The notion of a scientific consensus on climate change has become a common fixture in discussions of climate science and policy. What once may have been a useful concept has now become little more than a political touchstone used more often to advance political agendas than to support policy development. Even to discuss this issue, as I am here, is to risk being labeled a “climate skeptic” by the denizens of politically correct discourse on climate. Lest there be any confusion, as I’ve written many times, I accept the IPCC consensus on climate science, and base my readily available peer-reviewed climate policy research on the IPCC scientific consensus.

There is of course a scientific consensus on climate change. But what, exactly, does this mean? Checking in with the dictionary folks at Oxford we learn that a “consensus” is a “general agreement.” And a “general agreement” means that there is an “accordance in opinion or feeling … concerning all or most people.” But a “general agreement” is imprecise, so the Intergovernmental Panel on Climate Change (IPCC) has sought to institutionalize the process of defining exactly what it is that scientists are in accordance on and with what certainty. In 2001 the IPCC issued its Third Assessment Report and in its communiqué to policy makers summarized the Working Group I consensus in 17 pages (hereafter SPM, here as PDF, and the other working groups also produced SPMs).

This is where things get tricky. Consider for example this statement from the SPM:

In the light of new evidence and taking into account the remaining uncertainties, most of the observed warming over the last 50 years is likely to have been due to the increase in greenhouse gas concentrations.

How might one interpret this statement? Fortunately, the IPCC has standardized its terminology on uncertainty (see this document in PDF). So an equivalent phrase would be:

Relevant IPCC Lead Authors judge that over the past 50 years that the majority of observed warming has been due to the increase in greenhouse gas concentrations with between 64% and 90% certainty.

But even with the quantification, it is unclear how one should interpret this statement, because the IPCC relies on “expert judgment” – an opaque, negotiated process among IPCC co-authors — to arrive at its consensus statements and levels of uncertainty. There are other methods (e.g., formal expert elicitation) to derive a consensus perspective, but none can remove the need for judgment and corresponding ambiguity (for more discussion of this point see the End Note below). Think about the following questions that one might ask about this IPCC statement:

Would it be fair to interpret this statement as indicating that anyone who is less than 64% certain that “most of the observed warming over the last 50 years is likely to have been due to the increase in greenhouse gas concentrations” is a climate skeptic?

Similarly, would it be fair to interpret this statement as indicating that anyone who is more than 90% certain that “most of the observed warming over the last 50 years is likely to have been due to the increase in greenhouse gas concentrations” is a climate alarmist?

Does the statement in fact allow for scientifically valid views outside the 64%-90% range, with the range simply representing a measure of central tendency?

I have no idea how to answer these questions in a way faithful to the intent of the IPCC, and in my experience it turns out if you ask 5 IPCC contributors you get 5 different answers. The IPCC consensus is remarkably open to interpretation, and consequently we should not be surprised to see it used in very different ways.

For instance, when I read the 2001 IPCC statement from above as a non-expert, I see that the IPCC TAR decided to use the word “likely,” rather than “very likely” (>90%) or “virtually certain” (>99%), which seems to clearly suggest that there was some considerable uncertainty in this statement, i.e., up to a 1/3 chance that the warming over the past 50 years was due to factors other than greenhouse gases. The statement would thus seem to allow a large enough amount of uncertainty to warrant scientifically valid questions about the trend and its causality. And it would also seem that at least in 2001 the climate research community would have agreed with this interpretation, having subsequently focused a considerable amount of research attention on the issues of detection and .attribution related to temperature.

Others interpret the IPCC consensus statements in a much more categorical fashion, for instance, in 2004 Stefan Rahmstorf, a climate scientist and co-host of the RealClimate blog, appealed to and restated this same IPCC consensus statement but completely dropped the references to uncertainty writing (here in PDF),

In the 20th century, global climate warmed by ~0.6°C . . . Most of this warming is due to the rising concentration of CO2 and other anthropogenic gases.”

He then labels people who have doubts about the causes of the global temperature increase – “The attribution sceptics doubt that human activities are responsible for the observed trends.” To be fair, it may be that Rahmstorf and his colleagues at RealClimate believe that the science on this particular issue has now moved from “likely” into the realm of “virtually certain,” perhaps giving us a hint of what to expect in the next IPCC, explaining that

The following scientific findings can no longer credibly be argued to be in dispute:

(1) The observed large-scale warming of the atmosphere and ocean is an entirely expected, and in fact well-predicted, consequence of the human-caused accumulation of greenhouse gases in the atmosphere.

(2) There is no other reasonable scientific explanation for the observed warming.

But until the next IPCC process actually is completed and guidance is formally provided to policymakers, non-experts like me have no choice but to rely on the 2001 IPCC as the most recent authoritative consensus statement of the scientific community. The only alternative that we non-experts have is to pick and choose among more recent science and interpretations of that science according to whose views we trust or feel more politically comfortable with. Preventing this sort of cherry picking through a comprehensive community view was one of the main reasons for the creation of the IPCC in the first place.

In practice, the notion of “consensus” is typically portrayed in pure black and white. You are either with the consensus or you are one of those skeptics, who Rahmstorf characterizes as “frequently dishonest.” And we see this pattern whether the issue is temperature trends as discussed above, or a seemingly even more uncertain issue like hurricanes and climate change. Consider this quote from a recent ABC News story which focuses on the nefarious intentions of the oil industry-funded climate skeptics, “But some continue to promote the idea there are “uncertainties in the science”.”

Such characterizations by scientists and the media can have a chilling effect on scientific discourse because they create powerful incentives to avoid research which may be somehow critical of the consensus, leaving such research primarily to those seeking to challenge the consensus. So as a result, some chose to focus on critiquing the hockey stick, others choose to critique the satellite temperature record, still others focus on critiquing the role of the sun in the climate system, and so on. It seems clear that the orientation that each of these researchers take to the subject of study is profoundly shaped by an interest in either consolidating or expanding the “consensus,” with trans-scientific agendas shaping more than a few perspectives. As a result, many climate scientists have become much more tactical in how they select research topics. If this interpretation is anywhere close to the mark then climate science as a whole suffers because of an adversarial orientation in which scientists pick and choose research topics and stances according to what they want to (dis)prove, not simply what they want to learn.

Underlying much of the tactics of climate science are of course political perspectives. What the ABC News article cited above really means is that some assert that there are “uncertainties in the science” as a rationale for business-as-usual policies on climate change, and it is this sort of characterization of uncertainty that is unwelcome. At the same time, it seems that the mentioning of those very same uncertainties in climate science is OK, so long as one also accompanies that with an appropriate qualification about the need for political action.

Consider that among much of the environmental and scientific communities it is appropriate for James Hansen to state: “Climate models suggest that doubled CO2 would cause 3ºC global warming, with an uncertainty of at least 50%” (PDF). But among many of the same people, the following statement by George Bush is inappropriate, “We do not know how much our climate could, or will change in the future. We do not know how fast change will occur, or even how some of our actions could impact it.” Both statements seem to me to be obviously scientifically consistent the knowledge represented by IPCC. The difference is that James Hansen has openly advocated one political agenda and George Bush another. The defining characteristic of a “climate skeptic” does not seem to be only how one views climate science, but how one orients climate science in political context. This explains why it will be extremely unlikely to see Ted Nordhaus and Michael Schellenberger labelled climate skeptics, even though they wrote in Saturday’s New York Times a statement entirely consistent with Rahmstorf’s “attribution skeptics”:

Environmentalists and their opponents have spent far too much time debating whether global warming is caused by humans, and whether the transition to cleaner energy sources will be good or bad for the economy. Whatever the causes, warming is a genuine risk.

They won’t be called skeptics because they also advocate a “Global Warming Preparedness Act.” Politics manifests itself in how various people choose to interpret the notion of “consensus” and use it in politically convenient ways.

In its efforts to precisely delineate a “consensus” the IPCC and those who incessantly cite a universal consensus may be leading us astray. The focus on consensus diverts attention from policy development to endless scientific debates with the pockets of remaining resistance to the consensus, and in the process creates a false impression that securing total conformance to the “consensus” is a prerequisite for action.

As M. Granger Morgan and colleagues remind us, the exact definition of a scientific consensus can be a function of the method used to elicit scientific perspectives. For example, they find that the consensus view on uncertainties associated with aerosol forcing reported by the IPCC in 2001 dramatically underestimated the uncertainties actually held by the relevant experts, they conclude,

The range of uncertainty that a number of experts associated with their estimates, especially those for total aerosol forcing and for surface forcing, was often much larger than that suggested in 2001 by the IPCC Working Group 1 summary figure.

And as well, science advances with surprising findings, new discoveries, and unexpected results. For all of these reasons the precise details of a scientific consensus are always in flux, seeking to pin it down precisely in for it order to sit untouched for 5 or 6 years does not faithfully represent how science works, and more importantly, it does not effectively inform the needs of policy making. For these reasons I wrote last year:

So in addition to arguing about the science of climate change as a proxy for political debate on climate policy, we now can add arguments about the notion of consensus itself. These proxy debates are both a distraction from progress on climate change and a reflection of the tendency of all involved to politicize climate science. The actions that we take on climate change should be robust to (i) the diversity of scientific perspectives, and thus also to (ii) the diversity of perspectives of the nature of the consensus. A consensus is a measure of a central tendency and, as such, it necessarily has a distribution of perspectives around that central measure (1). On climate change, almost all of this distribution is well within the bounds of legitimate scientific debate and reflected within the full text of the IPCC reports. Our policies should not be optimized to reflect a single measure of the central tendency or, worse yet, caricatures of that measure, but instead they should be robust enough to accommodate the distribution of perspectives around that central measure, thus providing a buffer against the possibility that we might learn more in the future.

Until we have a consensus on the diminishing value of the notion of consensus as the keystone of the climate debate, we’ll continue to see the politicization of climate science and the continued gridlock on climate policy. Climate researchers will continue to array themselves tactically with respect to the consensus, ensuring a continuous stream of research results that shade the consensus to and fro. But the reality is, scientists are in general agreement, and at this point effective action on climate change does not depend on either strengthening or more precisely measuring that agreement. The important questions instead are what actions, when, by whom, at what cost, to whom, how, and why?

***End note: From Reilly et al. 2001:

Expert judgment was widely used in preparing the TAR, but the organizers were not able to impose a consistent procedure across the various components. The likelihood terms above were variously assigned on the basis of “judgmental estimates” in the discussion of the science of climate (1) and on using “collective judgment” when discussing the effects of climate change (2). However, little or no documentation is provided for how judgments were reached or whose estimates were reflected. In discussion of mitigation measures (3), the TAR did not report any analysis using these concepts. The TAR states that many hundreds of scientists contributed to the report. In the absence of documentation, readers could easily conclude that reported likelihoods represent a consensus among them (7). This is not necessarily the case (8). Many of the scientists listed as contributors were never consulted about these probability judgments.