Roger and I have both written on Obama’s scientific integrity memo over the last couple of days (here and here), pointing out the (misguided) persistence of many commentators in asserting that science should supersede politics and values in the policy making process. So far, it seems that Obama’s administration has avoided conflating scientific integrity with technocracy.

But all this talk about scientific integrity is really just a reaction against a few controversial, and highly visible incidents associated with the Bush Administration. It’s campaign politics. Now that we’ve had a cathartic and symbolic restoration of scientific integrity, how should we think about actually governing science policy going forward? In an article in the forthcoming edition of Issues in Science and Technology (PDF), ASU President Michael Crow offers one take on how we should judge Obama’s science policy (my emphasis):

So even as we applaud our new national science policy leaders, we should also encourage the Obama administration to make the necessary transition from a campaign posture focused on countering political interference in science to a governing posture that connects the $150 billion U.S. public investment in S&T to our most urgent problems.

One key obstacle to strengthening this connection is a culture that values “pure” research above other types, as if some invisible hand will steer scientists’ curiosity toward socially useful inquiries. There is no such hand. … Overall, we act as if the intellectual goals of scientists are automatically and inevitably aligned with our most important goals as a society. They are not.

As Crow points out, scientific integrity is not the major problem facing our R&D enterprise; its the governance of science. We don’t just need innovative scientists; we need innovative institutions that manage our research investments creatively, effectively, and with the goal of aligning science with the needs of society:

The success of President Obama’s new science team should be measured by its ability to break down the historical disconnect between science and policy. Our scientific enterprise excels at creating knowledge, but it continues to embrace the myth that new knowledge, emerging from the stubbornly disciplinary channels of today’s scientific programs, automatically and serendipitously turns into social benefit. A new administration facing a host of enormous challenges to human welfare can best unleash the power of S&T by rejecting this myth and building a government-wide knowledge creating enterprise that strengthens the linkages between research and social need.

It’s worth reading the whole article, as he provides examples of this kind of institutional innovation, as well as a brief discussion of how these ideas interact with the university setting.

The public must be able to trust the science and scientific process informing public policy decisions.  Political officials should not suppress or alter scientific or technological findings and conclusions.  If scientific and technological information is developed and used by the Federal Government, it should ordinarily be made available to the public.  To the extent permitted by law, there should be transparency in the preparation, identification, and use of scientific and technological information in policymaking.  The selection of scientists and technology professionals for positions in the executive branch should be based on their scientific and technological knowledge, credentials, experience, and integrity.

For many, this is not just about opening the door to a particular kind of medical research; they view it as a fundamental change in the role played by science in policy.

This seems to emerge in the media coverage of the event. There is considerable discrepancy between the actual contents of the memo, and what the media (and those they interview) have been saying about it. For example, on NPR:

DeGette says that during the Bush administration, scientific policy was often dictated by things other than scientific evidence.

Well, yes, of course it was. As is often said on this blog, policy is never dictated by science, and Obama’s memo says nothing that would suggest otherwise. It is very much focused on process and openness, but makes no statements about how science should influence decision making.

The Washington Post quotes Harold Varmus (former NIH director, Nobel laureate, and Obama advisor):

Today’s executive order “is consistent with the president’s determination to use sound scientific practice . . . instead of dogma in developing federal policy”

This suggests that you can somehow use science instead of values to develop policy. But Obama’s stem cell decision is no less value driven than was George Bush’s. Regardless of your position, to come to a conclusion on the ethics of stem cell research you must wrestle with difficult issues such as the acceptability of destroying human embryos. Obama’s words and actions suggest nothing like the “determination” Varmus describes.

It’s important to remember that this event represents a political success — a shift away from one set of values, and toward another (though of course, it is not so black and white). It is not, by any means, a shift from politics toward science. Even if that were possible, it’s hard to see why it would be desirable.

In a letter in this issue of Times Higher Education, the group calls for academics to rebel against new rules that state that the potential financial or social effects of research must be highlighted in a two-page “impact summary” in grant applications.

The requirement to provide a summary, answering questions about who might benefit from the research and how a financial return could be ensured, is being phased in by the UK’s seven research councils. The summary will be used by peer reviewers as a factor when determining which applications receive funding.

The scientists suggest two problems with this: 1. that they should not be expected to predict potential outcomes of their research, the results of which are highly uncertain, and 2. that this will shift the balance of funding away from “blue skies” research:

“The academic community must stand up,” said Professor Braben, adding that history showed that even the most seemingly inapplicable of scientific discoveries could yield huge economic benefits, such as the development of lasers.

“You cannot command developments at the frontier, it is not possible,” Professor Braben said.

He added that the new policy spelt the end for blue-skies research. “As soon as you identify a beneficiary for research … the councils are going to turn it around and say, right, deliver. And then it is applied research … You can’t have blue-skies research if you put caveats on it.”

For some reason it is still necessary to point out that the dichotomy between basic and applied research is false: applied research may lead to fundamental breakthroughs in knowledge, and “blue skies” research can lead to useful applications. But in any case, the scientists (not to mention one of the blog reactions) may be misunderstanding or misrepresenting just what these new expectations entail:

Philip Esler, chief executive of the Arts and Humanities Research Council, speaking on behalf of Research Councils UK, said: “The description of impact that the research councils work with is broad, encompassing not only the contribution research makes to the economy but also to society as a whole.

“It covers not only economic benefits, but also those related to public policy, quality of life, health and creative output. Research councils will not be disadvantaging blue-skies research, nor stifling creativity.

“The impact statement is not designed to ask peer reviewers or applicants to predict future benefits. It is intended to allow the applicant to highlight potential pathways to impact, especially through collaboration with partners, and to help the research councils support them in these activities.

This makes a great deal more sense. Almost all government funding for science is justified on the basis that it will contribute to these kinds of outcomes. Is it all that scandalous to suggest that scientists should have some understanding of the social contexts of the problems they are studying, or at least that they collaborate with someone who does?

One of the biggest challenges in science lies in connecting new knowledge with users who can actually benefit from it. This lesson is demonstrated in many different areas, particularly in climate science, where huge amounts of resources are spent on physical science that yields very little useful information to natural resource managers and other decision makers at whom this new knowledge is targeted. Making these connections requires partnerships between scientists and potential users, and it requires that scientists understand the context of knowledge use and/or decision making.

Of course, it is quite possible that a new requirement such as this could go the way of NSF’s criterion 2, which is seen by many as merely a formality with little meaning. Or, if it is poorly implemented, the new rules may place undue burden on researchers in terms of our expectations of them. But the emphasis on collaboration with partners is promising.

And the reaction of scientists to this effort reflect a dangerous sense of entitlement. Yes, the funding of science is important, but the science community is lucky to have it. If we take this at face value as an effort by the government to build stronger connections between science and the problems it is trying to solve, then resistance to such ideas starts to look pretty arrogant.

Last week, Science printed a response to that article, which, to our disappointment, was not the one submitted by Roger Pielke, Dan Sarewitz, Lisa Dilling, and me. So, after consulting with the original authors, we have decided to use Prometheus as a venue for continuing this discussion. I will include our unpublished letter below, and hope that Jim Baker will post their response in the comments, and we can go from there.

The short version of our response is that: 1.) there is no reason to think that the proposed consolidation would be good for science, and 2.) whether or not consolidation turns out to be good for science is irrelevant to the question of how best to orient scientific research so that it helps to address larger societal problems.

Here is our letter:

In a July 4 Policy Forum (1), a formidable group of experts on environmental research policy proposes the creation of an Earth Systems Science Agency (ESSA) by combining NOAA and USGS.

We agree that the U.S. faces “unprecedented environmental and economic challenges” at the same time as the environmental sciences have experienced declining national investments and increasing politicization.  But reducing the number of key environmental science agencies from six to five will neither alleviate the challenges of inter-agency coordination, nor create a comprehensive research capacity. Moreover, the politics of ripping NOAA and the USGS out of their home departments (Commerce and Interior) would certainly lead to bruising political battles with unpredictable outcomes.  In a time of budget constraint, consolidation could make the problem worse for science by creating more conspicuous targets for political manipulation and budget cutting (2, 3).

Even more troubling is the “science push” approach that the authors advocate. ESSA’s proposed design puts data acquisition and basic science at the core of efforts to develop decision-relevant information that is then delivered to the outside world.  Decades of experience and scholarship, including a recent NRC assessment of the Nation’s climate science enterprise (4), shows that this approach is good at advancing basic knowledge but largely ineffective at providing useful knowledge for decision makers.

The most successful environmental science programs—including those in the USGS and NOAA—have long ago moved away from a simple linear approach (“do the science, then communicate it”) to embrace an integrated model where research and engagement are tightly linked from the outset, so that research agendas and products are responsive to the needs and capabilities of information users, and users in turn know what they can expect from scientists (5-10).  What makes them successful is not the breadth of their research portfolios, but their approach to research and problem solving. A truly innovative approach to improving society’s capacity to respond to environmental challenges would focus investments on leveraging the experience and resources of agency programs with ongoing decision support and service capacity, such as hazards and water monitoring programs at USGS, and the Coastal Services Center and National Integrated Drought Information Systems at NOAA.  Absent such innovation, bureaucratic reshuffling is unlikely to improve the value of science in addressing society’s most urgent challenges.

1.    M. Schaefer et al., Science 321, 44 (July 4, 2008, 2008).
2.    D. Sarewitz, Issues in Science & Technology 23, 31 (Summer, 2007).
3.    D. Goldston, Nature 451,  (2008).
4.    NRC, “Evaluating Progress of the U.S. Climate Change Science Program:  Methods and Preliminary Results”  (National Academies Press, 2007).
5.    D. W. Cash, J. C. Borck, A. G. Patt, Science Technology Human Values 31, 465 (July 1, 2006, 2006).
6.    H. Meinke et al., Climate Research 33, 101 (2006).
7.    D. Sarewitz, R. A. Pielke, Environmental Science & Policy 10, 5 (2007).
8.    K. Jacobs, G. Garfin, M. Lenart, Environment 47, 6 (2005).
9.    S. Agrawala, K. Broad, D. H. Guston, Science Technology Human Values 26, 454 (October 1, 2001, 2001).
10.    D. Cash et al., Proceedings of the National Academy of Sciences of the United States of America 100, 8086 (2003).

“Given the many urgent scientific and technological challenges facing America and the rest of the world, the increasing need for accurate scientific information in political decision making, and the vital role scientific innovation plays in spurring economic growth and competitiveness, we call for a public debate in which the U.S. presidential candidates share their views on the issues of The Environment, Health and Medicine, and Science and Technology Policy.”

I won’t go into the various arguments for and against this idea, but I think it’s worth contrasting the title of this effort — “Science Debate 2008” – with what is actually being proposed in the above quotation. The issues listed span a political and cultural landscape of which science occupies only a very small piece. On the other hand, there are far more issues in which science plays a part (e.g. space, transportation, agriculture, …) that somehow did not make the cut. Why these issues in this particular debate? What is the goal of holding this debate?

Despite the title of this movement, what is advocated here is not a “science debate,” and as Goldston pointed out in Nature, applying such a name to it potentially does a great disservice to whatever discourse might emerge. One need only look at Kennedy’s suggested questions in Science to see why this is true. Some of these questions are about politics and values, others are about budgetary aspects of science policy, and still others are about criteria of scientific merit. While a sitting president could take a position on any of these example questions, few of them are high profile enough to reach that status.

Others are misleadingly framed to begin with. For example, no single person or entity determines the balance between “major-program project research and investigator-initiated basic research grants” and it is doubtful that it would be possible (let alone desirable) to alter this reality. (And I’ll leave aside the question of whether anyone at all in the general public would find this question interesting.)

It seems that the agenda of this movement is to raise the profile of a very specific set of issues. Why these issues should represented as inherently “scientific” is mystery to me. What should be clarified is the reasoning behind selecting these issues, and the overall goals of the proposed debate. Maybe then the supporters could make some progress in dealing with the issues raised by Goldston and others, and perhaps even make headway toward a truly useful event.