Earlier this year, the Trump Administration issued a proposed “Regulation for Federal Financial Assistance” to make federal grantmaking “consistent with the public purpose of federal authorizing legislation and aligned with administration policies and priorities.” To that end, in the domain of scientific research, the 108-page regulation “clarif[ies] that peer review remains advisory and does not replace agency discretion.” The rule proposes to allow federal agencies to terminate a science grant if it “no longer effectuates program goals, federal agency priorities, or the national interest as they exist at the time of the termination.”

In short, this rule proposes to strengthen and clarify what we might call civilian control of science and other areas of federal expenditure. Not boards of peers (fellow scientists shielded by anonymity) but named and, in most cases, politically appointed and Senate-confirmed agency heads are to be responsible and accountable to the president and to Congress—and through them, to the public—for grantmaking. Programs are not to go on any longer than their political masters believe them to serve the national interest—or, at least, when they do go on, we will know who is to blame.

When I put it that way, you might wonder how anybody can be against this change, or at least, anybody who claims to believe in democracy and democratic accountability. But a lot of people believe, or at least claim to believe, that scientists not only know science better than politicians and voters but also know better what science is for, and how to balance its prospective achievements against its prospective costs and harms.

Of course, that too is false to the point of absurdity. No scientist is an expert in all the fields on which any potentially serious piece of work might impinge. Moreover, no scientist has even the beginning of a notion of how to show rigorously, outside his own tiny subfield, that public money is best spent on the projects he favors.

Decisions like this have to be made through institutions that are politically accountable. “That which affects all must be decided by all.” As President Eisenhower said in a less-often-quoted portion of his 1961 Farewell Address,

[I]n holding scientific research and discovery in respect, as we should, we must also be alert to the…danger that public policy could itself become the captive of a scientific-technological elite. It is the task of statesmanship to mold, to balance, and to integrate these and other forces, new and old, within the principles of our democratic system-ever aiming toward the supreme goals of our free society.

Those are theoretical considerations. But our experience points in the same direction.

Ever since an influential 1945 report by FDR science official Vannevar Bush, the standard policy has been to fund peer-reviewed theory and “basic research” more heavily than applied science. After 80 years, there has been no evidence that this approach produces more progress on fundamental questions. All actual revolutionary progress on such questions has come either in the course of exploring the fundamentals of applied research or from the “unfunded” curiosity of those willing to question the wisdom of the scientific peers. The empirically proven way to support such curiosity is not by awarding grants to research proposals but by paying those willing to speculate full-time salaries for half-time jobs teaching or calculating astronomical tables or examining patents.

Under present conditions, peer review is an engine for enforcing consensus rather than discovering truth. Why don’t we have more Manhattan Projects? Because they are too risky, too offensive, and too expensive for “peers.” Peer review rewards scientists who play to the existing “scientific consensus” and are most successful at selling minor tinkerings as possible breakthroughs.

Worse, this system is an (unfortunately not often literally) astronomical waste of public money. Instead of funneling billions into a bureaucracy that forces researchers to spend half their time begging for grants to fund bloated projects, we could achieve real progress for a fraction of the cost. We could end these grant programs entirely and redirect a sliver of that money to hire more researchers on full-time salaries for half-time teaching or administrative jobs—buying them the actual time and freedom to pursue unfunded, unstructured curiosity.

The most spectacularly successful technological projects today are Elon Musk’s reusable rockets, and public excitement over them has made Musk the world’s first trillionaire. Musk employs a great many experts, but he does not run SpaceX or any of his other companies by peer review. Nor, and more to the point, do the government agencies that serve as SpaceX’s customers run their programs by giving the experts the power of decision. Program officers and agency heads who work with SpaceX know that when Musk’s rockets blow up, he and his employees are going to have to answer their questions, and they in turn are going to have to answer Congress’s. 

Going forward, we need bigger, more beautiful, and, frankly, more dangerous science if meaningful progress is to continue. Political accountability for publicly funded scientific and technological research and development cannot guarantee us those benefits—the rockets can still blow up, or the viruses might again escape from the lab. Yet without accountability we will be stuck measuring excellence by the mere size of grants and the citation counts of researchers, rather than by results that actually make us healthier, richer, and freer.