest, is not a modification of that system. It is a different system, one the founding generation would have recognized as precisely the kind of arbitrary power they revolted against.
Reconstructing the Castle
James Otis told the Boston court in 1761 that a man's house is his castle, guarded even against the king. The Fourth Amendment was the founding generation's attempt to build that castle into law — to guarantee that the government could not search and seize except on particularized cause, tested by oath, authorized by a neutral magistrate, and ultimately answerable in open court. The FISA Court, whatever the good intentions of its creators, has become a mechanism for tearing down that castle wall by wall, in secret, with no one present to object.
Reconstructing the constitutional order in this domain does not require abandoning the legitimate needs of national security. It requires insisting that those needs be pursued within the framework the Founders established, not outside it. It means real adversarial process — a permanent, adequately empowered advocate for privacy and civil liberties in every significant proceeding, not an occasional friend of the court with no client and limited access. It means genuine particularity, an end to the programmatic dragnet authorities that sweep up the innocent along with the suspected. It means transparency wherever transparency is possible — the publication of the court's legal reasoning, so that the public can know what the law actually is. It means real consequences for the kind of misconduct the Horowitz report exposed, so that the government's sworn representations to the court carry the weight that oaths are supposed to carry. And it means a willingness, on the part of the judges themselves, to say no — to be the check the framers intended, rather than the collaborator the statistics reveal.
None of this is a counsel of despair. The Fourth Amendment has not been repealed; it remains the supreme law of the land, waiting to be enforced. The founding generation gave us not only the principles but the tools — the insistence on particularity, on adversarial testing, on public law, on a judiciary willing to tell the government no. What has been lost can be recovered, but only if Americans first recognize how far the surveillance regime has drifted from the constitutional design, and refuse to accept the comforting fiction that a court which approves 99.9 percent of what the government asks is meaningfully protecting anyone. A rubber stamp is not a check. A secret opinion is not law in any sense the framers would recognize. And a proceeding in which only the government is heard is not a trial of the government’s claims — it is a ratification of them.
The seventeen errors in the Carter Page applications were not caught by the secret court. They were caught by an inspector general, looking backward, years later, in a report that only saw the light of day because the case had become a political firestorm. For every case that draws that kind of scrutiny, there are tens of thousands that never will — tens of thousands of applications approved in the dark, on the government's word alone, with no one to test the claims and no target who will ever know. That is not the system the Founders built. It is the system they fought a revolution to destroy. And it will remain in place until Americans remember what James Otis argued, what John Adams witnessed, and why they wrote the Fourth Amendment in the first place: because a government that can search in secret, unchecked and unaccountable, is not the servant of a free people. It is their master.