Congress is heading into dangerous territory as it decides what basic scientific research should be.
“What were you people thinking?”
This was the opening salvo from a staffer on the House Science and Technology Committee, a hint of the battles to come. The year was 1997, and I was a program officer for the Political Science Program at the National Science Foundation.
The research that had attracted the ire of the committee investigated why credible candidates chose to run, or not run, for Congress. The NSF had peer-reviewed and funded this research, which answered a fundamental question about how potential candidates weighed entering into an electoral contest. These insights illustrated the importance of timing, money, and the stage of one’s life. The research also pointed out, not surprisingly, that women faced even bigger barriers to entry than men do.
The staffer was incredulous that the NSF would fund such a study. While I thought this work was an application of excellent social science methods, members of Congress thought that the NSF was funding researchers to recruit people to run against them. In particular, members of the Congressional Black Caucus thought they were being targeted, and demanded that the NSF order the researchers to turn over all of the names of people who had been sampled, even though the respondents had been guaranteed anonymity. It was clear that these lawmakers wanted to know who might mount a challenge.
Then, as now, congressional incumbents were re-elected over 90 percent of the time, and challengers were usually weak candidates. Nonetheless, members of Congress were running scared. This example underscored just how much lawmakers resist being the subject of study. For political scientists, however, this is difficult to avoid, since we often study those who exercise power.
In the end, the researchers stuck to their guns and did not provide the names sought by Congress. In retaliation, Congressional Black Caucus members successfully introduced an amendment to punish the Political Science program, cutting the program by the amount of the study’s funding to signal that this research should not have been supported.
This was neither the first nor the last time politics interfered with the NSF. It was created in 1950, in the midst of the Cold War, to provide national resources for basic research as the legislative and executive branches recognized that basic science was crucial for innovation and building scientific leadership. They also understood the NSF’s importance as an independent agency that would rise above politics. Although its director has always been a political appointee, its staff is made up of civil servants and temporary “rotators” drawn from the science community.
Unlike other agencies in the federal government with a research charge, the NSF has no laboratories, and it does not carry out its own research. Instead, its mandate is to identify and fund the very best basic science. Indeed, the NSF is regarded as the gold standard for funding basic research. It is envied and emulated across the world. Three features are responsible for this reputation: a culture of scientific independence, an extraordinary system of peer review, and the long tradition of political independence.
In 1996, I went to the NSF as a “rotator.” My employer, Rice University, granted me a two-year leave, and the federal government temporarily employed me. At the outset, I was sent to a boot camp for new program officers. The message was clear: I had been entrusted with taxpayer money and my job was to fund the best and most exciting science that I could. It was pounded into my head that I should take risks, avoid funding work that was taking only miniscule steps, and encourage work that would move science forward.
The NSF relies heavily on its peer review system. In the case of my program, proposals were sent out for review to independent scholars around the world. Accordingly, a substantial part of my job was to identify who would be the most appropriate reviewer, which meant deciding who was working in a similar field and doing cutting-edge research. It also meant finding researchers who were a bit removed from the subfield in question, but who could still comment on the general direction of the research. It did not mean asking for non-scientific opinions from pundits.
The proposals detailed the theoretical contribution, the explicit hypotheses to be tested, the research design, and the justification for their budget. Peer reviewers then judged the scientific merit of each proposal, weighed the intellectual merit, and judged the broader social contribution of the research. A typical proposal received five outside evaluations. In addition, the Political Science program convened a panel of eight scholars who came to the NSF to discuss all of the proposals that had come in during the previous six-month period. The panel discussion covered each proposal, discussed the reviews, and provided the independent judgment of the panelists. It followed up by offering advice about how to rank the proposals for funding. There were always far more proposals than dollars to fund those proposals, meaning that my job was to take into account all the advice I had been given, and then decide how to allocate taxpayer money. The mantra at the NSF remained: fund the very best basic research.
We also made sure the NSF remained above politics. We were told to focus on what we felt was important for our scientific communities. This did not change with the fact that the Political Science Program touches on politics – it is the object of our study. But that did not make us politicized. When I was called to Congress in 1997, amid the storm over our study on candidate entry, the NSF director at the time, Neil Lane, called me to his office to ask me a simple question: “Is the study you funded high quality science?” My answer was simple and unequivocal – it was very high quality, and it answered a key question about political choices in a novel and innovative manner. He said he agreed, and promised that the program would be fully backed.
This mandate notwithstanding, Congress has often attacked individual projects at the NSF. The “Golden Fleece” awards handed out by the late Sen. William Proxmire (D-WI) poked fun at grants that seemed silly on their face. But he selected these individual grants for shock appeal, and neither he nor his budget-minded successors ever went after an entire discipline.
These days, however, Congress is wielding a far bigger anti-science axe. Lawmakers are now going after entire programs at the NSF, with political science on the chopping block. In 2009, Sen. Tom Coburn (R-OK) brought to the floor an amendment to a spending bill that was designed to eliminate completely the Political Science program from the NSF. This was a new tactic. Coburn argued that in times of economic distress, the study of politics was a luxury that the country could not afford. He conceded that eliminating the Political Science program was a drop in the bucket, but claimed that the money would add up as long as Congress continued to chip away at small things. Coburn also explained that he was not against political science as a discipline. “If there ought to be any political science study done, it is, why are members of Congress such cowards? That is the thing we ought to study. We ought to study why we refuse to do the right thing because it puts our job at risk.”
That amendment was defeated, but Coburn remained undeterred. The following year, Sen. John McCain (R-AZ) joined him to release a report titled “Summertime Blues,” a compendium of 100 stimulus projects that supposedly gave taxpayers “the blues.” A number of the highlighted and ridiculed projects were NSF funded – my NSF funded project was number 45 on the list.
In 2012, Rep. Jeff Flake (R-AZ) joined Coburn’s cause, offering an amendment to a House spending bill to cut almost $1.2 billion from the NSF. This amendment was defeated, but another to eliminate the Political Science program at the NSF followed. During the debate, Flake said that the NSF should not subsidize such work, even though he recognized political science’s value. This time, his amendment passed the House, largely along party lines, but the Democratic Senate declined to take up the House bill, thereby averting the elimination of the program.
Coburn scored a genuine victory this past spring as Congress considered a spending bill to fund the federal government through September 30. As the Senate considered the bill, Coburn first introduced an amendment that would eliminate the Political Science program at the NSF, arguing that the discipline does not save lives. When he saw that his amendment was not likely to win support, he rewrote it so that it would de-fund all political science research except that which promotes “national security or the economic interests of the United States.” This time, the amendment was adopted. Exhausted by its long string of fiscal battles, Congress quickly passed the full spending bill, and President Obama signed it.
At first glance, it appears that the program was saved. The language constraining research seems minimal, and the limits on what kind of science can be done apply only to the current fiscal year. However, the amendment requires that the Director of the NSF certify that every political science project adheres to the Coburn amendment’s language; the NSF has already followed through by putting such a mechanism in place. Furthermore, this language is expected to stay intact in the spending bill for the next fiscal year, which Congress is likely to pass through another continuing resolution this fall. The amendment’s success has emboldened others in Congress to place additional constraints on which sciences are legitimate, and what their scope will be.
The Coburn amendment has already produced unintended effects. The first casualty is the Ralph Bunche Summer Institute at Duke University, which lost its NSF funding this year because it does not meet the standards set by the legislation. The Institute, named after the 1950 Nobel Peace Prize winner, Ralph J. Bunche, encourages diversity students to pursue careers in political science. The program takes promising college sophomores and juniors and puts them through an intensive five-week summer course that exposes them to the scientific enterprise. It asks them to formulate an interesting scientific question and provides them with the tools to carry out their own research. The political science faculty mentor students on research, graduate school applications and navigating professional norms. The Institute has become an important gateway for diversity students entering political science.
Highly talented diversity students have enticing options other than political science. Many of these students choose professional schools instead, law school in particular. The choice is obvious: the lure of a high income and the chance to settle into the upper middle class. Students often make that choice because they have no experience with other options. The Bunche Summer Institute has shown them what these other options could be. More broadly, getting diversity students into the academe is critical. We study phenomena with which we are familiar, and our personal perspectives lead us to ask questions in our own ways. If diversity students opt out of the academe, we lose their critical input. Given the demographic shifts in the U.S., this is not a sustainable scientific path. While the Congressional Black Caucus led the charge against the Political Science program at NSF in 1997, it is difficult to believe that it would now support dismantling an institution that has been so valuable in providing a different perspective in political science.
Congress has a clear duty to oversee taxpayer expenditures, and basic research may indeed be a luxury item in a thinly stretched federal budget. But curtailing basic research because the government cannot afford it is a different matter than Congress dictating what accounts for “acceptable” science. The former is a budget issue, whereas the latter is an attack on the social sciences – and even on basic science as a principle. Basic research looks at the fundamentals of science, and it is difficult to predict its payoff. But we need to understand basic research as a public good, and therefore deserving federal support. This is all the more important today as strapped businesses eliminate their own research and development laboratories and rely on university researchers to carry out their work. Such work usually does not center on basic research, but on applied work designed to develop promising products. In contrast, everyone can share in the benefits of basic research, so individual firms have little interest in subsidizing such work.
Since the 1950s, Congress has funded basic research from the belief that the positive externalities are worth it even though they are difficult to predict. Transistors, computers and the Internet have all transformed the world economy. Who knew that carbon “buckey-balls” would spur the nanotech revolution?
Many members of Congress still understand the importance of basic research. But more and more are questioning which sciences are legitimate, and the social sciences are a tempting target. It is difficult to predict spinoffs from basic social science research. After all, the social sciences do not produce patents; they seldom produce widgets; and the object of study is not about manipulating the physical world. Instead, social scientists study humans and their interactions. This has prompted lawmakers to question whether the social sciences are even a science. They assume that their experience gives them a “commonsense” understanding of the political world that is far more astute than the analysis of political scientists. In this view, pundits and polls are all it takes to elaborate on what everyone commonly knows.
The hallmark of science, however, is the systematic and rigorous investigation of phenomena. Social scientists observe patterns in the world, formulate theories that explain the underlying mechanisms for those patterns, and then proceed to systematically test those theories. The theories come in all varieties, ranging from detailed descriptions of a social phenomenon to abstract mathematical models that strip away the rich context of the social environment to focus on a specific mechanism. These studies show that many “commonsense” understandings of the social world are, in fact, incorrect. Yet lawmakers still continue to pursue policy solutions based on opinion rather than fact.
Aside from systematic study, the social sciences help policymakers understand the intersection of institutions, culture and human behavior. For example, most scientists acknowledge that climate change is a serious global problem. The technical sources of change are well known: rising levels of carbon dioxide are a proximate cause. Yet the source of greenhouse gas emissions is largely human. Getting political entities to clamp down on emissions (say, by taxing carbon) or getting individuals to forego comfort (say, by dialing up one’s air conditioning) is a human challenge. Understanding the incentives for behavior is in the realm of social science. Reducing greenhouse gases is not just a technical problem; it is also a human problem.
A quick perusal of NSF awards shows the depth of what social scientists study. For example, considerable resources have been invested in understanding social networks. By understanding the links between people, we can shed new light on how the diffusion of information affects the way parents seek vaccinations for their children, how infectious diseases spread, or how natural resources are sustained. Other resources have been devoted to deciphering the neural basis of social cooperation. A common human trait is to be attentive to those around us, which causes us to constantly seek cues from others about when to trust people, or when to be wary.
More broadly, research investments by the NSF in optimizing resources have had enormous economic payoffs. Basic research on auction markets led to the Federal Communications Commission to adopt specific mechanisms for licensing the airwaves and netting the government more than $60 billion in revenues. Work by the Nobel Prize-winning economist Al Roth on how to match newly minted doctors to hospitals has led to real-time matching of kidney donors and recipients. Another good example is the NSF’s long-time support of data collection that provides critical information for social scientists to test new hypotheses. The Panel Study of Income Dynamics at the University of Michigan, which has long received NSF funds, has been critical in building our understanding of the effects of economic shocks on the American workforce. The American National Election Survey, which has been conducted since 1948, provides valuable insights into changes in public opinion, public attitudes toward government, and aspects of public participation in governance. These and other databases that the NSF has funded are used by thousands of scholars.
While the contributions of social science are many, the future of basic research funding for the social sciences is increasingly threatened. The Coburn amendment is just a start, as some of his like-minded colleagues launch two prongs of attack. The first is blatant and seeks to eliminate the social sciences from federal funding. The second prong of attack is far more pernicious: the attempt to require science to serve a narrowly defined mission. And this argument is gaining ground. The chairman of the House Science Committee, Lamar Smith (R-TX), has recently proposed changing the mandate of the NSF. At present, all applicants for research funds must make the case for their project’s intellectual merit and broader impact on society. Smith is proposing two more criteria. First, all research must be “in the interests of the United States to advance the national health, prosperity, or welfare, and to secure the national defense.” Second, it must solve “problems that are of the utmost importance to society at large.” His legislation has met with a good deal of resistance from the Democratic minority on the committee, but the proposal is still in play – and it could be revived in later spending and authorization bills.
Smith argues that his proposal would ensure accountability over NSF expenditures, but it affects the entire scientific community in unintended ways. By focusing on solving (undefined) problems of “utmost importance” to society, Congress would transform the NSF from an agency funding basic research to one of applied research. It also means that it is up to lawmakers to determine the crucial problems facing society. With Congress’ approval rating hovering around 10 percent and “must-pass” bills stalled amid gridlock, it is hard to imagine the public would agree that it has a respectable track record on addressing the core problems facing the country.
A very real likelihood is that the Smith proposal, if enacted, would eliminate much of the basic research in the physical sciences. It is difficult to see how the contribution of pure mathematics advances social needs. Understanding the evolutionary basis for why certain slime mold cells “cooperate” with each other does little to advance the interests of the United States. High-energy physics seeking to understand “dark matter” does little to advance national prosperity. The list could go on, but the point is simple: basic research will not be valued. This may not be the intended goal of Congress. But those of us who study complex systems understand that unintended consequences can often be pernicious.
Some scientists may view these attacks as minor matters. After all, the focus has been with a small program at NSF, and many in the natural sciences may believe that the study of politics cannot be scientific. But the larger scientific community should not ignore the shackling of one program at the NSF. If politics dictates what is worth studying, all disciplines are at risk. If politicians decide that they can judge the merit of cutting-edge research, then the peer review system is at risk. Why stop at political science, when the entire NSF Directorate of Social, Behavioral and Economic Sciences could be eliminated? Why stop there if biology continues to insist on using evolutionary models? The challenge to science is clear. If politics inserts itself into science, we must ask ourselves whether any of our fields survive — and who will be the next target.