Inside the National Gallery’s misleading “public” archive

When the National Gallery of Art announced its new “public” archival catalog, it sounded like a breakthrough. The press release promised a new era of transparency, inviting “audiences around the country, including those who may never be able to visit us in Washington,” to peer inside the institution’s history.

As a journalist who covers governance and accountability, I was curious. I logged on expecting a tool akin to the federal court system’s PACER database or the National Archives’ growing digital repository—a place where keyword searches yield documents and public access is the baseline expectation.

In the news release, the National Gallery said the online archive allows users to explore “more than 86,000 resources documenting the National Gallery’s history and activities since its founding in 1937.” Since its founding—implying up to the current day. “Researchers and the general public will now be able to directly search the National Gallery’s archival records,” the institution said.

I wasn’t expecting state secrets. But I did expect some understanding of the museum’s recent history: the decisions of its directors, basic administrative context. That is how the public tends to use archives—to understand institutions that shape civic and cultural life. What I found instead was a system that revealed almost nothing about contemporary operations, without any warning that this would be the case.

The empty search results

My first test was routine: a background check on leadership. Earl Powell III served as NGA director for 26 years—nearly the entire modern era of the museum. A functioning federal archive should return a deluge of records: memos, policy directives, board minutes. A search for “Earl Powell” returned just 167 entries, mostly press clips and event material.

Searching for current director Kaywin Feldman yielded six records, two of which were duplicates. The “director records” link led only to collections for leaders who served before 1992. To archivists, this may reflect standard accessioning policy, which delays the transfer of recent administrative files for privacy and legal review. But this absence was not explained anywhere. A public-facing archive that excludes three decades of leadership records requires at least a sentence acknowledging why.

The NGA creates an accountability gap when it provides silence instead of an explanation of its internal rules.

The National Gallery of Art is a massive physical space, and its archive fills many rooms. The argument is not that all of these paper files should be instantly digitized—that is a budgetary issue. The failure lies in the descriptive level of the digital catalog.

Archivists defend that the catalog only describes collections at the "box" or "series" level, not the individual item. But for a federally funded, FOIA-exempt institution, this low level of description is a systemic choice.

And to be clear: the point is not that every document deserves item-level description, but that a public-facing archive must clearly communicate what is included, what is excluded, and what remains unprocessed. Without that, the user cannot even know what questions the archive can answer.

Born-digital irony

But the second failure felt concerning. Even when the catalog listed “born-digital” files—records that already exist as PDFs—the system did not allow visitors to view them. Instead, users were told to email NGA staff to “request” the files.

This staff-mediated access introduces a bottleneck no modern system can withstand. If hundreds of people requested records tomorrow, the process would be overwhelmed. And most people will never email, never travel to a reading room, and never know what to request because they cannot see what exists. The system assumes the public knows what’s in the boxes before they look.

The museum gestured at improvements, stating: “In the future, direct access to digital archival materials will be available through the catalog.” It did not say when.

What the NGA actually launched

After publishing an article about these gaps, I posted it on a Reddit forum full of archivists and museum workers. I asked: “What is the actual benefit of this archive? From a journalistic perspective, the archive seems lacking.”

The reaction was swift. “What an asinine article,” one commenter wrote. Another said I “shouldn’t be writing articles about archives.”

I clarified the article to reflect archival terminology: “The catalog launched by the NGA is, in archival terms, a finding aid, a standard professional tool analogous to a library index that describes what materials exist for researchers.”

But criticism continued, centered on the idea that I “lacked understanding of how archives function,” despite my quoting the NGA’s own framing: that this tool is for “researchers and the general public.” These expectations don’t come from me—they come from the museum’s own language promising a searchable archive for “the general public.”

If the NGA’s new tool is only a finding-aids database—an index for insiders—then why did its press release promise a tool for “audiences around the country” to “directly search” archival records? The public should not have to understand archival theory to use a resource marketed to them.

The FOIA issue no one acknowledged

The biggest gap in the conversation was legal. The National Gallery of Art is exempt from the Freedom of Information Act. That means the public cannot compel the release of internal correspondence, board minutes, or administrative records.

Several commenters assumed FOIA applied: “All of Earl Powell’s records are accessible through FOIA,” one wrote. When I pointed out this was false, I was told, “Then you have even less of an ability to access records from them”—yes—and that “You have the same access… as any nonprofit,” which is incorrect. The NGA is not a nonprofit. It is a federal cultural institution created by Congress.

If the NGA chooses to release only 167 records covering a 26-year directorate, there is no legal mechanism to challenge that decision. The “finding aid” database becomes the only window into the institution—and if it reveals virtually nothing, that window is ornamental, not functional.

NGA’s response—and the silence afterward

After publication, an NGA spokesperson emailed me: “Much of our archival material is indeed digitized,” spokesman Christopher Abanavas wrote. “Once users find a resource… they can easily get in touch with our archivists, who will share a digital copy.”

I wrote back with follow-up questions: What percentage of material is digitized? Why not allow automated access to existing digital files? How many staff are dedicated to fulfilling requests?

Forty-eight hours later, I had heard nothing. If they cannot answer a journalist quickly after contacting him for a correction, what hope does the average citizen have?

The argument archivists kept missing

Archivists responding on Reddit leaned heavily on the idea that finding aids describe collections at the “box” or “series” level, not the “item” level. They said recent director records weren’t absent—they were simply unprocessed, legally sensitive or too new. They pointed to the Americans with Disabilities Act and copyright laws as barriers. “It’s a lawsuit,” one warned.

Fair enough—those are real constraints. But none were mentioned in the NGA’s public messaging. If the archive excludes anything from the past 30 years, the museum should say so plainly. If viewing a PDF requires staff mediation, that should be stated upfront.

When I pointed out that other federal databases provide item-level, searchable access to millions of documents, the argument that such access is impossible grew weaker. Those agencies handle ADA compliance, redaction, and scale. The fact that the NGA has not invested in technology to automate these processes reflects institutional choices.

A larger issue begins to reveal itself

The more archivists insisted the NGA’s archive was normal, the clearer it became that the problem was not the National Gallery alone.

This is where Part II begins.