The Counsel's Exception: Presidential Records, Government Secrecy, and the Limits of Transparency Law

Credit: Stock Images

On April 1, 2026, the Justice Department’s Office of Legal Counsel released a slip opinion declaring the Presidential Records Act of 1978 unconstitutional—not unconstitutional in some narrow application, not unconstitutional as applied to a particular dispute, but facially and entirely void. The PRA, it concluded, “exceeds Congress’s enumerated and implied powers and aggrandizes the Legislative Branch at the expense of the constitutional independence and autonomy of the Executive.”

The statute being declared void is the same law that produced the presidential libraries, made the Nixon tapes public, and gives historians and journalists and citizens access to the internal workings of every administration since Reagan. It is the legal architecture by which the public eventually learns what its government did. Yet, the OLC opinion, with casual temerity, says the executive branch no longer considers itself bound by it.

The memo’s argument rests on a historical narrative: for the first 200 years of American constitutional practice, presidents owned their papers as personal property. Congress obtained presidential records, when it obtained them at all, through political negotiation and what the OLC calls, “interbranch accommodation.” The memo traces this practice from George Washington’s refusal to fully comply with the 1792 House inquiry into the St. Clair expedition through Jefferson, Jackson, and Tyler, presenting two centuries of presidential discretion as a “liquidated” constitutional norm.

The PRA, in this telling, was a radical departure. Where the earlier Presidential Recordings and Materials Preservation Act of 1974 (PRMPA) targeted a specific crisis (Nixon’s resignation and the imminent destruction of the Watergate tapes), the PRA imposed a permanent, prospective regime on every future president. The OLC argues this fails the four-factor test the Supreme Court articulated in Trump v. Mazars: it lacks a valid legislative purpose, it isn’t tailored to any specific congressional need, its information could be obtained from other sources, and it imposes burdens on the executive that chill candid advice.

The memo also distinguishes Nixon v. Administrator of General Services, the 1977 case that upheld the PRMPA against a separation-of-powers challenge. According to the OLC, that case involved extraordinary circumstances and a narrow statute, and its broader language concerningCongress’s interest in preserving presidential records ought to be read as situated reasoning, not as a general grant of legislative authority. In one of the memo’s more striking passages, the OLC simply declares that Nixon v. Administrator was “wrong” on the separation-of-powers question–an unusual move for an executive branch legal opinion to make for a settled Supreme Court precedent.

Still, every historical argument is also a choice about what to leave out. The memo’s 200-year narrative treats the pre-PRA era as a stable equilibrium of executive discretion and interbranch accommodation. What the narrative omits is that this equilibrium produced Watergate: the destroyed tapes, the shredded documents, the eighteen-and-a-half-minute gap. The PRA was Congress’s response to a demonstrated and catastrophic failure of the very accommodation model the OLC now seeks to restore as a constitutional baseline.

The omission matters because the memo’s logic depends on treating presidential good faith as the constitutional default. It invokes the “presumption of regularity” to argue that Congress cannot mandate records preservation on the speculative theory that a future president might destroy evidence. Watergate is the answer to that argument, and the memo essentially asks the reader to forget it ever happened.

There is also a quieter problem the memo acknowledges only in passing: the PRA was always weakly enforced. The D.C. Circuit’s 1991 decision in Armstrong v. Bush held that most of the statute commits records decisions to presidential discretion, with no private right of action and limited tools for the Archivist to compel compliance. Even under existing law, a president inclined toward noncompliance faced few real-time consequences. The OLC memo does not so much abolish a robust enforcement regime as it does formalize what was already a weak one into something effectively unenforceable. It declares the emperor has no clothes only after spending decades quietly removing them.

The memo did not arrive in a vacuum. Throughout 2025, the administration systematically dismantled the operational infrastructure that makes federal transparency law work. At the Centers for Disease Control and Prevention, FOIA offices were eliminated entirely. At the FDA and NIH, which together processed roughly 14,000 records requests in 2024, multiple FOIA teams were fired. When CNN filed a FOIA request with the Office of Personnel Management about security clearances for DOGE personnel, an OPM email replied: “Good luck with that, they just fired the whole privacy team.”

In March 2025, the administration fired Bobak Talebian, the longtime director of DOJ’s Office of Information Policy, which is charged with overseeing government-wide FOIA compliance. By the end of fiscal 2025, the Defense Department’s FOIA backlog had risen 42 percent to more than 30,000 cases. The Justice Department, meanwhile, mandated a federal court to require a $50,000 bond from journalists seeking expedited processing of a FOIA suit, a move that would effectively create pay-to-play access to public records.

The DOGE litigation reveals how the PRA has been weaponized as a shield rather than serving as a guarantee. The administration has argued throughout that DOGE is a presidential advisory body covered by the PRA, not an agency subject to FOIA, meaning its records are unavailable for at least five years after Trump leaves office. The OLC memo now adds a second layer: even after that five-year window, the records may not be accessible at all, because the statute that mandates eventual disclosure has been declared void.

The intellectual scaffolding for all of this is the unitary executive theory, a constitutional doctrine that has been developing in conservative legal circles for forty years. In its strongest form, the theory holds that Article II vests the entirety of the executive power in the president personally, and that congressional efforts to create independent structures within the executive branch (independent agencies, inspectors general, records mandates) are unconstitutional intrusions on presidential autonomy.

The Supreme Court has absorbed some of this thinking. Seila Law v. CFPB (2020) struck down removal protections for the head of the Consumer Financial Protection Bureau. Trump v. United States (2024) granted sweeping immunity for official presidential acts. The Court’s June 2025 DOGE emergency docket ruling invoked separation-of-powers concerns to narrow discovery into whether DOGE qualifies as an agency at all. Five of the nine current justices worked in Reagan or George W. Bush executive branch legal offices that focused explicitly on finding ways to augment presidential authority.

The PRA memo is the unitary executive theory applied to records. Its argument–that the Presidency is a constitutional office that Congress did not create and cannot regulate; that appropriations authority gives Congress no power to control how the president exercises his duties; and that even the “accommodation process” of political negotiation is what the Constitution requires rather than what the Constitution permits—flows directly from the doctrinal architecture that has been forty years in construction. 

The judicial response has been more complicated than either “the courts will save us” or “the courts have capitulated.” Both caricatures are wrong.

The D.C. Circuit has resisted in places. In December 2025, a unanimous panel including judges appointed by presidents of both parties denied DOGE’s effort to block discovery in the CREW transparency lawsuit. A May 2024 D.C. Circuit decision narrowed the consultant corollary to FOIA’s deliberative process exemption, potentially exposing more agency communications to disclosure. Individual judges have ordered agencies to produce records, preserve documents, and comply with statutory deadlines.

But the same D.C. Circuit ruled in October 2025, in Campaign for Accountability v. DOJ, that OLC opinions are not subject to FOIA’s reading-room provision. The ruling means that OLC opinions like the PRA memo itself are shielded from automatic public disclosure. The Supreme Court has been still more sympathetic to executive secrecy claims, broadening Exemption 4 in Food Marketing Institute v. Argus Leader Media (2019), expanding the deliberative process privilege in U.S. Fish and Wildlife Service v. Sierra Club (2021), and signaling in the DOGE ruling that lower courts should defer to executive branch characterizations of internal structure.

The post-Watergate transparency architecture was built as a system: FOIA (1966, strengthened repeatedly), the Presidential Records Act (1978), the Inspector General Act (1978), and the Ethics in Government Act (1978). Each statute reinforces the others. Remove the PRA, and you have not merely lost one law–you have punched a hole in the architecture itself, rendering the entire structure unstable. 

The questions this leaves are serious ones. What does a presidency look like in which no external entity has enforceable access to its internal deliberations? What does journalism look like when the operational infrastructure for records access has been gutted, and the legal infrastructure has been declared void? What does democratic accountability mean when the public learns what its government did only when its government chooses to tell them?

These questions have answers, and these answers come from places that have walked this path before. Democracies that have experienced democratic backsliding (Hungary under Orbán, Poland under PiS, Turkey under Erdoğan, Venezuela under Chávez and Maduro) almost always begin with attacks on transparency mechanisms before moving on to more visible forms of authoritarian consolidation. Hungary’s Fidesz government weakened freedom-of-information laws, restructured the data protection authority, and pushed sensitive records into categories exempt from disclosure–all years before its more dramatic moves against the courts and the press. Poland’s PiS government used similar techniques. The pattern is consistent enough that comparative scholars treat it as diagnostic.

The sequence matters because transparency infrastructure tends to go first. By the time the more visible erosions arrive, the public no longer has the documentary record it would need to understand what happened, or how. The records that would tell the story have been quietly placed beyond reach.

Olha Burdeina is a junior at Brown University concentrating in Political Science and International and Public Affairs. She is a staff writer for the Brown Undergraduate Law Review and can be contacted at olha_burdeina@brown.edu. 

Ashley Park is a sophomore at Brown University concentrating in Cognitive Neuroscience and Political Science. She is an editor for the Brown Undergraduate Law Review and can be contacted at ashley_h_park@brown.edu.