My younger colleagues have a hard time. They spend at least a decade in specialized graduate and postdoctoral study, during which they train to stay on task. There is minimal breadth of thought. Do the research, write the papers, win the grants.

In spite of its narrowness, the system has sustained decades of significant research. But now that system is breaking down, as its reward disappears: government research grants are becoming harder to obtain. National Institutes of Health funding has been stagnant since 2008 and, taking inflation into account, has declined by 20 percent since 2003. Despite NIH efforts to award independent grants to younger scientists, the average age at which researchers receive their first (called an RO1) remains stuck at forty-two. Many professors must be considered for tenure before this age, but an RO1 is generally a prerequisite. This ensures intense pressure in the early years of teaching and research. The system is good for neither education nor creative thinking. Nor does it encourage scientists to learn how to teach or communicate with the public.

This was not the case as World War II was drawing to a close. Wartime scientific research was highly productive, and President Roosevelt asked Vannevar Bush, director of the Office of Research and Development, how it might be transformed into a peace dividend. In his 1945 report “Science: The Endless Frontier,” Bush set down the principles that resulted, five years later, in the formation of the National Science Foundation: even in peacetime, the government should support basic research; it should continue military research, relying on civilian scientists working with the armed services; it should promote full employment by funding basic research that could lead to innovative businesses; and, finally, it should renew the nation’s scientific talent by funding students during the undergraduate and graduate years.

During the 1950s and ’60s, this democratic contract worked well. Voters agreed that their taxes were well spent funding a range of projects selected by experts. This research would eventually yield huge payoffs for the public in medical research, computers, and telecommunications. But as the postwar quiet exploded among demands for voting rights, the end of Jim Crow, equal rights for women, and an antiwar movement that nourished the idea of science for the people, some products of scientific research—napalm, defoliants, DDT, etc.—tarnished. Senator William Proxmire’s Golden Fleece Award (1975–1988) became an early and prominent symptom of and contributor to the fragmentation of consensus surrounding government research funding, particularly in the social sciences. Although Proxmire is long gone, others have taken up his mantle. Senate conservative Tom Coburn has successfully targeted NSF support for political science research, authoring a measure that allows funding for such projects only if the NSF’s director offers written assurance that they promote “national security or the economic interests of the United States.”

This dilemma has horns. In a democracy the public has the right to question how the government allocates spending. But scientific research is often arcane and seemingly far-removed from the urgent need to promote public health and welfare. This makes it easy for a Proxmire or a Coburn to take good work out of context and make fun of it. Furthermore, the choice of scientific projects always reflects political priorities. Coburn opposes research he believes supports a liberal view of government. Meanwhile, left-wingers disagree with other priorities, such as military research and studies concerning supposed biological bases of criminal behavior.

Our system is good neither for education nor creative thinking.

While we can’t return to the past, we can try to resolve the contradiction between popular government and top-down decision making through innovative mechanisms that engage citizens with science and science funding.

We could invite public participation in debates about social and scientific priorities, thereby encouraging scientists to respond to more diverse voices. Such experiments have been successful in the past. For example, there are Denmark’s citizen panels, established in the 1980s by the Danish Board of Technology to study technological problems and recommended solutions. Over the years their conferences have covered topics ranging from the challenges of 100 percent organic agriculture to the development and proper use of electronic medical records. Board activities sometimes lead to concrete political action. One panel discovered strong citizen opposition to a rule requiring that job applicants be blood-tested, resulting in a parliamentary ban. Closer to home, there was broad public participation in the establishment of the California Institute for Regenerative Medicine, founded in 2004 after citizens directly authorized $3 billion for stem cell research. Citizen panels can not only demand transparency and evaluate research options, they also can offer a counterweight to research funding directed by commercial interests and powerful advocacy groups.

Scientific research could also be done by non-specialist citizens themselves. Here we find precedents in popular citizen science projects, such as SETI at Home. Many of these have an expert scientist in the lead but harness the labor and computer power of laypeople to collect and analyze data.

Yet another approach to mobilizing citizen support for basic research involves recasting science in a more collaborative light. Many projects require international cooperation and a huge cast of characters. Research papers with hundreds of authors are not uncommon, but in the minds of policymakers and the general public, the model of the expert genius—the Stephen Hawkings of the world—has endured. In reality science today usually works through large, dispersed networks. These are not amenable to vast conspiracies of the sort feared, for example, by those who believe human-generated climate change is a hoax. If non-scientists better understood the process of producing scientific knowledge, some of the suspicions of government-funded science might abate.

And the public could be motivated to buy into scientific research through crowd-sourced starter grants. There are dozens of websites scientists use to raise money for pilot projects that can later morph into larger, government-funded efforts. Contributors on petridish.org, for example, have funded everything from tracking butterfly and whale migration to a study of the physics of tiny algal spores that grow into giant underwater plants.

But it is not enough to devise participatory forms of decision-making about funding. A more democratic science demands that we also rejuvenate education at the primary and secondary levels. Efforts are underway to implement the next generation of standards for science teaching, but full, nationwide compliance is years away. And even then it seems certain that science education will follow a sadly uneven path—stronger in wealthy, white school districts and weak to nonexistent in poor, rural, and heavily minority ones. Democracy, including democratically based support for science funding, works best with a well-educated public. In the long run, solving the problems of the NSF and the NIH cannot happen unless more money is available to elementary and secondary schools—for supplies, educator training, and salaries commensurate with the skill, hard work, and dedication displayed by so many teachers.

Research funding is a widespread concern among scientists, and not just in the United States. The British Nobel Laureate and cell biologist Paul Nurse has been particularly vocal. But he assigns the primary role for deciding research priorities to the scientists themselves. Not any old scientists, mind you—only the most talented and creative ones. In Nurse’s view this means the most effective reform is to screen for highly motivated, productive researchers. While he does want scientists to improve their communication with the hoi polloi, and his approach provides a needed antidote to legislative micromanaging of laboratory activities, he basically calls on the public to trust the experts. It is the Vannevar Bush program in modern dress, and it doesn’t work any more. We cannot emerge from our current funding drought without building a new, more democratic contract between scientists, science policymakers, and the taxpayers who finance it all.