abstract. The U.S. classification system leaves billions of documents hidden from view. Any departure from secrecy therefore grabs headlines worldwide. Presidents and others leverage the system by leaking or planting information with the press to bring attention to selected topics. A lesser-studied, and more insidious, way Presidents take advantage of widespread secrecy is by selective declassification—declassifying documents that fit their chosen narratives, while keeping conflicting documents classified. The first sustained analysis in the scholarly literature, this Note explores the uses and dangers of selective declassification through case studies from the three most recent presidencies. President Bush selectively declassified documents regarding Iraq’s possession of weapons of mass destruction; President Obama did so with documents from the Osama Bin Laden raid; and President Trump did the same with documents on the investigation of Russian interference in the 2016 election. As these case studies show, selective declassification misleads the public and harms the free flow of information by skewing our discussions and choices. To understand why selective declassification is concerning, this Note uses First Amendment theory to show how its harms are rooted in the manipulation of the marketplace of ideas. To see why those harms last, this Note turns to cognitive political science and examines our predisposition to believe information aligned with our partisan beliefs. The Note concludes by offering solutions for the Executive, Congress, the judiciary, and the press.

author.Yale Law School, J.D. 2020; Dartmouth College, B.A. 2015. Special thanks to Oona Hathaway and the students in her spring 2019 seminar, “Top Secret: The U.S. National Security State,” for invaluable feedback and dedication to student scholarship. Thanks also to Steven Aftergood, David Pozen, Dave Schulz, my friends and family, and the Yale Law Journal editors, particularly Caroline Wallace, Joshua Feinzig, and Alexander Nabavi-Noori. All opinions (and errors) in this Note are my own.

Introduction

The Bush Administration needed to boost support for the Iraq War. This effort involved making hundreds of public statements but, to gain maximum traction, officials needed to cite evidence for their claims. Much of it came from a classified National Intelligence Estimate (NIE), a compilation of intelligence agencies’ findings, written at Congress’s request.1 The document concluded—based on faulty intelligence—that Saddam Hussein had weapons of mass destruction (WMD) and was actively pursuing a nuclear program.2 But it also included a dissent attacking these claims.3 Days before Congress was scheduled to vote on authorizing military force, the Administration released an unclassified summary of the NIE.4 It excised all dissents and exaggerated the NIE’s conclusions. Based in part on this information, Congress voted to authorize the invasion. It went poorly, and no WMD were found.5

Months later, a former diplomat blasted the Administration’s key claims on WMDin a New York Times editorial.6 President Bush soon declassified the NIE’s actual eight-page “Key Judgments” section, which helped refute the criticism.7 But it was not until 2004 that the public could see the entire NIE—still heavily redacted—at which point the war was well underway.8 Throughout, these selective declassifications amplified the Administration’s faulty intelligence and false statements.9

President Bush’s NIE declassification was just one of many times Presidents have selectively declassified information to serve their political goals, misleading the public. The extensive classification system has sparked criticism from legal scholars,10 journalists,11transparency advocates,12 politicians,13 and even officials in charge of the system.14 Restricting this much information, they contend, hinders the transparency people need to hold leaders accountable and make informed choices. Government secrecy is indeed concerning. But focusing only on the information the government conceals overlooks another critical issue—that the government also has the power to reveal information, often by telling only one side of a story. Presidents have especially wide latitude in this regard. They may give classified information to favored reporters, use it in speeches, and spread it through social media. And, as this Note explores, they may declassify documents at will while keeping contradictory information classified. Whether their goal is to garner support for a congressional vote, a war, or a political agenda, selective declassification can be a powerful tool for shaping conversations and outcomes. Politicians and the public see only the chosen documents, form a one-sided view, and act accordingly. In making selective disclosures, Presidents are not promoting transparency, although they say they are. They are spinning secrets.

Selective declassification is arguably the President’s strongest and most dangerous information-shaping tool. Secrecy is the norm, so declassifications grab public attention. But the information revealed is one-sided, released only if it advances a chosen agenda. Unlike other ways in which national-security information comes to light—such as unauthorized leaks of government documents or authorized plants of information with favored reporters—Presidents can openly discuss selective disclosures while praising their own transparency. They can use the declassified documents as evidence for their claims and benefit from a false sense of credibility while hiding inconsistent information. This “spin” on national-security information creates long-lasting harms, as politicians and the public base policy choices on partial, misleading information.15 Because our brains cling to misperceptions, the effects can last even after accurate information comes to light.16

Each of the past three Presidents selectively declassified materials. President Bush targeted his NIE disclosures at Congress before it voted to authorize the use of force in Iraq. President Obama released some documents captured in the Osama Bin Laden raid, which supported the White House’s narrative that Bin Laden remained a major player whose demise was strategically important.17 President Trump ordered the partial declassification of a Foreign Intelligence Surveillance Act (FISA) surveillance warrant of former campaign adviser Carter Page, seeking to reveal anti-Trump bias in the Justice Department and discredit Robert Mueller’s investigation of Russian interference.18 In each case, Presidents used their declassification authority to shape politics and public opinion in their favor while concealing undesirable truths. These disclosures were part of broader, largely successful public-relations campaigns: Congress authorized the use of military force in Iraq with solid public support;19 heroic accounts of the Bin Laden raid predominate;20 and most Republicans see the Russia investigation as a “political witch hunt.”21 In each case, the President’s selective declassification helped supplant a more complicated, less favorable truth.

Despite selective declassification’s distorting effects and serious policy impacts, the phenomenon has not yet received in-depth scholarly attention. It is “a strange lacuna in the secrecy literature,” as David Pozen has observed.22 Scholars and others have identified the phenomenon23—which has also been called “ad hoc”24 or “instant”25 declassification—but have not given selective declassification the attention it deserves. This Note begins to fill that gap. Part I defines “selective declassification,” with a focus on its misleading effects. Section I.A distinguishes selective declassification from the other irregular ways in which classified information comes to light: leaks and plants. These phenomena are more common and have received greater academic treatment, but are a less potent tool for Presidents because they cannot publicly stand behind the revealed documents. Section I.B situates selective declassification in the context of the current classification system.

Part II examines selective declassifications by the three most recent Presidents. Part III draws on these case studies to identify selective declassification’s uses and dangers. Selective declassification has long-lasting effects and can lead to policy choices the public may have rejected given accurate information. Part III also responds to the counterarguments that, in the face of widespread government secrecy, all disclosures are beneficial—or that misleading disclosures can at least be corrected. To the contrary, affirmatively misleading information is worse than none, creating misperceptions that last even after more complete information surfaces. To explore these effects further, Part III looks beyond the national-security literature, drawing on First Amendment theory and the political-science literature on misinformation. It finds that selective declassification has a unique ability to distort the marketplace of ideas by authoritatively spreading inaccurate narratives. When these inaccuracies align with people’s preexisting views, people are likely to believe them even after corrective information is provided.

Part IV concludes by recommending potential remedies to the distorting effects of selective declassification. While the ideal solution would be for Presidents to make only declassifications that provide an accurate picture, they are unlikely to give up a powerful political tool. Congress can help by increasing declassification funding and legislatively ordering declassification. When there are fewer secrets overall, there are fewer secrets to spin. Strengthening the Freedom of Information Act (FOIA) in the national-security context would also help, as would a statutory shield law and Espionage Act reforms. Last, the press plays a critical role in reducing the impact of selective declassification by revealing the full story.