In mid-August 2016, as America was gripped by the wildest presidential campaign in memory, an essay in an obscure, conservative-leaning science and technology magazine called The New Atlantis jolted an already unsettled scientific community. The author, science policy scholar Daniel Sarewitz, dispensed with near-universally held pieties — that scientists’ work is inherently noble, that basic research necessarily benefits society. He called science “a failing institution” whose values reflect “a world that no longer exists.” And he proposed a radical reconstruction of the scientific funding system: Rather than set scientists loose to follow their curiosity, it would instead yoke research funding to industry and societal needs.
The piece went viral in a way that few works of science policy ever have.
A few months later, on the night Donald Trump was elected president, Sarewitz, along with many others in America, was watching in shock. The consequences for science — like the consequences for everything else — were unknowable, though they seemed ominous. Trump had denied long-established science on climate change and vaccines, and had called the country’s medical research agency, widely admired by both Democrats and Republicans, “terrible.” Within hours of his victory, Trump would be labeled “the first anti-science president” by veteran science policy advocate Michael Lubell.
Sarewitz got an email from an editor at Nature asking him to write a column. A shaken Sarewitz responded that he didn’t think he had the heart to do it. But after some cajoling, he agreed. His essay was a far cry from his polemic of a few months earlier, without the far-reaching proposals to overhaul the U.S. scientific funding system. It suggested instead that the president-elect champion programs such as Manufacturing USA — an Obama program aimed at helping industry capitalize on scientific advances — to better extend science’s benefits to disaffected working class voters.
The pivot reflects a careful balancing act that Sarewitz, the Washington, D.C.-based co-director of Arizona State University’s Consortium for Science, Policy, and Outcomes, is trying to pull off: putting his ideas in front of sympathetic power players without giving fodder to a president who seems as likely to destroy science as to reform it.
While science advocates and politicians on both sides of the aisle extol “basic research” — the kind where you give a scientist a grant and tell them to follow their curiosity, with the expectation that societal benefits will come out the other end — Sarewitz has taken the opposite tack. He has championed science programs — often small ones tucked into corners of federal agencies — that direct research toward specific goals set by non-scientists. He has also questioned, in increasingly sharp terms, how much the U.S.’s basic research enterprise, which puts around $40 billion a year under scientists’ control, is giving us in return. And he has found the most receptive ears among Republicans, the political party increasingly viewed as the enemy by much of the U.S. science establishment. Republican lawmakers have invited him to testify on Capitol Hill and cited his work to support their legislative agendas.
“The idea that as long as you give scientists enough money to do all the undirected curiosity-driven basic research they want, we will be able to solve all the problems we need to solve,” Sarewitz says, “it’s just bullshit.”
That kind of tough talk has made Sarewitz, long a Washington insider, an increasingly visible crusader for scientific reform. “Dan is right at the belly of the beast,” says Michael Crow, president of Arizona State University, who hired Sarewitz and champions his work, “questioning the logic of the policy design.”
For some, Sarewitz’s message is long overdue. “I think he’s very brave,” says Susan Fitzpatrick, president of the James S. McDonnell Foundation, which funds human cognition-related research. But to others, Sarewitz is putting forward precisely the wrong message at the worst possible time, and even sympathizers worry that Sarewitz’s arguments could backfire. They “could be used by people who are looking to just make cuts,” says Arati Prabhakar, a fellow at Stanford University who ran the Defense Advanced Research Projects Agency during the Obama administration.
By that view, and with policymakers in Congress and the executive now increasingly questioning government support of science — and even well-established scientific truths — some critics see Sarewitz’s message as something akin to an accelerant being poured on an anti-science fire that had been smoldering for years.
“We’re living in an era in which we have a large fraction of the public that is extremely distrustful of institutions, the government being very prominent in that area,” says Lubell. “Having somebody who basically says the government is wasting your money, the scientific community is a bunch of money grubbers, they’re out to line their own pockets — that can have a very serious resonance.”
How can governments best make science serve the public good? It’s a question that has vexed policymakers since the beginnings of scientific research. In the mid-1800s, some members of Congress pushed for a major government investment in agricultural science, while others worried about meddling in the economy. The boosters won and Congress passed a law granting states land to establish agricultural universities, which hired scientists to study farming practices and extension agents to educate farmers. U.S. agricultural product soared.
The debate flared up again around the end of World War II. Physicists had persuaded the government to make massive investments in developing the atomic bomb, which ended the war quickly and decisively in 1945 after the bombings of Hiroshima and Nagasaki in Japan. But should the government continue to spend public money to keep scientists employed in peacetime? Again, the science boosters won the day, when Vannevar Bush — an influential government scientist and administrator of the Manhattan Project, the research and development program that produced the bomb — convinced a skeptical Harry Truman to create a peacetime agency to fund basic research, the National Science Foundation.
Bush proposed an ultimate $120 million annual budget (just over $1.6 billion in today’s dollars) for medical research, natural sciences, defense, and education. The National Science Foundation launched in the early 1950s with a meager budget of $3.5 million, but after the Sputnik launch it grew rapidly, and now gets around $7.5 billion a year. Federally funded research helped create science-based technologies — the laser, new cancer drugs, MRIs, the internet — that supercharged the economy and transformed American life.
Sarewitz entered the scene in the late 1980s, having abandoned academic research for the chance to shape national policy as a staff member for the U.S. House of Representatives’ Committee on Science, Space, and Technology, which guides Congress’s investments in the science funding agencies. One of his first assignments was to write an op-ed for a California newspaper after the 1989 “World Series” earthquake devastated the San Francisco area. “I just remember thinking how much more gratifying it is to spend half a day writing an op-ed that tens of thousands of people will read, and that might help people think about issues in a new way,” he recalls, “than write academic publications for journals that are never read by anybody.”
Sarewitz was shocked, however, when he discovered that political calculations largely drove billion-dollar funding decisions about science policy. When it came time to push for increased earthquake research funding, for example, George Brown Jr., a Democratic representative from California and the committee chair (and Sarewitz’s boss), suggested a figure to add to the existing budget. Sarewitz asked where Brown got the number. “That’s what I think we can get,” Sarewitz recalled Brown saying.
Ever since Vannevar Bush, advocates have argued for unleashing science’s potential by setting inquisitive scientists free with piles of public cash and minimal rules. Such research has indeed borne fruit: Blockbuster drugs and compounds have emerged from scientists’ obsessions with obscure microbes living in soil or hydrothermal vents. Google began with an NSF grant. All told, Harvard physicist and public policy scholar Cherry Murray wrote recently, some 85 percent of the U.S.’s economic growth is built on the fruits of research and development. Other times, however, basic research has not delivered promised results. The more than $100 billion spent on the War on Cancer have yielded mixed results, with advances against some kinds of cancers and few or no new effective treatments for others. Other research fads — gallium-arsenide semiconductors, graphene, nuclear fusion — have consumed millions of dollars and generated thousands of journal papers, but so far have resulted in little real-world payoff.
There have been efforts to figure out what makes some research more productive and other research less. In the mid 2000s, George W. Bush’s science advisor John Marburger, after speaking with Sarewitz, convinced the National Science Foundation to create a program to fund research on improving science policy to try to answer such questions. Expert panels such as those convened at a 2011 National Academies workshop have also debated how to measure research’s impact.
But it can be hard to see the weaknesses of a system when you’re on the inside. Sarewitz has become the consummate outsider, but he still knows all the insiders. In 2009, he took over an opinion column in the high-profile journal Nature. In columns, Sarewitz eschews academic obfuscation and conservative language for the clear and combative prose of a polemicist. That pure research leads to societal goods is a “simplistic and self-serving myth,” he wrote in a 2013 column. In another piece from that year, he excoriated scientists for aligning themselves with the political left: “… [S]cience is revealing itself, like the unions, the civil service, environmentalists and tort lawyers, to be a Democratic interest,” he wrote, “not a democratic one.”
“He cuts right to the core,” says Roger Pielke, Jr., a science policy scholar at the University of Colorado Boulder who worked with Sarewitz on the Hill.
Scientists have not always appreciated Sarewitz’s critiques. He says he has received critical emails; online commenters accuse him of misunderstanding science and fatally oversimplifying complex issues. Gavin Schmidt, director of NASA’s Goddard Institute for Space Studies, took to Twitter to provide a rejoinder to a 2015 column in which Sarewitz accused scientists of interpreting research results in ways that support their political ideologies:
like most sarewitz articles:
– valid issue
– interesting point
– overreaching conclusion
– blame scientists
— Gavin Schmidt (@ClimateOfGavin) September 9, 2015
As his name got out, Sarewitz began being courted by members of Congress. In 2010, he testified at a House subcommittee hearing that the government needed better ways to measure what it was getting for its investment in science. Thereafter, though, his plaudits and invitations in Washington came increasingly not from the Democrats he once worked with — and whose party he still belongs to — but from Republicans who felt emboldened to question Congressional support of science agencies.
In 2011, Senator Tom Coburn, a conservative Oklahoma Republican, cited a Sarewitz column in a report on what he saw as National Science Foundation’s shortcomings, taking specific aim at its social science program, which funds research on human behavior and how socioeconomic, political, and cultural forces influence society (including some of Sarewitz’s own research). Sarewitz quibbled with Coburn’s report while celebrating his appearance in it in a blog post on Pielke’s website.
In 2013, with the House controlled by Republicans, Sarewitz testified at a science committee hearing on the artfully — and perhaps presciently — then-named Keeping America FIRST Act (“FIRST” standing for Federal Investments in Research, Science, and Technology). The bill, pushed by committee Republicans, directed NSF to create a process for determining whether each of the 10,000-plus grants it awards annually is in the “national interest.”
Science advocates were warning that the legislation would drain NSF officers’ time and could open the door to Congressional meddling in the agency’s grant review process, which had always been managed by scientists. Sarewitz viewed this as “hysteria.” His opinion? The NSF claims to advance science for the public good, so what was wrong with asking the agency to certify that its grants do what it says they do? He told the committee that the proposal was a “meaningless level of rubber stamping” that nearly any grant would easily earn, and he urged them to come up with accountability rules that had teeth.
“I’m not afraid of working with the ‘other side,’” he says. “I think the Republicans are more open to this stuff right now, because the Democrats just see themselves as playing defense.”
Sarewitz’s willingness to publicly question the value of publicly funded research struck some advocates as naïve, if not just reckless. “Dan’s got many legitimate things to say,” says Dahlia Sokolov, a Democratic staff director on the House science committee who opposed the act. But she thinks his devil-may-care argumentation style has needlessly given fodder to powerful anti-science forces. “His approach is to stir the pot by playing devil’s advocate.”
In the end, the scientific establishment won that battle. The bill never made it out of committee.
Sarewitz has gained notoriety as an incendiary polemicist, but his swashbuckling authorial persona belies a self-effacing, professorial, almost shy man, sporting a beard and thinning brown hair. When asked about specific impacts he’s had, Sarewitz demurs. No outcome of a complex policy discussion can be traced to any one person’s input, he protests.
Still, Washington is paying attention to critiques such as his. In 2006, the National Institutes of Health created a program to award grants for studies specifically designed to produce new drugs and therapies, not just a better understanding of disease
Lawrence Friedl, director of NASA’s Applied Sciences Program, was inspired by a 2007 paper Sarewitz and Pielke wrote on the supply of and demand for science — in other words, applying free-market thinking to science. Friedl is working to transform NASA’s internal culture by involving potential users of satellite data in the mission design process. About a year ago Friedl invited Sarewitz to join the program’s advisory committee. “I want to do more with Dan,” Friedl says.
For Sarewitz, this slow, laborious work inside bureaucracies is necessary but not sufficient to create the change he wants to see. So when an editor at The New Atlantis invited him to write a long think piece about the problems facing science, he jumped at the opportunity. It was an unusual venue for a science policy wonk. The magazine was founded by a small, conservative think tank with a religion-based agenda; even the magazine’s name, hearkening to a supposed lost continent, evokes pseudoscience more than science.
“Science,” Sarewitz wrote by way of introduction, “isn’t self-correcting; it’s self-destructing.” The core problem, he diagnosed, is a funding and incentive system that, instead of directing scientists to solve real-world problems, induces them to compete mercilessly for grants, chase fads such as the current obsession with big data, and publish so many papers that they have no time to think about how research can meet societal needs — or even learn what those needs are.
High-profile studies and books over the past decade have documented a “reproducibility crisis” in biomedical science and psychology — some estimate that up to half of all biomedical research produces results that cannot be replicated in other labs. Sarewitz added his own diagnosis for why these fields have gone so far off the rails: Because scientists must design their research to get published in journals rather than solve real-world problems, there is often no reality check on their results.
Sarewitz believes funding agencies should take cues from two institutions whose mere mention grates on the ears of many scientists — industry and the Department of Defense. These arenas, where research is strictly circumscribed by consumer wants or military needs, have actually generated a lot of scientific successes, Sarewitz argues. These include the transistor, which was developed at the industrial research facility Bell Labs, and which now populates our computers and smartphones by the billions. Sarewitz also lauds a cancer research program funded by the defense department with an explicit mission to cure breast cancer, rather than simply produce more knowledge about the disease. “When scientists are more aware of the context of use, knowledge advances in ways that are more useful,” he says.
“Saving Science” electrified the science policy community. Those who share Sarewitz’s jaundiced view were delighted to hear the voice of a kindred spirit. For others, the essay distilled and amplified what they see as Sarewitz’s weaknesses — anecdotes often favored over data, sweeping statements presented with scant supporting evidence. “While most of the evidence of poor scientific quality is coming from fields related to health, biomedicine, and psychology, the problems are likely to be as bad or worse in many other research areas,” Sarewitz wrote, writing off much of physics, chemistry and many other disciplines with a few keystrokes. Elsewhere he asserted that the scientific system “each year generates 25,000 promising new Ph.D. scientists and nearly 2 million new articles of mostly dubious value.”
“He basically believes that other than the defense department, federally supported science has not yielded very much. And that’s absolutely false,” says Lubell, former director of public affairs at the American Physical Society. “Is [the system] perfect? No. Are there abuses, is there waste? Yes. But how do you do it?”
Some fellow academics were also unimpressed. Sheila Jasanoff, a science and technology studies professor at Harvard, says Sarewitz ignored scholarship that should have added nuance to his black-and-white argument. She calls the piece “a polemic … a kind of primal scream at the universe.”
Primal screams do have an advantage: They can be heard when ordinary speech gets drowned out. Prabhakar had never heard of Sarewitz until “Saving Science” landed on her desk at DARPA. “He’s doing a service by reminding our community that we’re doing this for society,” she says.
Sarewitz also teamed up with University of British Columbia physicist Lorne Whitehead and University of Maryland computer scientist Ben Shneiderman and pitched a proposal for “highly integrative basic and responsive” research to the Association of Public and Land-grant Universities, a Washington-based group representing the same universities that long ago turbocharged America’s farms. Thirty research universities are now engaged in the effort, says Howard Gobstein, APLU’s executive vice president. It’s a long way from reinventing the system, Sarewitz admits, but it’s a start.
Of course, over the next few years, nobody is likely to have more influence on American science than Donald Trump, a president who has shown little interest in it.
After Sarewitz’s post-election article championing programs like Manufacturing USA, Trump’s budget proposal went after that program and other applied science initiatives with particular vigor. Though the document provided scant justification for the cuts, Trump’s budget director Mick Mulvaney has cited a concept that had been floating around conservative circles for years — that government should get out of applied science to avoid competing with industry. That’s diametrically opposite Sarewitz’s view that government should be playing a stronger role in connecting scientists to industry and other users of research. “It’s bizarre, given the president’s concerns about economic opportunity in areas that have had their manufacturing base wrecked, that he wouldn’t support those kinds of programs,” Sarewitz says.
But even if Trump’s lack of interest in science is insurmountable (a recent op-ed by a prominent geneticist noted that none of Trump’s nearly 4,700 tweets since becoming president mention “science” or “technology”), other powerful players in Washington are still listening.
This past fall, a representative from the libertarian Cato Institute cited the line “It’s technology that keeps science honest” from “Saving Science” at a Senate hearing, arguing that industry would do basic research better if the government just got out of the way. Rand Paul, a Republican Senator from Kentucky who was sponsoring a bill that included NSF reforms, and who often argues for smaller government, took a keen interest in the quote, according to a news report.
When I sent Sarewitz the report, he replied, “Yikes, I had no idea!”
Sarewitz does not endorse industry taking over basic research. He does, however, think business should be at the table when research projects are conceived, not just years later when research results are presented. And in Washington, where opportunities for major change are few, you often take what bedfellows you can find.
“I’m utterly open to finding receptors for these ideas,” he says.
He’s now considering trying to get a meeting with Paul.
Gabriel Popkin writes about science and the environment from Mount Rainier, Maryland, just outside the nation’s capital. His work has appeared in a variety of general-interest publications, including The New York Times, Nature, Science, The Washington Post, Scientific American, and National Geographic, among others.