Trump Takes on Big Science

Throughout 2025, the science establishment has warned that the Trump administration poses a lethal threat to “American science expertise as we know it.” The alarums have been overwrought and misleading—until now. The administration’s 2026 budget request for the National Science Foundation does raise legitimate concerns about funding cuts. Yet in other respects, the reforms decried by Big Science have not gone far enough.

Congress created the National Science Foundation in 1950 to “promote the progress of science; to advance the national health, prosperity, and welfare; [and] to secure the national defense.” The foundation disburses up to $9 billion annually to support university researchers in physics, engineering, computing, biology, and chemistry. It boasts an unmatched record of seeding Nobel Prize–winning science.

Before the 2026 budget release, the foundation had already terminated more than 1,700 grants, totaling $1.4 billion. It had capped the amount of overhead that it will henceforth pay universities at 15 percent of a research grant, though it was immediately blocked by the usual federal court injunction. (See “Racist—But Underfunded?,” Spring 2025.) The foundation had embraced a reorganization plan that consolidates divisions and demotes high-level bureaucrats to nonexecutive positions.

Trump opponents cried foul. “The American people deserve a scientific enterprise free from political interference,” California Democrat Zoe Lofgren, ranking member of the House Committee on Science, said in a press release. The termination of grants will lead to the “complete gutting of America’s scientific enterprise,” Sarah Spreitzer, vice president of government relations at the American Council on Education, told Inside Higher Ed. A Columbia University psychologist and NSF grant recipient asserted to the New York Times that the cuts will cede “American leadership in science and technology to China and to other countries.” Science wrote that the reorganization plan plunged an “already battered” NSF into “deeper turmoil.” The restructuring would leave the agency vulnerable to White House pressure to “fund research that suits its ideological bent,” unnamed sources told the magazine.

The claim that the Trump administration might push the NSF to fund research with an “ideological bent” was rich. The NSF has been supporting ideologically driven research for years, much of it through its Directorate for STEM Education. The directorate’s $1.15 billion budget in 2024—a full ninth of the foundation’s $9.2 billion budget and much higher than funding for biology, computer science, and engineering—is just a starting point for gauging how much the NSF spent on education projects. Other directorates, nominally focused on hard science, also distributed education grants.

The NSF’s education grant-making has been focused on racial victimhood. The education directorate plays a key role in boosting the NSF’s diversity metrics. Its program managers—who approve and oversee grants—are disproportionately minorities, especially minority women. Grant recipients also tend to be disproportionately minority. This imbalance may reflect the composition of the applicant pool for once, since America’s schools of education, the feeders for NSF education program managers and education awardees, are themselves disproportionately minority. This skew is even greater in STEM-related education specialties, and not just because those specialties are devoted to formulating racism-based explanations for the underrepresentation of minorities in STEM. These education fields serve as a safe harbor for STEM graduates who opt out of STEM careers, and this category, too, is disproportionately minority.

NSF grant recipient James Holly Jr. is a typical case. In 2023, Holly received nearly $600,000 from the NSF’s Division of Engineering Education and Centers—part of the Directorate for Engineering, illustrating how education-related spending extended beyond the NSF’s Directorate for Education. Holly earned an M.S. in mechanical engineering from Michigan State University in 2014 and then pivoted to education, completing a Ph.D. in engineering education at Purdue University in 2018. Whatever his strengths as a mechanical engineer, his command of antiracism discourse is impeccable.

The abstract of Holly’s NSF project, “Learning from Black Intellectualism: Broadening Epistemic Foundations in Engineering Education to Empower Black Students and Faculty,” deserves an extended excerpt, since it epitomizes what had been the NSF’s education portfolio:

The current discourse around the minimal presence of Black people in engineering is framed in terms of underrepresentation—the disparity between Black people’s demographic representation in the general populace and within the discipline. However, this narrative preserves Whiteness by passively neglecting the culture of racism in engineering. A discourse centered on who can be physically included without engaging the implications of power in knowledge production neglects the ways Black people are forced to give meaning to their experiences through the lens of Whiteness. Recent scholarship within engineering education suggests a need for (1) a modern, reparatory framework for helping engineering faculty and students understand political implications of engineering knowledge; and (2) an equity-focused resource to foster constructive evaluation of teaching. . . .

This CAREER project will 1) examine the effects of recasting engineering knowledge through the legacy of Black intellectualism, and 2) advance educational justice by countering the epistemic violence within engineering and its sense-making practices. The anticipated outcomes of this study will equip engineering faculty with tools for equitable instruction, and more importantly, enhance Black students’ sense of belonging by bridging the gap between their engineering learning and social reality. Fugitive pedagogy will be used to investigate engineering faculty epistemic norms and explore ways to reconstruct disciplinary knowledge through Black intellectualism. The project will implement a social design experimentation methodology to study how engineering education can be transformed toward epistemic equity. Epistemic equity is operationalized through the idea of re-politicizing—grappling with cultural and political implications of technical systems—engineering courses and curriculum. The overarching question guiding the research plan is: How can Black intellectualism be used to re-politicize engineering pedagogy? Engineering faculty will develop a schema (Phase 1), engage in revising a course based on the schema (Phase 2), and develop a teaching evaluation tool to assess the outcomes (Phase 3). Phases 2 and 3 will be repeated in an iterative cycle three times, centering faculty and student voice is the hallmark of the integrated research and education plans.

When Big Science proclaimed throughout spring 2025 that Trump’s budget cuts would devastate American scientific prowess, especially vis-à-vis China, “Learning from Black Intellectualism” was what the science establishment was referring to.

The following features of the Holly abstract were standard. It ignores blacks’ on average rock-bottom mathematical skills. It is this skills gap that causes black underrepresentation in engineering, not “Whiteness,” a “culture of racism,” or “epistemic violence.” (The number of black 12th-graders who are advanced in math nationwide is a statistical zero; 60 percent of black 12th-graders do not possess even basic 12th-grade math skills. The average black score on the math SAT in 2023 was 440 on an 800-point scale, compared with Asians’ average 629 math score.)

The abstract places all responsibility for increasing the representation of blacks in engineering on everyone and everything besides black students and their families. Engineering pedagogy must be “re-politicized” with “Black intellectualism.” Engineering education must be “transformed toward epistemic equity.” Not a word about cracking the books and completing problem sets.

Holly uses scientistic and hothouse rhetoric—“fugitive pedagogy,” “social design experimentation methodology,” “schema” and “phases” “repeated in an iterative cycle three times”—to create the illusion of exacting research protocols.

Multiply “Learning from Black Intellectualism” several hundredfold for a picture of the projects that the NSF had started to shed in the spring.

On April 18, the NSF announced that it would no longer fund projects that “give preference to some groups [based on] protected class or characteristics”—in other words, based on race and sex. On May 9, the NSF announced that it was disbanding its most concentrated source of racially ideological grant-making: the Division of Equity for Excellence in STEM, housed within the Directorate for STEM Education. By May 21, it had cut off funding for 1,752 grants in progress, including Holly’s. Naturally, the press played the race card, noting that the cuts “reduced the diversity of NSF’s pool of funded scientists,” as Science put it. Blacks suffered the heaviest blow, according to Science, with a cancellation rate four times higher than their representation among total NSF grantees. Such a disparity is hardly surprising, given that racism-themed grants serve as a vehicle for increasing black representation among NSF awardees.

Getting rid of the Division of Equity for Excellence in STEM was a good start, but the education directorate contains three other divisions as well: Graduate Education, Research on Learning in Formal and Informal Settings, and Undergraduate Education. All should be eliminated.

Like the Division of Equity for Excellence, those three additional divisions are mere extensions of education schools, whose effect on the transmission of knowledge has been disastrous. The main purpose of graduate schools of education is to obscure a basic truth: student learning results from self-discipline, demanding coursework, and unbending expectations from teachers who are masters of content. Education schools conceal that truth with vast clouds of verbal squid ink. Dip randomly into the remaining portfolio of the Directorate for STEM Education, and you come up with grant solicitations such as the following:

The first goal of this funding opportunity is to encourage the scientific study of theories, frameworks, and models for the translation and diffusion of knowledge, especially between fields and across contexts and levels-of-analysis (e.g., biological to cognitive/socioemotional to behavioral; individual to classroom to broader demographic variables; lab to classroom to school to district). . . . Proposals may also address the leveraging of effective classroom practices toward the enrichment of foundational research, constructs and models. We note that bi-directional movement across boundaries is a mutually beneficial reciprocal process. . . . The outcome of such a project would lay the methodological, theoretical, empirical, design, or social foundation for conducting systematic work at the next stage of development or at the next level of analysis.

The NSF has no business funding such vapidity; it contributes nothing to our knowledge of nature’s laws. The whole education directorate should go.

The NSF’s Directorate for Social, Behavioral, and Economic Sciences is another source of grant-making premised on academic leftism. Consider the aforementioned Columbia University psychologist who lamented that the U.S. is ceding scientific ground to China. His terminated grant—from the Division of Behavioral and Cognitive Sciences within the Directorate for Social, Behavioral, and Economic Sciences—focused on how “social inequities such as gender and racial disparities” are shaped by facial stereotypes and other “learned stereotypes” about race and gender. It’s doubtful that China is directing its science funding toward combating race and gender stereotyping. Rerouting U.S. taxpayer dollars away from such studies would strengthen—not hinder—our ability to compete scientifically with China.

Such was the state of play before the FY 2026 funding request: the science establishment was crying bloody murder because the NSF had begun lopping off some of its most egregiously politicized grants. The establishment portrayed those grants as key to future scientific progress, relying on the public’s ignorance about their content and the press’s unwillingness to reveal it.

Enter the 2026 budget, released on May 30, 2025. Funding for research and related activities has been cut 61 percent, or $5 billion. The NSF’s total budget has been cut 55 percent, or $5.12 billion. Sadly, the education and social-sciences directorates survived the budget ax, even if they now exist in a much diminished state. The education directorate went from a $1.15 billion allocation in 2024 to $288 million in 2026, a 75 percent drop. The Directorate of Social, Behavioral, and Economic Sciences went from a $290 million allocation in 2024 to $94 million in 2026, a 67 percent drop. The biggest surprise is that the Division of Equity for Excellence in STEM within the education directorate has been exhumed, presumably in response to yet another federal court order, this time out of San Francisco, blocking the Trump administration from laying off any government employees. True, the division’s budget has been cut nearly 80 percent—from $214 million in 2024 to $43 million in 2026—but that $43 million can do a lot of mischief.

Most bracingly, the 2026 budget zeros out nearly the entirety of a category of grants known as “Broadening Participation.” These grants represented the culmination of Congress’s decades-long mania for imposing nonscientific goals on the foundation. In 1980, the Science and Technology Equal Opportunities Act, sponsored by Senator Edward Kennedy, established that it was U.S. policy to ensure that “women and minorities have equal opportunity in science and technical fields.” In 2010, Congress forbade the NSF from evaluating grants solely on scientific merit. Instead, according to the America COMPETES Reauthorization Act, scientists would have to justify their research according to its “broader impacts.” Any notion that the NSF would focus exclusively on basic science became a relic of a distant age. Insufficiently rousing “broader impacts” statements have torpedoed otherwise vital scientific proposals.

The broader-impacts mandate gave rise to the Broadening Participation portfolio. Most of NSF’s Broadening Participation portfolio consists of the predictable racial and gender equity grants. They are gone, with these exceptions: grants to faculty and undergraduates at HBCUs (Historically Black Colleges and Universities) and to tribal colleges. Those categories are required by statute.

But the Broadening Participation portfolio contains another type of grant—those designed to spread science funding around on the basis of geography. These so-called EPSCoR (Established Program to Stimulate Competitive Research) grants are dearly beloved of Congress, for they allow representatives to brag of bringing the science bacon home, regardless of whether their district’s colleges are likely to make breakthrough discoveries.

EPSCoR grants came out relatively unscathed, compared with identity-based Broadening Participation grants. But anti-meritocratic diversity policies are no more acceptable when diversity is defined by geography than when it is defined by race and sex.

While the cuts to the education and social-science directorates were too timid, cuts to the hard-science directorates were arguably too sweeping: biological sciences is down 71.5 percent; mathematical and physical sciences, which includes chemistry, physics, and astronomy, is down 67 percent; engineering is down 75 percent, and computer and information science and engineering is down 65 percent. Some major projects have been eliminated entirely. The NSF will stop funding Hawaii’s Thirty Meter Telescope, in favor of the Giant Magellan Telescope, planned for Chile. An astrophysicist (and Trump supporter) laments: “We would appear to be handing over a century of world leadership in astronomy back to our European competitors. At least for a generation or two.”

Part of the science reductions can be explained by the NSF’s intention to cap its reimbursements for the indirect costs of research (if the courts allow it to do so). Another part can be attributed to the administration’s overzealous animosity toward climate and clean energy research, though the budget does not spell out the extent of those climate and energy cuts. It is also true that “equity” and its offshoot, “sustainability,” have infiltrated the hard sciences. A 2022 White House report recommended that scientific research focus on “research questions, samples, and settings that reflect the diversity of the U.S. population.” A lawsuit filed on May 28 by 16 state attorneys general against the NSF’s DEI cuts provides an example of such research: a project to integrate the “knowledge” of “Indigenous communities” into the science of protecting the Great Lakes ecosystem. It’s hard to believe that such silly grants make up the bulk of the hard-science portfolios, however, even after the Biden administration’s diversity push.

Two additional things may be going on: a determination to cut as an end in itself; and a preference for the latest applied technology over basic discovery research.

The government cannot fund everything, obviously, and there is no reason to think that previous levels of funding represented a Platonic ideal. The Trump administration’s science advisor, Michael Kratsios, claimed in May that federal science funding is seeing diminishing returns. Money spent does not correlate with scientific impact, Kratsios said. But cuts on the order of 70 percent to core fields, even taking into account indirect cost caps and climate-related cuts, risk X-ing out breakthrough findings.

The May 30 budget request reads like a prospectus for a tech startup. Its “prioritized” activities are artificial intelligence, quantum information science, and the Directorate for Technology, Innovation, and Partnerships, the last akin to a tech incubator for small businesses. These “investments” have not been cut, explains the budget proposal, because they “complement private-sector R&D and offer strong potential to drive economic growth and strengthen U.S. technological leadership.” Other favored areas are advanced manufacturing, microelectronics and semiconductors, and advanced wireless, presumably because they, too, help “harness the full power of American innovation by empowering entrepreneurs and unleashing private-sector creativity.”

But however much Congress itself has been imposing “broader impacts” mandates on the NSF, it would be as much of a mistake to reorient the agency toward research perceived to be economically useful as it was for the agency to adopt social-justice goals. The private sector is already moving full-speed ahead on high-tech research and applications. It has less incentive to fund curiosity-driven research into the laws of the universe. There is no such thing as too much knowledge, as long as it is pursued through objective means, free from political assumptions and open to falsification.

The administration could fight more worthwhile battles. It should persuade congressional Republicans to provide the White House with an unambiguous charter for its reform efforts. Congress should strip all identity-politics language from NSF budgetary authorizations and reject the notion that researchers must justify their work on nonscientific “broader impacts” grounds. As long as that “broader impacts” and “broadening participation” language remains in the statutes, lawsuits like the one filed by the 16 state attorneys general are likely to succeed.

Congress should also extricate the NSF from all teacher training and education research. While the foundation’s original charter did establish a Division of Scientific Personnel and Education, the “education” in question referred to graduate-level study. According to the 1950 charter, the division’s role was to oversee programs “relating to the granting of scholarships and graduate fellowships,” with selections made “solely on the basis of ability.” The original drafters could hardly have imagined the assault on meritocracy that would follow.

The Trump administration should launch a campaign to treat scientists like adults again. It should invite nominations for the most burdensome or insulting bureaucratic requirements and restore discretion to project managers and researchers. University of Southern California chemistry professor Anna Krylov provided a harrowing description in April of the clerical demands associated with a simple single-researcher project. Only 15 pages of Krylov’s 64-page proposal related to her project—computer-modeling the interaction of molecules with X-rays—and even those 15 ostensibly science-based pages included the usual broader-impacts section. The proposal also included a mentoring plan, a list of Krylov’s “synergistic activities”—such as editorial work or conference organizing—a data-management plan, and a conflict-of-interest spreadsheet listing collaborators and former associates. Preparing the proposal took the assistance of three USC administrators over many days.

The White House has started a long-overdue shake-up in Big Science and Big Academia. While it might have proceeded more surgically, it will now need to be better armed against the end-of-times prophesying arising from those intertwined establishments. It should make the case that federal science funding should not go to social or economic impact, “equity,” or any particular worldview, but only to the unleashing of human genius in its confrontation with natural mystery.

Top Photo: The Trump administration argues that NSF funding has seen diminishing scientific returns. (Mehmet Eser/ZUMA Press Wire/ZUMA Wire/Alamy Live News)