Social Evolution Forum
FIND sef:
What are the roles of scientists in policy-making?

As Peter Turchin has reported in a recent SEF blog post, the formation of a Cultural Evolution Society is in progress. We have signed up about 1,200 Founding Members and have surveyed them for their ideas about the grand challenges facing our field. Of those who responded to the survey, 75% come from the research world and 35% work on practical applications. An additional 38% are teachers. Some obviously do more than one of these things.

Given that many evolutionary social scientists are interested in influencing public policy debates I thought that ESF readers might be interested in my late colleague Paul Sabatier’s ideas on how scientists can be effective in this role. Paul was an influential political scientist interested in the policy making process. He was, among other things, interested in the role of scientists and other experts in the policy process. In the late 1970s he conducted a survey of the kinds of participation in the policy arena UC Davis faculty had and wanted to have in the policy arena. More generally, he conducted surveys of policy elites. The elites include elected officials and their high level staff members, the agency chiefs and staff members who have an important role in regulation setting and enforcement, the active representatives of businesses and other entities with an economic stake in issues, issue oriented voluntary organizations, and independent consultants and technical advisors to activist organizations and government agencies. Academy-based scientists are a small minority of members. Over the course of his career he carried out these studies in various environmental policy arenas including national air pollution regulation, coastal zone management in California and France, and the management of Lake Tahoe, in which the federal government played a large role due to the bi-state situation of the lake.

From his early work Paul derived what he called the Advocacy Coalition Framework (ACF) as a general picture of the policy making process. He observed that the policy world is divided into many policy sub-systems in which a fairly stable set of policy elites participate in policy making. Roughly a thousand people are influential actors in any one large-state or federal sub-system at any one time. These actors are divided into advocacy coalitions that collectively advocate for particular policies in the arena concerned.

Often there are two advocacy coalitions, sometimes more. For example, in making policy for California’s big inland estuary, the San Francisco Bay—Delta System, one advocacy coalition is organized around agricultural and urban water users in the Central Valley and Southern California for whom the Bay—Delta is a key link in California’s massive plumbing system conveying water from the rainy north to the arid south of the state. The other major coalition is composed of those for whom the Bay—Delta is an environmental resource for the ten million or so people who live adjacent to it. This latter coalition includes local irrigators, sport and commercial fishers, and recreation oriented business interests in addition to environmentalists devoted to non-commercial interests.

Over decades, advocacy coalitions change as the larger society changes. For example, the roots of the Bay—Delta policy go back to conflicts between hydraulic miners, who released massive quantities of silt into the system in the late 19th Century, and farmers whose irrigation facilities were inundated by that silt. But mining ceased to be important many decades ago. The modern advocacy coalitions organized in the mid-20th Century in the wake of the massive federal Central Valley Project organized to deliver water stored in Shasta Lake (closed in 1945) to municipalities and irrigators in the Central Valley, and the state California Water Project that distributes water stored in Lake Oroville (closed in 1969) to San Joaquin Valley irrigators and Southern California municipalities.

The ACF model became quite influential in the policy analysis community. Paul tirelessly advocated the comparative testing of theories of the policy process and many others found virtue in the ACF. A half-dozen or so of his papers and books have attracted over 1,000 Google Scholar citations.

What does the ACF imply for the role of scientists in public policy formulation? The simple answer is that to have an effect on a policy arena a scientist has to become a member of an advocacy coalition. Paul contrasted his empirical findings with the textbook picture of the role of science in public policy that was current when he started his career. The textbook picture cast elected officials as policy formulators and scientists and non-elected officials as objective, disinterested technocrats. The technocrats were in charge of making sure elected officials had the objective facts before them when they made policy decisions, and that, once made, the policies were competently executed.

In Paul’s view, no one is completely objective and disinterested. Scientists tend to belong to particular advocacy coalitions because disciplines themselves come with policy objectives partly built into them. Economists classically study markets and their ability to create economic growth. Ecologists come to that discipline because they are interested in organisms and ecosystems. Engineers like to design and build things. Anthropologists like to study marginalized or non-Western societies and subcultures. The very ethos of a discipline tends to incline people to join predictable advocacy coalitions.

Most disciplines have an official commitment to scientific objectivity, and it is important not to sell this commitment short. Still, every discipline has its inherent biases and every individual scientist her own personal interests and ideological commitments. To get to the special role that science and other forms of technical expertise play in policy formulation we must first concede this interestedness of scientists.

As members of advocacy coalitions, scientists will grapple with scientists and technicians who are members of other advocacy coalitions. Each side’s scientists and technicians attempt to create facts in the debate that the other side cannot deny without obviously abandoning the standards of scientific objectivity. As policy debates unfold on the decadal time scale, some scientific evidence can become impossible to deny. For example, smoking was linked to lung cancer by very convincing studies by around 1970, and tobacco companies were forced to resort to concealing information and using deceptive pseudo-scientific campaigns to defend their businesses. Paul described the gradual accumulation of relatively sound science and practical experience in a policy arena as a process of policy learning.

Scientists who wish to remain above the rough-and-tumble of advocacy coalition politics face a fatal problem. Policy thought pieces delivered from a completely disinterested perspective are outside the policy discourse. Members of advocacy coalitions have thought hard and struggled long with the issues of their policy arena. Rightly or wrongly, they think they know the issues involved rather well. Outsiders, no matter how well credentialed, are treated like disciplinary scientists treat outside scholars and cocktail party boors. They may be smart, but they just don’t know the issues as they are in play in a policy sub-system. As such, disinterested experts seldom play a constructive part in policy debates. They are often seen as obviously not motivated to stick with the policy arena long enough learn the ropes and then to have an impact.

Paul and his colleagues have three pieces of advice to scientists and other experts who want to make an impact on public policy:

Develop a belief system awareness and deep knowledge about policy sub-system affairs. Belief system awareness is about knowing how your own belief system works as well as that of other members of your advocacy coalition and those of the opposing coalition. In effect, the policy debate is about strengthening the acceptance of your coalition’s belief systems and undermining those of your opponents. Belief systems are hard to change, but the opinions of less deeply committed outsiders also matter.

1. Analytic methods are a key to the application of expertise to policy. The analytic methods of technical experts tend to vary by discipline and the strengths and weakness of methods will be highlighted in policy debates. Economists and engineers cultivate a reputation for advanced quantitative methods and are ruthless in using their assumed, if rather mythological, superiority to try to drive “soft” social scientists on the other side out of the debate. If you can be portrayed as a softy you’d better have a convincing defense ready for such an attack.

Lay people are often not shy about using their own experience and ideology especially when they detect inadequacies or an ideological slant in the science and technology involved in a political debate. You can’t expect exaggerated respect for self-proclaimed expertise or for having a ticket like a PhD degree. In this regard, local knowledge and sentiment often play a large role in policy debates. Local technical experts and laypeople often have a different picture of issues than people from other localities, or those representing larger scale parts of the system, and they are often right.Knowledge about other policy arenas is often important; no policy arena is self-contained. For example, elected officials must be to some extent policy generalists. An elected official or staffer that you normally count as a member of your advocacy coalition may have to withdraw support because of conflicting issues arising in another arena.

2. Build networks. Advocacy coalitions are complex entities composed typically of hundreds of people who are collaborating to produce a public good, the implementation of policies they collectively favor. There is a division of labor within the coalition and you often need help from people with complementary expertise.

Inevitably, differences of interest arise within the coalition, not to mention between them. Personal relations develop trust between individuals that can offer ways to make progress despite conflicts of interest. Currently, in the US federal government, bitter partisanship has reduced trust to the point that Congress has virtually ceased to function. Personal relationships give one access to information and expertise that is vital to one’s own role.

3. Participate for long periods of time. Knowledge acquisition and network building require time and effort. Herbert Simon estimated that ten years are required to acquire a cutting edge level of expertise in a scientific subject, and likely the same is true for technical experts and lay members of an advocacy coalition. The trust and connections a participant forms in the course of such learning are of as much value as the technical knowledge. A substantial commitment of time and effort is required if one expects to become an influential member of an advocacy coalition.

Policy issues often require decades to resolve. To have the experience of decisively winning (or losing) on a particular issue may require a career-long dedication. You might have to wait for a rare moment of opportunity to enact some policy that has seemed obvious to you and your coalition for a lifetime. As a scientist, the acceptance of innovative methods and findings often takes a decade of effort to diffuse throughout the policy community, especially to the experts of opposing coalitions.

Given the effort involved, no one should be surprised that a division of labor tends to arise between applied and basic scientists. Nevertheless, a substantial number of basic, academy-based scientists make important contributions to policy issues.

What role do scientific societies play in the policy process? Scientific societies are part of a complex of what Paul called system maintaining organizations that encapsulate the policy process. Some of these organizations are formal government ones maintaining the existing constitutional order such as the courts, and ones that maintain the knowledge infrastructure like the educational system. Some are organs of civil society like scientific societies.

Scientific societies play an essential system-sustaining role in providing the basic infrastructure of science. In essence, universities and other employers of scientists outsource hiring, promotion, and funding decisions to the scientific societies that publish the journals in which a scientific reputation is built. The well published get hired, promoted, and funded. Scientific societies are also a locus for the promotion of a field as a whole, such as lobbying funding agencies and universities for a larger share of the funding pie and faculty lines. They can point out opportunities for curriculum reform. They can tout the importance of a discipline or area of study for applied purposes. In doing these things they must take care to maintain their reputation for objectivity.

In the policy realm, objectivity partly means a commitment to democratic values. No science or scientist has a monopoly on truth. Different scientists will have commitments to many different advocacy coalitions. It is almost like lawyers. Every side in a political debate deserves to have whatever scientists they can recruit to their cause. Scientific societies do close their journals to bad science, judging that some issues are settled. In the smoking and cancer debate and others that followed, the “merchants of doubt” scientists excessively motivated by ideology properly lost access to prestigious academic journals.

To my way of thinking, Paul’s picture of the role of scientists in public policy, and his advice to scientists who want to participate, are hard-headed realism. I commend it to readers on this account. And I encourage those involved in forming the Cultural Evolution Society to carefully distinguish the roles of scientists across the array of activities from basic research to education and advocacy. The institutions that manage these activities differ, so the roles of scientists will differ accordingly.

Further reading:

Weible, C., Heikkila, T., deLeon, P., & Sabatier, P. (2012). Understanding and influencing the policy process. Policy Sciences, 45(1), 1-21. doi:10.1007/s11077-011-9143-5


Join the discussion


  1. Chris Dowdeswell says:

    Excellent article. I think you come to a very important place near the end, noting that, “In the policy realm, objectivity partly means a commitment to democratic values.” Here is the reason why the US congress is paralyzed, and Ostrom offers a quick-fix solution in the development and efficient enforcement of graduated sanctions. To find a long-term and lasting solution, however, we need to understand the issue of values-formation much better. The reasonable assumption of a genotypic prosociality is one thing, but its phenotypic expression is another. As values-formation is better understood psychologically, we might be able to help our existing values-formative cultural practices self-consciously evolve in a more democratic manner.

  2. Tom Dietz says:

    I had the great privilege of being at UCDavis when Paul Sabatier was formulating ACF and Pete Richerson and Rob Boyd were beginning their work on cultural evolution. So my own work has been influenced by these perspectives. I would emphasize that scientists need to be very careful to understand what values they are bringing to the policy arena. We all have values–we can’t make choices without them, and thus wouldn’t have policy preferences without using our values. But they are often buried in the “taken for granted assumptions” of our field. It’s also important to note that we often have to apply science to particular problems or setting that are very different contexts from where the knowledge was established. This increases uncertainty, makes “local” knowledge important and requires scientists to be a bit humble and open-minded. I’ve written a bit about this in: Dietz, Thomas. 2013. “Bringing Values and Deliberation to Science Communication.” Proceedings of the National Academy of Sciences 110:14081-14087.

  3. Kelly Gerling says:

    Another way to influence public policy would be to include university professors as actual members of legislatures at both state and national levels as Ireland has done.