11 minute read

This piece called "Framing Science" by Matthew Nisbet and Chris Mooney is driving people crazy. It's in the "Policy Forum" section of Science this week, and it's addressed to all those scientists who want to make a difference in politics. At least, I guess so, since it is entirely about how to make a political argument that people will like.

I'm strictly apolitical here, but I think it's an interesting question, to what extent scientists should craft a message to appeal outside their fields, and to whom?

Greg Laden writes about how the article gets the frame concept wrong:

Nisbet and Mooney apply the concept of "frames" to the problem of science and scientists communicating with the public, and being understood in a public forum. With a little digging, it is possible to ascertain that when they speak of "frames" they are speaking of Goffman's Frame Analysis Theory, a widely known concept in Anthropology, Education theory, etc. This is an idea that anyone writing (in a great journal like Science) about communication, etc. would know about and understand. But, what is actually happening here is that the authors are using the word "frame" in a way totally unrelated to Goffmans ideas. This is not a modification of Goffman, or an update, or a better use. Its just misuse and misunderstanding. It borders on being just dumb shit. Considering the value of column inches in science, it is a bit of a shame.

I recommend Laden's post.

Despite any confusion about "framing", Nisbet and Mooney appear to mean something fairly simple: scientists should appeal to non-scientists by presenting their work in a way that conforms to common biases. That may help to communicate (in a way that Coturnix discusses).

But I sort of doubt it. This kind of cynical strategy is the province of used car salesmen and other charlatans. And it's easily exposed by any clever critic who happens to be watching -- why just within the last few days there is this Powerline post fisking an AP article that transparently "frames" William Gray as a crank, and this Ron Bailey article pointing out that the "scientific consensus" "frame" deployed in support of global climate change arguments somehow is illegitimate as applied to genetically modified crops.

My point isn't that these critics are right, but that such criticisms pretty much write themselves! A scientist trying to "frame" in this way is going to end up discredited unless they retreat to the facts anyway. This is, after all, why scientists are typically so cautious in print -- because they work in a field where bad arguments are quickly torn apart by their critics. Why in the world would anyone think politics would be any easier?

UPDATE (4/17/2007): This post got some attention, including a mention on, so I thought it would be worthwhile to post an update. Greg Laden has been busily carrying on a conversation with Nisbet about framing. Nisbet's blog, Framing Science has a roundup of some of the reactions to their commentary. Nisbet himself was kind enough to forward to me a few references in framing in the context of political science -- a field that has undergone substantial development since the 1970's. Laden comments on this literature as well. As an anthropologist, I have to say that I agree with Laden on this: the political science context appears to have developed by using a simplified version of a richer concept from linguistics.

Nisbet forwarded a book chapter (PDF) by Dietram Scheufele reviewing the concept of framing science-related public policy debates. Scheufele has a post on the subject on his own blog. One of the pullout points of the chapter was this:

The popular notion of 'spin', while used more broadly, often refers
to the idea of framing.

That's basically the point. If you want your general concept to be received positively by an uninformed public, then "frame" your message to draw upon their already-extant knowledge and beliefs in a positive light.

As politicians generally discover (and often exploit), spinning a message also tends to change it. For instance, a scientist trying to communicate about climate change two years ago might have tried to frame the issue in the context of the increased number and severity of Atlantic hurricanes in that year. Hurricanes were a major focus of the news; people knew about the damage they caused, nobody likes them. The sound byte, "Global warming causes more hurricanes" might have done more to move public opinion on the issue than any other single argument.

But it was clearly misleading. A number of climate change scientists pointed this out at the time -- noting that a long-term trend in hurricanes could not predict hurricane severity in any given year, that 2005 was exceptional but not outside the range of historic hurricane incidences, and that weather events are generally not the best evidence for long-term climate change.

Noting all these details was the responsible thing to do -- it was not only scientifically correct, but it was also practically correct.

Because last year did not have a high number of Atlantic hurricanes. The "climate change is happening" message might have been momentarily strengthened by the "global warming causes more hurricanes" frame -- the benefit of the spin might even seem (to some) to outweigh the cost of its misleading nature. But it ain't science. In this case, the frame created a backlash.

Public opinion has moved on this issue because scientists have been (mostly) responsible in representing the science, and because they have convinced other opinion leaders (like news anchors and politicians) that climate change is happening. I don't think that has to do with framing; I think it has more to do with status-seeking, social relationships, and the political payoffs of the moment.

A more practical issue is that scientists already have a pretty effective frame working for them. "We're smart and rational. We've thought about this. We're looking for the truth. We're not starting from a biased conclusion."

Now, you'll notice that this particular frame has taken some damage over the years. That damage increases every time it looks like scientists are in fact working to confirm their biased conclusions. In other words, every time we start spinning our results.

Coturnix of Blog Around the Clock has several posts on the topic, which are recommended by Nisbet and Mooney. His consistent position has been that "framing" is not a scientific endeavor, but a necessary political one in which scientists must engage. He writes:

But we need to start funding stem cell research today. We need to start stopping global warming today. We need to rethink our energy use and energy production today. We need to rethink about food production and use today. We need to rethink our economic system, our electoral system, our foreign policy - everything. And science can inform all of those areas. And to an audience that is not interested in (or is even hostile to) science, the policies have to be sold on other merits, on the economic, medical, emotional and esthetic interests of the voters, with the underlying science being brought up as needed and in small, palatable measures.
Framing science is not teaching science. Framing science is persuading voters that a policy (which, in this case is based on some underlying science) is good. It has little to do with science, and all to do with politics. But we have to win some political battles first (hello, see who is running all branches of the government these days!?) if we want to survive and if we want science to survive as an endeavor.

I think that Nisbet and Mooney framed their original commentary very carefully. Clearly their intent is that scientists take up the banner of political activism, as reflected in their first sentence:

Issues at the intersection of science and politics, such as climate change, evolution, and embryonic stem cell research, receive considerable public attention, which is likely to grow, especially in the United States as the 2008 presidential election heats up.

These are marching orders. Yes, all good scientists must rise up to influence the 2008 U. S. presidential election. It'll be just like that "Simpsons" episode where Mayor Quimby resigns and the council of learned elders takes over the city. One Frink to rule us all!

But a consistent remark by critics (notably, Larry Moran) has been the failure of the authors to show any evidence that their strategy actually works. And in their commentary, Nisbet and Mooney don't even specify their goals! Consider the paragraphs just following the first sentence:

Without misrepresenting scientific information on highly contested issues, scientists must learn to actively "frame" information to make it relevant to different audiences. Some in the scientific community have been receptive to this message (1). However, many scientists retain the well-intentioned belief that, if laypeople better understood technical complexities from news coverage, their viewpoints would be more like scientists', and controversy would subside.
In reality, citizens do not use the news media as scientists assume. Research shows that people are rarely well enough informed or motivated to weigh competing ideas and arguments. Faced with a daily torrent of news, citizens use their value predispositions (such as political or religious beliefs) as perceptual screens, selecting news outlets and Web sites whose outlooks match their own (2). Such screening reduces the choices of what to pay attention to and accept as valid (3).
Frames organize central ideas, defining a controversy to resonate with core values and assumptions. Frames pare down complex issues by giving some aspects greater emphasis. They allow citizens to rapidly identify why an issue matters, who might be responsible, and what should be done (4, 5).

As they say on the SAT, the main idea is clearly this "frame" thing. The thesis is "scientists must learn to actively 'frame' information."

But the article never tells us why scientists would want to do this. It leaves us with several possible goals, without specifying which is correct:

(a) to make information relevant to different audiences.

(b) to make controversy subside.

(c) to tell citizens what to do.

(d) to identify why an issue matters.

(e) to identify "who might be responsible."

Of course, all of these are the sort of things that "well-intentioned" scientists might like. But only (a) and (d) are arguably proper roles for science, and (c) and (e) are downright scary. None of them logically follow as proper goals for "framing"; for instance, if we want to make information relevant to different audiences, why not use plain language? Why not say, "The information from my science is relevant because of X?" Why "frame?"

Why am I so exercised about this?

Clearly most people do not know the scientific concepts that would be necessary to really evaluate climate change science or any number of other topics. These definitely are relevant in a political context. Communicating the relevant issues widely requires simplification. People will not accept simplified arguments (which are often easily contradicted) that disagree with their core assumptions. So why not smooth things over with a good frame?

The reason, at least in my mind, is that science is never unanimous. Consensus itself is a meaningless concept in science. Even a crank can be right, if his predictions come true. Science depends on the concept of testability, which in my mind makes it the most special system of knowledge ever devised. It is good that science is conservative, and it is good that it is slow, because both these things contribute to claims being tested properly.

But even though science is not governed by consensus, a great number of scientists behave as if it were. That is almost enough to make it true. There are no end of scientific papers that begin a literature review with words like "the scientific consensus on X is...", or "X idea is intellectually dead." These are frames, but they are frames in a scientific context. Indeed, every literature review begins with assumptions (open or hidden) which in effect frame the results of analyses. Change the assumptions and you change the conclusions.

It is the job of the scientist to figure all this out. It is the job of the scientist to be skeptical of claims, and to root out hidden assumptions. Not all scientists are good at this job.

Reasonable people may disagree with me, but I think that more framing will make this worse, not better. I think that scientists are unlikely to be unchanged by the public discourse. I think that cautious, conservative scientists who stick to the facts represent themselves in the best light -- and really, in the most effective frame. I think that if this frame is the accepted role of scientists, we will have an enriched public discourse.

It may be very difficult for science to affect politics, but it is very easy for politics to affect science. Science has far more to lose by accepting the strategy of framing than politicians have to gain by hearing us more easily.


Scheufele DA. 2007. Messages and heuristics: how audiences form attitudes about emerging technologies. Pp. 20-25 in Engaging Science: Thoughts, Deeds, Analysis and Action, Turney J, ed. Wellcome Trust, London. Online