Help Us Stay Paywall-Free
Foreground: White House science advisor Michael Kratsios (far left), President Donald Trump, and Office of Management and Budget director Russell Vought (far right) in June 2026. Background: President Lyndon Johnson visits the National Institutes of Health. Images: The White House / NIH via Flickr, illustration by Boston Review
Trump has killed the postwar social contract, but its demise was long in the making. To protect science as a public good, we need a different approach.
The second Trump administration has presided over the greatest attack on scientific research since the end of World War II. It has frozen or terminated $2.3 billion in unspent funds at the National Institutes of Health (NIH) and $700 million at the National Science Foundation (NSF) across thousands of grants. It has targeted research on cancer, vaccines, environmental protection, climate change, and public health as well as diversity, equity, and inclusion initiatives. It has overseen thousands of layoffs from key scientific agencies including the Centers for Disease Control and Prevention, the Environmental Protection Agency, and the National Aeronautics and Space Administration. And it proposed unprecedented budget cuts to both the NIH and NSF, promising to slash them by nearly half or more.
Though Congress rejected the cuts when it passed the 2026 federal budget, these moves have created a climate of such fear and confusion that three-quarters of scientists report they have considered leaving the United States. According to Science, the U.S. government has lost over ten thousand STEM PhDs since Trump took office. Faced with growing uncertainty, universities have paused hiring and curtailed PhD admissions in STEM fields. And the campaign to overhaul science policy hasn’t stopped. In May, the Office of Management and Budget (OMB) proposed sweeping changes to the federal grantmaking process, including mandatory pre-issuance review of all grants by political appointees; in late July, the director of the White House Office of Science and Technology Policy, Michael Kratsios, released a report calling for investment in AI-driven research, tighter integration with the private sector, a deprioritization of “legacy institutions,” and new funding mechanisms besides peer review.
As jarring as these developments are, many have deeper roots than Trump and his controversial appointees: Robert F. Kennedy Jr. at Health and Human Services and Jay Bhattacharya at NIH. Every year between 2009 and 2014, Senator Tom Coburn of Oklahoma released a lengthy report arguing that a significant percentage of taxpayer money for science went to “fraud, waste, and abuse.” All the while, conservative think tanks such as the Heritage Foundation and the Cato Institute have run reports and editorials on “How to Stop Wasting Money on Science,” “The Case Against Public Science,” and “Indirect Costs: How Taxpayers Subsidize University Nonsense.” With guidance from Heritage’s Project 2025, the Trump administration is finally carrying out many changes that Republican policymakers, think tanks, and scholars have been advocating for decades.
With guidance from Project 2025, the Trump administration is finally carrying out changes the GOP has been advocating for decades.
To understand how we got here, we have to understand the unusual structure that has governed the allocation of U.S. public funding for the better part of a century. The prevailing system, established after World War II, combined the largest amount of public funding for basic research in the world—around $45 billion annually in recent years, or about 5 percent of federal, non-defense discretionary spending—with a high degree of autonomy to scientists in the distribution of funds. While the private sector has been the largest source of overall U.S. research and development spending since the early 1980s, the government remains the largest source of funding for basic research. And few areas of public funding, apart from central banking and certain areas of national security, have combined so much money with such high delegation to nongovernmental experts. While Congress controls total scientific budgets and agency structures, individual funding decisions—through agencies like the NSF—are effectively made by program officers who rely heavily on peer review by independent scientists, based mostly at research universities. Since 1997, the NSF process has evaluated proposals on the basis of “broader impacts” as well as intrinsic scientific merit, but it remains highly insulated from public input and deliberation.
This structure worked to the extent that the public accepted the idea that scientists pursuing their own curiosity brings about the greatest public benefit. But as American politics has become increasingly polarized—often around scientific issues, including climate change and vaccines—that conviction is less certain than ever. There is now a 25-point gap in Democrats’ and Republicans’ trust in scientists, and only 51 percent of Republicans say that science has had a mostly positive role in society, down from 70 percent before the COVID-19 pandemic.
That’s not to say that all of Trump’s actions on science are popular, much less the result of a democratic mandate. A large majority of Americans continue to believe that government investment in science is worthwhile for society over time, and many of the administration’s attacks contravene areas of rare bipartisan agreement. To defend American science and chart a new vision for its future, we should build on the public’s enduring support for public funding while addressing real democratic deficits. The truth is, there never was a convincing democratic justification for the postwar scientific compact. But by learning from more democratic paths not taken, we can build a system that more fully serves the public.
In late 1944 President Franklin Roosevelt asked engineer Vannevar Bush, then head of the Office of Scientific Research and Development (OSRD) and overseer of the Manhattan Project, to develop a vision for the role of scientific research in peacetime. Americans largely appreciated the role scientists played in winning the war, but no comparable case had been made for high levels of funding for nonmilitary research. At the same time, wartime investment had changed the nature and scope of scientific research; cutting-edge science in fields like nuclear physics and electronics had become increasingly dependent on large amounts of funding. Bush’s challenge was to devise a persuasive justification for continued public investment in basic research.
His answer—developed in a fifty-page report called Science, The Endless Frontier—became one of the most influential public policy documents of the twentieth century, laying the foundation for publicly funded science in modern democracies. Bush’s argument had two key features. First, public funding could be justified on the basis of expected material benefits. “Advances in science,” he wrote,
when put to practical use mean more jobs, higher wages, shorter hours, more abundant crops, more leisure for recreation, for study, for learning how to live without the deadening drudgery which has been the burden of the common man for ages past. Advances in science will also bring higher standards of living, will lead to the prevention or cure of diseases, will promote conservation of our limited national resources, and will assure means of defense against aggression.
Second, the public would benefit most if scientists themselves—not political appointees—decided what to work on and how to distribute funds. According to Bush, “scientific progress on a broad front results from the free play of intellects, working on subjects of their own choice, in the manner dictated by their curiosity for exploration of the unknown.”
This vision was quickly put into practice. The NIH, established in 1930, expanded dramatically in the 1940s and 1950s, becoming the largest federal funder of biomedical research, and the NSF was created in 1950 to support basic science across a variety of disciplines. Overall, annual public spending on science grew from hundreds of millions of dollars during World War II to several billion by the late 1950s.
These arrangements gave science funding an unusual status within the burgeoning administrative state. Although spending was justified on the basis of expected public benefits, the peer review and allocation process would not consider public need, develop missions, or respond to democratic steering. In a mechanism not unlike Adam Smith’s invisible hand, taxpayers were asked to believe that public benefits would follow if scientists were left to their own devices. Just as Smith had claimed that society as a whole would benefit from the individual pursuit of self-interest, so too would society benefit from individual scientists’ free pursuit of their curiosity. In these key respects, Bush’s vision was more radical than Progressive and New Deal models of expert administration. His ideal wasn’t a bureaucratic technocracy—government by experts—but a self-governing professional community shielded from bureaucracy, even as it relied on government funding.
The similarity between Bush’s and Smith’s arguments ran deep. In the postwar era, the specter of Lysenkoism in the Soviet Union—among other pressures of the Cold War—inspired comparisons between scientific freedom in the West and the freedom of capitalist markets from state intervention. The analogy was given fullest expression in the philosophical writings of Hungarian-British scientist Michael Polanyi, a staunch critic of communism and central planning. In a series of papers in the 1940s, collected in The Logic of Liberty (1951), Polanyi insisted that scientific progress required freedom from political interference, not least because the course of science was very difficult to predict. On his view, the independent pursuits of individual scientists would be coordinated “as by an invisible hand” toward the greatest discoveries. “To plan science” with a “view to [its] interest to some industry or other practical concern,” Polanyi wrote, “would be simply to kill science.”
These were not the only visions of science on offer, however. As the historian of science Madeleine Baker has traced, Polanyi developed his critique of scientific planning in opposition to arguments made by left-wing figures in the 1930s and ’40s. From the liberal philosopher John Dewey to Vienna Circle logical positivists Otto Neurath, Philipp Frank, and Rudolf Carnap to the Marxist crystallographer and historian J. D. Bernal, all started from the premise that science was already being shaped by political, economic, and military goals. The choice, as they saw it, was not whether to plan science, but by whom it would be planned. Corporate interference and restrictive patent laws were diverting scientific research away from socially beneficial projects, toward private profit; the state could redirect it toward human welfare. Moreover, leaving the direction of science entirely to scientists would mean that a scientific elite—“the gerontocracy”—would come to hold disproportionate power.
These kinds of considerations were precisely what drove U.S. Senator Harley Kilgore, a Democrat from West Virginia, to propose the creation of a National Science Foundation in 1942. While Kilgore agreed with Bush about the need for a new scientific agency, he worried that Bush’s vision—already clear from his work at OSRD—would create a scientific elite and concentrate public funds in the hands of a few top institutions. Instead, inspired by New Deal ideals of accountability and antimonopolism, Kilgore introduced legislation for an agency with strong congressional oversight, regional and mission-oriented distribution of funds, and public ownership of patents resulting from federally funded research.
The choice, according to this rival vision, was not whether to plan science, but by whom it would be planned.
An acrimonious debate ensued. The nation’s scientific leadership vigorously opposed Kilgore’s proposal, objecting that its provisions would expose science to undue political pressure. The presidents of Harvard, MIT, Johns Hopkins, and Caltech feared the model would shift research strength away from their universities and toward the federal government. To turn Congress against the proposal, Bush and his allies compared it to Soviet-style planning. After prolonged legislative battles—and a 1947 veto by Harry Truman, who wanted a presidentially appointed director—a bill creating the NSF finally became law in 1950. The result was a far cry from Kilgore’s original vision and largely in the mold of Bush’s proposal: the new agency was governed by a board of prominent scientists and distributed grants through a competitive peer review process solely according to scientific merit. As Kilgore anticipated, the NSF was soon channeling most of its funding toward basic research at a small number of top universities. In 1965, Lyndon Johnson noted with concern that twenty universities were receiving half of all federal money.
Meanwhile, the philosophical dispute about which approach to funding would bring about the most significant discoveries and the greatest public benefit was never resolved. The unpredictability of scientific progress—the lynchpin of Polanyi’s argument—could just as well justify distributing funds randomly, equally, or through some other decision procedure. In the end, the choice between Bush’s and Kilgore’s visions was settled not on the merits but by politics. Even though Kilgore’s proposal was less technocratic and centralized than the vision endorsed by left-wing science planners, the charge that it smacked of Soviet-like political interference proved decisive in the context of Cold War anticommunism. Ironically, within a few decades, Bush’s own model—elitist in practice, but justified by appeal to the common good—would itself come to be criticized as an unjustified exercise of state power.
This critique came to a head beginning in the 1970s as government funding in the name of the public interest came under attack, both politically and philosophically. In this context, influential liberal and neoliberal thinkers converged on the view that it was difficult to justify state-funded science by appealing to the common good.
In A Theory of Justice (1971), for example, John Rawls divided state expenditures into two categories: those that are required by the principles of justice (as agreed upon in the “original position,” behind the veil of ignorance), and those that are permissible but not required. The former—the institutions that make up the “basic structure of society”—are not subject to a popular vote because their imposition on all citizens is justified by the fact that they are the basis of a just, mutually advantageous cooperative venture among free and equal citizens. But funding for both science and for the arts falls in the second category. We can’t justify funding these by reference to the “common good,” Rawls argued, because what counts as the good is controversial, and the state should avoid imposing unwanted burdens on citizens based on values they do not hold. The principle regulating their provision should instead be solely that of individual benefit, and individuals should be taxed in proportion to the benefits that they receive.
In making this argument, Rawls drew on the so-called “unanimity principle” developed by Swedish economist Knut Wicksell. In an 1896 article, Wicksell had argued that if a public good is an efficient use of social resources, there must be a distribution of tax burdens that would gain unanimous approval. On this line of thought, decision-makers should consider proposals for public goods, together with different schemes for the distribution of tax burdens, and only provide those that are unanimously approved. This would ensure that people who derive no benefit from the good would not be forced to pay, and the distribution of burdens across individuals would track the value of the good for each person.
But this is a strange theory of public goods. For one thing, it ignores that individuals might game the system to secure better deals for themselves. The system gives individuals an incentive to misrepresent their preferences in order to secure a lower tax rate for goods that they would like to have provided. Since everyone has this incentive, the unanimity principle will result in the under-provision of public goods—a version of the classic “free rider” problem. But even if we bracket self-serving strategic behavior, the rule delimits an extremely narrow scope for state provision, allowing only those goods that constitute a Pareto improvement—those that make someone better off without making anyone else worse off. “The principles of justice,” Rawls wrote, “do not permit subsidizing universities and institutes, or opera and the theater, on the grounds that these institutions are intrinsically valuable and those who engage in them are to be supported even at some significant expense to others who do not receive compensating benefits.” Doing so, he argued, would be equivalent to forcing people to subsidize the private expenses of others.

To sustain the vitality of science and secure funding for the long term, we have to make science a true public good—and give the public a stake in its realization.
This became a fundamental tenet of liberal conceptions of state funding. In the 1990s, when Republicans attacked government funding for the National Endowment for the Arts, most liberal theorists agreed it was difficult to justify, assuming that support for “high” art went against the preferences of the majority. (In a notable exception, Ronald Dworkin attempted to defend state support, but his argument rested on the premise that forms of art often associated with “high” culture contributed more to a rich and diverse culture than more popular forms—a highly controversial claim from a liberal point of view.)
In fact, Rawls’s view wasn’t all that different from economist Milton Friedman’s—despite their very different foundational commitments. Friedman strongly opposed public funding for science and called for the abolition of the NSF and the NIH, along with all federal funding for higher education. “What ethical justification do you have for extracting tax money from people for purposes that do not yield them some greater benefit?” Friedman asked in 1980. “You have to be able to say that the extra dollar spent on research will produce more than a dollar’s worth of benefit to the person from whom the dollar was taken. This is a hard proposition to establish in the case of science.” Rawls refrained from taking a position on the issue; he merely said it should be left to individuals to decide. He would refine his view in the mid-1990s, arguing that a fair decision procedure such as majority rule could be used to determine which public goods should be funded and provided. But even then, Rawls shied away from a more robust defense of science.
While these philosophical debates were unfolding, publicly funded science became increasingly vulnerable. The contention that it is wasteful, inefficient, and difficult to justify became an influential strand of the Republican Party’s rhetoric, sometimes with Democratic support. (It was in this context that the NSF began to evaluate grant proposals on the basis of “utility or relevance of the research”—alongside merit—beginning in 1974.) In the absence of a powerful rival vision, and in keeping with trends throughout the public sector in the early years of the neoliberal turn, American science became increasingly marketized, and its benefits were increasingly privatized. In 1980, private R&D spending overtook public spending for the first time since the end of World War II. That same year, the Bayh-Dole Act revised federal procurement regulation to allow universities to patent and commercially benefit from publicly funded research. Scientists and universities increasingly began to operate like businesses—focusing on technological applications for research, seeking private sources of funding, and pursuing projects with clear startup or profit potential. As Philip Mirowski argues in Science-Mart: Privatizing American Science (2011), the result was a new model and ethos for science that has endured to the present: the scientist as entrepreneur, research findings as potential commodities, and the university as a partner to industry, run according to the same corporate logic.
These changes only further undermined the alleged justification for the NSF’s founding vision of scientific autonomy. If scientists began pursuing projects for profit, the link between scientific autonomy and the common good—tenuous from the start—became impossible to maintain. Much as the advocates of science planning had predicted, scientists wound up beholden to the pressures of markets and private corporations. Private companies not only profited off technologies developed on the basis of publicly funded research; they also took credit for them, benefitting from political rhetoric and popular narratives portraying the private sector as the true source of innovation.
In reality, the U.S. state’s significant funding for basic research has helped produce some of the most significant innovations of the past few decades, from the smartphone to GPS, voice recognition, breakthroughs in robotics and batteries, and many new drugs and biomedical discoveries, including vaccines and CRISPR gene editing. The long time horizon and risk-taking of public funding made these technologies possible, but they were patented by private companies and sold back to the public whose taxes had financed the basic research on which these innovations relied. In this context, it is hardly surprising that taxpayers failed to see the benefits of the research they were supporting.
All this has dovetailed with the increasing politicization of science, starting with the changing political alignment of the scientific community itself. Along with bitter partisan debates about the public health response to COVID-19, the growing salience of climate change science since the 2000s has bolstered the perception that scientists increasingly support the Democratic Party. This perception is not wrong: a landmark Pew research poll from 2009—the latest comprehensive study available—showed that only 6 percent of scientists identified as Republicans. (Other evidence suggests the fraction used to be four times as large, or more.) Since then, trust in scientists has gradually declined among Republicans and increased among Democrats, and political polarization has only intensified. In these circumstances, it’s no surprise that GOP attacks on publicly funded science have gained traction: the more science comes to seem identified with Democrats, the easier it is for Republicans to argue that the status quo doesn’t truly serve the public. The foundation of the postwar science compact was always fragile, and all signs suggest it has now collapsed.
Where do we go from here? It is hard to see how we could possibly restore consensus around a top-down vision for the role of science in society. Nor is it likely we can renew the public’s faith in the claim that scientists simply pursuing their curiosity will serve the public interest—not after decades of private companies profiting on the back of publicly funded research. To sustain the vitality of science and secure funding for the long term, we need a more democratic vision that makes science truly a public good and gives the public a stake in its realization.
What has always been missing in these debates is the idea that the public good of science goes far beyond the material benefits it delivers. Science certainly plays a key role in fueling economic growth, driving innovation, and improving health, but these functions do not exhaust its significance. On the contrary, science is the precondition of democratic government in modern societies. Unlike other public goods like public infrastructure or even art, science claims to provide fundamental truths that shape our most basic and objective understanding of the world, which in turn shape our ability to deliberate about political issues and hold policymakers accountable. Defenses of science that double down on its contribution to GDP or the treatment of disease miss this fundamental political function of scientific knowledge—something the Trump administration seems to understand.
It’s not just that claims to scientific truth are ubiquitous in policymaking, helping societies cope with uncertainty and change. Scientific claims can themselves drive new political agendas and become sources of uncertainty and change. Moreover, the way that scientists define problems and pursue solutions both offers political possibilities and restricts them. Whether the relationship between science and democracy is a productive one depends not only on the quality and reliability of knowledge a society produces but also on decisions about what types of knowledge are pursued and how. The power to fund science and decide on the distribution of funds is thus a fundamentally political power, and making science a truly public good requires giving citizens more say over its direction.
At first glance, the political review mandated by OMB’s recently proposed change to the federal grant system might seem to address these democratic deficits. But replacing the unaccountable judgment of scientific elites with the unaccountable judgment of presidential appointees—empowered to terminate research “for convenience” whenever priorities shift—is no improvement in democratic legitimacy. Neither is Kratsios’s recent report, Science: A New Golden Age, despite its embrace of mission-oriented projects. Heavy reliance on AI, increased emphasis on commercial incentives, and a reallocation of funding to the private sector will only further erode the public interest that publicly funded science should serve.
A better place to start is Kilgore’s vision: making science answerable to the public through stronger congressional oversight, regional equity in funding, and public ownership of patents. But given the outsize influence of interest groups, including scientific lobbies, over the budget process today, even broad congressional guidance is not enough. We need new institutions that can direct research toward the needs and problems of the public and give ordinary people more say in the determination of scientific priorities. To this end, we could establish adversarial science courts, where scientists would present the case for research agendas to a jury of randomly selected citizens, who would then deliberate about the value of different scientific agendas and make broad recommendations for funding priorities. These could be modeled after successful participatory budgeting processes such as those in Portugal, Porto Alegre, Paris, and New York City.
These participatory institutions could be combined with several structural changes at the federal level in the spirit of Kilgore’s rejected proposal. A greater portion of funds could be distributed for mission-oriented projects with more coordination across a range of actors and institutions to meet broad societal goals identified by science courts. Grant allocation could incorporate equity concerns across geographic regions as well as at the level of communities, prioritizing funding for issues affecting neglected areas and underserved populations. The government could also include stronger public-interest requirements on federally funded research to prevent monetization at the expense of taxpayers. Revenue sharing and public equity clauses as well as mandatory public access rules for products could shift the incentives for researchers pursuing profits and ensure that the public benefits from research that it pays for. Finally, in a nod to Kilgore’s antimonopoly concerns, traditional peer review processes could be reformed to include experiments with lotteries among proposals that pass a quality threshold. Among other things, lotteries could help prevent elite institutions and dominant paradigms from monopolizing funding at the expense of more innovative proposals from untested researchers.
Unfortunately, in the face of the Trump administration’s attacks, any discussion of the need for change will likely be treated with defensiveness and skepticism—and existing experiments in the funding process will be put on hold—as the scientific community tries to keep core projects running and protect existing jobs. This is a major problem for a democratic society. A true new golden age for science will arise only when we restore mutual trust and openness between scientists and the public, and collaborate in realizing its future as a genuine public good.
Independent and nonprofit, Boston Review relies on reader funding. To support work like this, please donate here.
Zeynep Pamuk, Associate Professor in Contemporary Political Theory and Professorial Fellow of Nuffield College at the University of Oxford, is author of Politics and Expertise: How to Use Science in a Democratic Society.
Moderation used to help Democrats win, but its advantages now have been greatly exaggerated.
The problem isn’t immigration. It’s the failure of liberal democracy itself.
Speaking the truth and exposing lies is not enough.
Vital reading on politics, ideas, and culture to your inbox
A political and literary forum, independent and nonprofit since 1975
Registered 501(c)(3) organization
Help Us Stay Paywall-Free
Become a Member
© 2026 All Rights Reserved
Featuring Zeynep Pamuk and Michael Harris on science under fire; Omer Bartov and Olúfẹ́mi O. Táíwò on memory politics; China Miéville on his new novel; and much more.
Subscribe now