On Friday, April 25, each of the 24 members of the National Science Board received a short email from the Presidential Personnel Office. It read, in full: “On behalf of President Donald J. Trump, I am writing to inform you that your position as a member of the National Science Board is terminated, effective immediately. Thank you for your service.”
No explanation. No warning. No phone call.
I’ve worked closely with the National Science Board — both as NSF’s Assistant Director for Biological Sciences and as a co-author of a 2022 piece in Issues in Science and Technology that argued, in some detail, for restructuring the Board’s role. I know what the NSB does. I know where its governance model has worked, and where it has struggled. And I want to explain why Friday’s action represents a significant break from the historical pattern of how administrations — including ones that were frustrated with the Board — have navigated this relationship. I also want to suggest that the disruption, as damaging as it is in the short term, might create an opening for a structural reform that is long overdue.
The founding compromise
NSF and the National Science Board were created together in 1950. Their joint structure was itself a political compromise. Vannevar Bush had wanted something like a corporate board — presidentially appointed, with authority to hire and fire the foundation’s director. Senator Kilgore wanted a director accountable directly to the president, not to any board. What emerged was a blend of both visions, with one key question settled in Kilgore’s favor: the director would be accountable to the president, not to the board. But Bush got his presidentially appointed board, and both entities were given shared executive authority over the foundation.
This “two-headed structure,” as NSF historian J. Merton England called it, has no real analog elsewhere in the federal government. NASA’s administrator, for instance, holds unambiguous authority over that agency. NSF’s director manages day-to-day operations and serves as the public face of the foundation — but the Board retains statutory authority to set policy and to approve the foundation’s largest awards and infrastructure investments. That authority is written into the National Science Foundation Act. It cannot be waived by an OMB directive, and it cannot be delegated away.
A structure that has survived previous tensions
The dual-authority structure has never been frictionless. Every administration since Truman has, at some point, found the NSB’s independence inconvenient. Nixon’s science advisers viewed the Board’s independence as an obstacle to the president’s control over research priorities. Reagan’s OMB pushed to reduce NSF’s budget in ways the Board opposed. The pattern across 76 years is consistent: administrations applied pressure, the Board pushed back, and the institutional structure held. The resolution was always a political negotiation, not a structural rupture.
What made that equilibrium possible was a shared understanding — across Democratic and Republican administrations alike — that the NSB’s statutory authority wasn’t optional. Appointments could be shaped to favor policy directions, and they often were. But the institution itself, and the staggered six-year terms that insulate members from any single administration’s full control, remained intact.
Friday’s action is the first time in that 76-year history that a president has simply removed the entire Board at once.
What the NSB does — and where it has struggled
In a 2022 article in Issues in Science and Technology, my colleagues Jessica Rosenberg, Nicholas Robichaud, and I argued that NSF’s governance structure needed reform. The NEON crisis — the near-collapse of the National Ecological Observatory Network during construction, which I was directly involved in managing as BIO’s assistant director — had exposed a real tension at the heart of the dual structure. NSB members are selected for scientific distinction, not for the project management and business expertise required to oversee a $500 million infrastructure build. The crisis produced competing ad hoc processes, overlapping oversight bodies, and structural confusion about who was ultimately responsible for what.
Our recommendation was not to eliminate the Board. It was to refocus it — to pull NSB back from day-to-day management functions and lean into what it can do that no other body in the federal government does: provide independent scientific advice to the president and Congress, across administrations, insulated from the political moment by staggered six-year terms.
That’s the NSB that was dismissed on Friday.
What triggered this
The Board’s public criticism last May of the proposed 55% cut to NSF’s budget appears to be the proximate cause. NSB chair Victor McCrary and his colleagues advised Congress to reject the cut. Congress did. Board member Keivan Stassun, an astrophysicist at Vanderbilt, described the dynamic with characteristic precision: this group of presidential appointees was advising Congress not to follow the president’s wishes.
That is, in fact, exactly what the Board’s design anticipates. Staggered six-year terms exist precisely to create a body capable of offering independent scientific judgment even when the political moment runs against it. The same design logic underlies independent inspectors general and the Federal Reserve. The underlying theory is that some decisions — particularly investments in basic research that only pay off over decades — are too consequential to be made entirely within the horizon of any single election cycle.
The Board had also been bypassed on a major facilities decision. OMB told NSF’s chief of research facilities directly that it would build a new Antarctic research icebreaker — a commitment that required statutory Board approval — with no Board involvement whatsoever. When Board members asked what had happened, the answer was essentially: OMB said so. That sequence illustrates the structural tension that preceded Friday’s action.
What the historical record tells us about what comes next
The deeper historical question is not what happens to the 24 dismissed members — it’s what the institution becomes under whatever replaces them.
The NSB’s value as an advisory body has always rested on a specific institutional property: members who serve across administrations, whose six-year terms mean that no single president ever fully controls the Board’s composition. That property is what allows the Board to offer advice — to Congress, to the president, to the public — that is credibly independent of the political moment. It is also, of course, precisely what makes the Board structurally resistant to any administration that would prefer unanimous alignment.
The historical record of analogous institutions suggests the outcome depends heavily on what happens next. Advisory bodies reconstituted with members selected primarily for political alignment tend to lose their epistemic authority relatively quickly — not because their members are unqualified, but because the perception of independence, once lost, is difficult to restore. The Board’s advice to Congress carried weight partly because it came from a body that had, under previous administrations, offered advice those administrations didn’t always want to hear.
Former Board member Alondra Nelson, who resigned in May 2025 after concluding that the Board had been strategically neutralized, described Friday’s development as the progression from erosion to open elimination. The institutional history supports that framing: what changed Friday was not the direction of travel, but the pace.
A note on what NSF does
The foundation is often described in coverage of this story simply as a “grant agency.” That undersells it considerably.
NSF is the primary federal funder of basic research across non-biomedical science and engineering — the Antarctic stations, the major telescopes, the oceanographic vessels, the long-term ecological observatories. The science behind MRI imaging, cellphone technology, and LASIK surgery — all of it traces partly through NSF. More fundamentally, NSF supports the graduate students and early-career investigators who will produce the next generation of discoveries we cannot yet anticipate. The foundation does this on timescales that no administration and no budget cycle can fully capture.
The NSB that was dismissed on Friday was the statutory body designed to protect that investment across exactly those timescales. That design has survived fourteen presidents. What it looks like under the fifteenth remains to be seen.
A possible way forward: rethinking the relationship between NSB and PCAST
Disruptions of this magnitude, whatever their cause, sometimes create space for structural reforms that would otherwise be impossible to negotiate. This may be one of those moments — though the opportunity it presents is more pointed than a simple merger.
The President’s Council of Advisors on Science and Technology — PCAST — is a presidential advisory body created by executive order, serving at the president’s discretion, with no independent statutory authority over federal research agencies. The National Science Board, by contrast, has deep statutory roots, staggered terms, and dual authority over both NSF governance and national science policy advice. In practice, the two bodies have often operated on parallel tracks, producing overlapping reports and competing for the attention of the same senior officials.
What makes the current moment structurally interesting is not that both bodies are vacant — they are not. The current administration reestablished PCAST in January 2025. It announced its membership in March 2026: a roster dominated by technology sector leaders, including Marc Andreessen, Sergey Brin, Larry Ellison, Jensen Huang, and Mark Zuckerberg. PCAST is active. It is the NSB that is now empty.
That juxtaposition defines the real question Friday’s action raises. The current PCAST represents one coherent vision of what science advice to the president should look like: private sector technologists, oriented toward near-term innovation and commercial application, serving entirely at the president’s pleasure. The NSB tradition represents a different vision: academic researchers and scientists, oriented toward long-term basic research and cross-administration continuity, with statutory independence from any single administration’s priorities. These are not merely different rosters — they reflect genuinely different theories of what science policy is for.
The NSB vacancy is the moment for Congress to ask which model it wants to enshrine, and whether a better design might draw from both. One approach worth considering: a consolidated body with tiered membership — a statutory core of Senate-confirmed members with staggered terms, responsible for NSF governance and independent science policy advice to Congress, combined with a rotating presidential advisory panel of shorter-term appointees focused on near-term innovation priorities and direct White House engagement. The statutory core would preserve the independence and cross-administration continuity that gives science advice its credibility with Congress. The presidential panel would provide the responsiveness to current priorities that PCAST has historically supplied.
The design challenge is real: a body with statutory authority and staggered terms is, by design, harder for any sitting president to fully control, which is precisely what gives its advice credibility. Resolving that tension in legislation has never been easy. But the scientific community and its allies on Capitol Hill would be better served by proposing a durable structural solution than by simply advocating for restoration of the status quo ante — a status quo that, as Friday demonstrated, was more fragile than anyone had fully reckoned with.
There were real arguments to be made about reforming how the Board exercises its authority. I made some of them. But the historical record doesn’t offer a precedent for what a mass dismissal — rather than strategic appointment pressure — does to an institution whose core value is independence. We are in new territory.
Victor McCrary said it plainly: “If the White House wants the golden age of science that Trump has promised, now is not the time to go backward. Instead, we need to spend more.”
The history of American science policy suggests he’s right about the investment argument. The structural question now is whether the vacancy created on Friday becomes simply a void or the opening for something more durable.
Discoverers and Discoveries 