Social Science for What? Battles over Public Funding for the ‘Other Sciences’ at the National Science Foundation

Responsible innovation (RI) is one of the most recent movements to advocate that science, engineering and technology develop in ways that better reflect and respond to diverse societal values and concerns (Owen, Macnaghten, and Stilgoe 2012; Foley, Bernstein, and Wiek 2016; European Union 2014). Shanley (2021) traces a history of movements that led to RI beginning in the mid-twentieth century, including technology assessment, science, technology and society (STS), and appropriate technology. Recently, the new Biden presidential administration in the U.S. announced that its Office of Science and Technology Policy (OSTP) would include a new Deputy Director for Science and Society (Thomas and Ambrose 2021). The nature of this role remains to be seen, but it holds the potential to encourage deeper and broader reflection on the societal outcomes of research and development.

Those interested in the U.S. federal agencies seriously assessing the societal outcomes of science and engineering must read ‘Social Science for What?¹‘ In it, the historian Mark Solovey provides a thoroughly researched and fascinating history of how managers at the National Science Foundation (NSF) overcame scientific and political opposition to create the NSF’s social science research programs that exist today. The book well captures the skeptical impressions that NSF’s early scientific and political leaders had of the social sciences, much of which also still exist today. Solovey masterfully shows how a combination of NSF ‘scholar bureaucrats,’ armed with arguments about the nature of science, and outside political pressures from social scientists themselves justified and established today’s NSF social science research programs. As RI is arguably an emerging social science, the challenges that NSF’s social science programs faced in becoming institutionally and politically established may well apply to future RI research and efforts to better consider ‘science and society’ within government.

The social sciences as a unit of analysis can be quite challenging, as they cover an incredibly diverse range of research on humans and society. Solovey’s task is made partially easier because he focuses on social science activities that were actively funded by the NSF. The early NSF’s 1950s research programs included classical and foundational social science disciplines, namely anthropology, sociology, economics, political science, and the history and philosophy of science. Solovey situates social science within contemporary 1940s debates about the nature of science. He describes a ‘scientistic’ view, where NSF leaders and prominent scientists believed in what he calls a ‘unity’ of social and natural science, but also argued—as Vannevar Bush did in his 1945-6 debate with Senator Harley Kilgore who wished to integrate the social sciences into NSF—that the social sciences were immature, and needed to emulate the physical sciences before gaining funding. Bush mostly won this debate. The NSF charter primarily enabled the physical and life sciences, and called for the support of ‘other sciences,’ notably not calling the social sciences by name. Many, especially physical scientists but also including some social scientists, still agree with Bush’s assessment of the social sciences today.

Solovey shows how NSF cultivated more quantified approaches for these ‘other sciences’ to be seen as legitimate and mature, such as funding research activities on ‘mathematical theory of economic models, … multiple factor analysis, suicide rates, [and] social psychological experiments on cognitive dissonance’ (58). Certain types of social science, such as cognitive psychology and quantitative approaches in economics, more easily fit the narratives of quantification from the natural sciences, and they gradually received more funding within the agency budget. Such an approach, called ‘underdogging,’ was the easiest way to develop, defend and grow the NSF programs, and is certainly relevant today (56).

Solovey traces present-day Congressional efforts to change NSF’s social science programs to NSF’s early controversies over qualitative social science. The perceived and at times self-professed immaturity of the social sciences led some to believe they were more ideological than scientific. McCarthyist ‘red scare’ attacks in the 1950s made similar and unsupported claims, and Solovey quotes the sociologist Louis Worth: ‘Since the problems with which the social scientist deals are characteristically controversial issues among the rival groups comprising society, social science runs the risk of being considered ‘dangerous thought,’ and the social scientist a ‘subversive’ character’ (54). Champions of the social sciences within NSF lobbied the agency internally to let private funders of social science research (such as the Ford Foundation and others) take the risk of tackling more potentially controversial topics such as ‘sex, religion, race, [and] politics’ (68). Solovey shows that critics of NSF from the 1980s to 2010s similarly argue for a narrowed conception of research for the public good, with discussions on classic events such as Senator William Proxmire’s ‘golden fleece’ awards for all kinds of science projects deemed wasteful and irrelevant.

The formidable cultural challenges of getting physical scientist leaders at NSF to embrace social science were summarized by NSF staffer ‘Bel’ Rubenstein: ‘Sisyphus must have had [an] easier task’ (p.64) than those who supported government social science, as they continually had to defend and articulate their intellectual foundations. Solovey strongly summarizes how some early and present-day social scientists argued that social sciences are fundamentally different from natural sciences, in part due to the complexity of making claims about people’s intentions and beliefs, which are not directly examinable by others. Solovey chronicles a long history of social scientists and NSF insiders defending qualitative research as a distinct and worthwhile endeavor. While qualitative social scientists did have strong internal disagreements about the societal value of the social sciences, they were largely united in their attempts to convince varyingly sympathetic NSF officials the necessity of qualitative descriptors to capture the complexity of people and how they engage in the world—phenomena that could not be definitively reduced to quantitative descriptions alone. In the 1970s, major critics of NSF’s quantitative social science focus voiced their dissatisfaction, with anthropologists using the work of Clifford Geertz as an example of excellent qualitative research that would not meet the informal quantitative research standards of the time. NSF’s methodology for research awards became part of the debate. Critics argued that peer reviewers chosen by NSF to assess research were biased towards quantitative projects. Today, NSF’s qualitative peer reviews are now used to justify the rigor of NSF’s qualitative social science research.

NSF’s embrace of more qualitative areas of social science research was shaped by direct political pressures as well, and Solovey provides useful detail on the discipline of political science in particular. While some political science research falls into the qualitative areas that ‘hard’ scientists sought to avoid, an influx of political science-trained individuals who entered Congress, such as Hubert Humphrey, encouraged additional research in this area. In the 1970s, the head of the American Political Science association also began an effort to write members of Congress, hoping to pressure NSF leaders to support more qualitative political science work (96). More broadly, as the Reagan administration proposed major reductions to social science funding in the 1980s, Solovey notes how social scientists banded together through the Consortium of Social Science Agencies (COSSA) to help lobby the hill and to defend the social science budget. This helps underscore that the science policy process is often indistinguishable from the budget process, which necessarily involves normative, political judgments about priority (Sarewitz 2007),² and concerted strategic action by decision makers in multiple parts of government. Efforts to institutionalize responsible innovation require funding and as such require careful attention to the politics of budgeting, as the physical and natural science communities already do.

Solovey concludes the book with considerations on how future social science research could be used to help contribute to the public good, and he embraces a provocative proposal from the late 1960s. Senator Fred Harris (D-OK) proposed a National Social Science Foundation (NSSF), a new agency, separate from NSF, to focus on social sciences (Harris 1967). Harris envisioned this as allowing social sciences to be supported in ways that did not force them to follow the conventions of the NSF, and to enable more innovative thinking on societal problems. Harris appreciated the increased complexity of an evolving urban population facing massive inequities, in ways that may resonate with responsible innovation scholars today. The NSSF was envisioned as being able to advise other mission-focused federal agencies. Solovey convincingly argues that NSSF had a significant amount of support, both from scientists as well as over 20 Senators who co-authored a proposal for it, and shows it primarily lost traction when the more hostile Nixon administration took power in 1969 (127).

The proposed NSSF may serve as an important historical reference point for recent efforts to resurrect the U.S. legislative branch’s now defunct Office of Technology Assessment (OTA). While OTA used both natural and social scientists to assess policy options surrounding upcoming technologies, it conducted such studies as one-off assessments requested by legislators. Many scholars look at the practice and theory of ‘technology assessment’ as another framework to support ethical outcomes from engineering and science (Graves and Cook-Deegan 2019). While not assessing the OTA proposal himself, Solovey seems right that the proposed NSSF could more directly support decision making within executive branch agencies than NSF’s social sciences programs currently do. Further, we could imagine that an NSSF could go beyond the OTA’s proposed decision support role, by improving the fundamentals of social science disciplines themselves, and in a way that would not be marred by conceptions about the primacy of natural science among the social sciences.

These debates raise a deep question about which organizational banner should be used to push to institutionalize efforts to get more responsible science and engineering research and practice. Solovey’s focus on the collected social sciences as a unit of analysis evokes a few existential concerns on this score. As he notes, some areas of social science, such as economics, embrace a quantification-focused approach, and have secured themselves key roles in federal policy-making (Berman forthcoming). The most traumatic odyssey described in Solovey’s book belongs to the various qualitative social sciences, whom may have distinct political interests and goals that differ more quantitative social sciences.

Given these tensions within social science, should RI efforts in government self-identify as being part of social science, or perhaps just qualitative social science, of whose arduous history Solovey chronicles? Would having a new agency separate from NSF, like NSSF or OTA, better enable critical reflection? One could imagine also supporting such research from a reinvigorated National Endowment for the Humanities, where the lack of a ‘science’ framing could allow for a less hindered discussion of key qualitative and normative issues. Having personally attempted to inform engineering decisions in government based on participatory technology assessment (Tomblin et al. 2017; Bertrand, Pirtle, and Tomblin 2017; Weller, Govani, and Farooque 2020), I prefer a simpler argument for qualitative research on science, engineering and society, that does not rely on a ‘social sciences’ framing. The complexity of society and our world require an attention to people, values and the emergent interactions between social and technical systems. Qualitative work is a necessary input to good, values-focused governance and management of science and engineering. Responsible innovation requires using these insights to inform how we live at the intersection of science and society.

As policy-makers consider issues related to ‘science and society’ at OSTP and across the multitude of federal agencies, Solovey’s book is a reminder that social science generally has been hard to establish in government. Beyond the influence of any one individual, NSF social science required a broad constellation of grassroots, Congressional and Executive branch support to become established. Solovey’s story at times feels like the creation of a program at odds with the core identity of an institution, where qualitative social science was seen as different and lesser. Solovey’s history suggests that ‘science and society’ and RI research could also require decades to mature and be established, and would benefit from having permanent (and funded) institutional homes. Establishing a good governance approach for reflection about science and engineering in society may require continued additional ‘Sisyphean’ efforts to help push U.S. science and engineering to more beneficial ends

Additional information

Notes on contributors

Zachary Pirtle

Zachary Pirtle has a Ph.D. in systems engineering from George Washington University and undergraduate degrees in philosophy and mechanical engineering from Arizona State University. He works as an engineer supporting space transportation systems, and his research publications include work on systems engineering, the philosophy of engineering, and science and engineering policy. He previously co-chaired the 2018 Forum on Philosophy, Engineering and Technology and was a Fulbright scholar to Mexico.

Notes

1 The book is free to download from MIT Press. Available at: https://direct.mit.edu/books/book/4852/Social-Science-for-What-Battles-over-Public

2 Solovey relies on the work of an NSF scholar bureaucrat from the 1980s, Otto Larsen, to evaluate final appropriated NSF funding levels, arguing that most NSF social science budgets stayed flat with inflation or went into a slight decline during a time of trenchant criticism (Larsen 2019). Solovey argues that reduced social science funding percentages as a percentage of total NSF funding showed a decline for social science, but one might question whether near-flat absolute budgets is actually a sign of health, and also whether research in the physical sciences should be more expensive than social science.