State Secrets Privilege: Doctrine, Court Review, and Recent Cases

The state secrets privilege is a doctrine of federal law that lets the executive branch block evidence from a lawsuit, and sometimes end the lawsuit entirely, when disclosure would threaten national security. It traces to the Supreme Court’s 1953 decision in United States v. Reynolds and gives agency heads extraordinary power to shield military plans, intelligence operations, and diplomatic communications from courts. Whether a case survives after the privilege is asserted depends on which branch of the doctrine applies and how central the secret evidence is to the claims.

The Two Branches of the Doctrine

The privilege operates through two distinct rules, and the difference matters enormously for anyone caught in a case touching classified information.

The Totten bar is the harsher of the two. It applies when the subject matter of the lawsuit is itself a secret government relationship, most commonly an espionage agreement. Under this rule, the case is dismissed at the threshold, before any evidence is gathered or any balancing test is applied. The Supreme Court reaffirmed the categorical nature of this bar in Tenet v. Doe (2005), holding that it “bars suits—including those alleging due process and tort violations—if they are predicated on an alleged espionage contract with the government.” The Court explicitly rejected the argument that the Totten rule had been folded into the more flexible Reynolds framework.¹

The Reynolds privilege is an evidentiary rule, not an automatic dismissal. It allows the government to withhold specific pieces of evidence that would jeopardize national security. The lawsuit itself may continue if the plaintiff can prove the claims with other evidence. Only when the privileged information is so central that the case cannot proceed fairly does the court dismiss the whole action. This distinction matters because the Reynolds privilege triggers judicial review and balancing; the Totten bar does not.

How the Government Invokes the Privilege

The privilege cannot be invoked casually. Under the framework set out in Reynolds, a formal claim must be “lodged by the head of the department which has control over the matter” and may be made only “after actual personal consideration by that officer.”² The Secretary of Defense, the Director of National Intelligence, or a comparable agency head must personally review the materials and sign a declaration explaining why disclosure would harm national security. Lower-level attorneys cannot make this call on their own.

Since 2009, the Department of Justice has imposed additional internal requirements. The agency seeking protection must submit a detailed declaration specifying the nature of the information, the harm disclosure would cause, and why that harm is reasonably likely. A State Secrets Review Committee of senior DOJ officials evaluates the claim and consults with the originating agency and the Office of the Director of National Intelligence. The committee’s recommendation goes up through the Deputy Attorney General, and the Attorney General’s personal approval is required before the privilege moves forward.³

That policy also sets explicit boundaries. The government may not invoke the privilege to conceal violations of the law, prevent embarrassment to a person or agency, restrain competition, or delay the release of information that would not actually harm national security. If the Attorney General approves the privilege in a case involving credible allegations of government wrongdoing, the Department must refer those allegations to the appropriate agency Inspector General for investigation.³ This is an internal guideline, not a statute. A future administration could revise or abandon it without congressional action.

How Courts Evaluate the Claim

Once the government files a formal assertion, the judge must independently decide whether the privilege applies. Reynolds described the court’s job as determining “whether the circumstances are appropriate for the claim of privilege, and yet do so without forcing a disclosure of the very thing the privilege is designed to protect.”²

The Reasonable Danger Standard

The test is not whether disclosure would certainly cause harm. It is whether there is a “reasonable danger that compulsion of the evidence will expose military matters which, in the interest of national security, should not be divulged.”² A realistic possibility of exposure is enough. Reynolds also recognized a sliding scale: the stronger the plaintiff’s need for the information, the more probing the court’s review should be. In practice, when national security concerns are strong, courts routinely defer to the executive branch’s judgment about what would cause damage.

In Camera Review

Judges sometimes review the classified materials privately, without the plaintiff or their attorneys present, to assess the government’s claim. But Reynolds was clear that in camera inspection is not always required. If the surrounding circumstances are enough to establish a reasonable danger, “the court should not jeopardize the security which the privilege is meant to protect by insisting upon an examination of the evidence, even by the judge alone, in chambers.”² Some privilege claims are sustained on the government’s declaration alone, without the judge ever seeing the underlying documents.

Segregability

When the government claims privilege over a broad set of documents, courts have an obligation to consider whether non-secret portions can be separated out and disclosed. A judge may require a careful line-by-line review and release of anything not genuinely sensitive. In practice, the government often argues that the secret and non-secret portions are too intertwined to separate.

What Happens to the Lawsuit

The privilege does not automatically kill a case. Its effect depends on how central the shielded evidence is to the claims and defenses.

Evidence Exclusion

In the most limited scenario, the court removes the privileged evidence from the case and the lawsuit continues on whatever remains. If the plaintiff can still prove the claims with non-privileged information, they get their day in court, often with a weaker evidentiary hand.

Dismissal

When the excluded evidence is so central that the case cannot be tried fairly, the court must dismiss the lawsuit. This can happen in two directions. If the plaintiff cannot establish the claims without the secret evidence, there is no viable case. If the defendant cannot mount a meaningful defense without revealing classified facts, due process requires ending the proceedings.

The Fourth Circuit’s decision in El-Masri v. Tenet shows how sweeping this can be. Khaled El-Masri, a German citizen, alleged the CIA had kidnapped him, flown him to a secret detention site in Afghanistan, and interrogated him for months before releasing him without charge. The Council of Europe later concluded his account was “substantially accurate,” and President Bush publicly acknowledged the existence of the CIA’s overseas detention program. The court affirmed dismissal anyway, holding that “sensitive military secrets will be so central to the subject matter of the litigation that any attempt to proceed will threaten disclosure of the privileged matters.”⁴

The Ninth Circuit reached a similar result in Mohamed v. Jeppesen Dataplan, dismissing claims against a private contractor that allegedly provided flight services for the CIA’s extraordinary rendition program. The court found “no feasible way to litigate Jeppesen’s alleged liability without creating an unjustifiable risk of divulging state secrets.”⁵

A plaintiff whose case is dismissed on state secrets grounds typically has no meaningful path to a legal remedy. No established alternative forum can hear claims the courts cannot. Congress could theoretically pass a private bill providing compensation, and some dismissed cases have ended in quiet government settlements, but neither outcome is guaranteed or common. The DOJ’s 2009 policy requires referral of credible wrongdoing allegations to an Inspector General, but an IG investigation is not a substitute for a court judgment.³

Criminal Cases Are Governed by a Separate Statute

When classified information intersects with a criminal prosecution, the Classified Information Procedures Act (CIPA) provides a separate procedural framework. CIPA does not create or limit any substantive privilege. It establishes ground rules for handling classified evidence so that trials can proceed without unnecessary disclosure.

Under CIPA, the court holds pretrial conferences to address classified information issues before they derail the trial. The government can request protective orders covering classified material shared with a defendant during discovery. Judges may authorize the government to substitute summaries or stipulations of fact in place of the actual classified documents, allowing the jury to consider the relevant information without seeing the underlying secrets.⁶ If a defendant plans to disclose classified information at trial, they must notify the government and the court in advance. CIPA applies only in criminal cases. Civil suits involving classified information rely on the Reynolds framework.

Recent Supreme Court Developments

Two 2022 decisions reshaped the privilege’s boundaries in ways that strengthened the government’s hand.

United States v. Zubaydah

Abu Zubaydah, detained by the CIA after the September 11 attacks, sought discovery from two former CIA contractors to support proceedings before a foreign court. The information he requested concerned the location of a CIA “black site” in Poland, a fact widely reported in the press and acknowledged by European investigations. The Supreme Court sustained the government’s privilege assertion, holding that even when information has entered the public domain through unofficial sources, the government can still block official confirmation or denial. The Court reasoned that “public knowledge might be untrue” and that official confirmation “could still threaten U.S. national security and complicate U.S. relationships with foreign intelligence partners.”⁷ After Zubaydah, the government can block evidence even when anyone with an internet connection can find the same facts in news reports.

FBI v. Fazaga

Three Muslim residents of Southern California sued the FBI, alleging an informant had infiltrated their mosques and surveilled them based on their religion. They argued that Section 1806(f) of the Foreign Intelligence Surveillance Act, which provides a procedure for courts to review surveillance evidence in camera, displaced the state secrets privilege. In a unanimous decision, the Court held that “Congress did not eliminate, curtail, or modify the state secrets privilege when it enacted §1806(f).”⁸ FISA’s review procedures exist alongside the privilege rather than replacing it.

Why the Doctrine Remains Controversial

Reynolds itself carries an uncomfortable footnote. The case involved a military B-29 testing secret electronic equipment that crashed, killing three civilian observers. Their widows sued and requested the Air Force’s accident investigation report. The Secretary of the Air Force filed a formal claim of privilege, stating the aircraft and personnel were “engaged in a highly secret mission,” and the Court sided with the government.² Decades later, the accident report was declassified. It contained no military secrets. It contained evidence of Air Force negligence in maintaining the aircraft.

That revelation drives much of the ongoing criticism. If the privilege was used in 1953 to conceal negligence rather than genuine military secrets, critics argue, there is no structural reason to believe current assertions are more trustworthy. The judge evaluates the claim based on the government’s own declaration, sometimes without ever seeing the underlying evidence, and the plaintiff has no access to the materials needed to challenge the claim. Cases like El-Masri and Jeppesen amplify the concern: plaintiffs alleged serious government misconduct, had substantial corroboration from foreign investigations and public statements by U.S. officials, and still saw their cases dismissed because the underlying facts were too entangled with classified operations. Several legislative proposals have been introduced over the years to codify judicial oversight requirements, mandate in camera review of actual documents, and create procedures for cleared counsel to challenge the government’s assertions. None have been enacted. The privilege remains a judge-made rule, governed primarily by a 1953 decision whose factual underpinnings turned out to be false.

1
Justia U.S. Supreme Court Center. Tenet v. Doe, 544 U.S. 1 (2005)
2
Justia U.S. Supreme Court Center. United States v. Reynolds, 345 U.S. 1 (1953)
3
U.S. Department of Justice. Policies and Procedures Governing Invocation of the State Secrets Privilege
4
United States Court of Appeals for the Fourth Circuit. El-Masri v. Tenet, No. 06-1667 (4th Cir. 2007)
5
United States Court of Appeals for the Ninth Circuit. Mohamed v. Jeppesen Dataplan Inc., 614 F.3d 1070 (9th Cir. 2010)
6
Office of the Law Revision Counsel. Classified Information Procedures Act
7
Supreme Court of the United States. United States v. Zubaydah, 595 U.S. 1 (2022)
8
Supreme Court of the United States. FBI v. Fazaga, 595 U.S. 344 (2022)