Consensus, Truisms and Straw Men

In a recent op-ed published in the Washington Post, science historian Naomi Oreskes, elaborating on her essay for Science magazine, argued that the nation's leaders were ignoring a unanimous agreement in the scientific literature that man is responsible for global warming and that something must therefore be done about it. Yet an examination of the form the much-touted scientific consensus actually takes reveals that it does not mandate policy choices. Moreover, the charge that people are denying what Orsekes defines as the consensus appears to be a straw man. It is therefore worth asking what the point is of this argument, which is growing increasingly popular.<?xml:namespace prefix = u1 /> <?xml:namespace prefix = o ns = “urn:schemas-microsoft-com:office:office” /> 

What do scientists mean when they talk about the “scientific consensus on climate change”? The answer is helpfully provided by the new web log set up by a variety of climate scientists entitled There, British Antarctic Survey scientist William Connolley defines the consensus in these terms:


“The main points that most would agree on as 'the consensus' are:


1.       The earth is getting warmer (0.6 +/- 0.2° C in the past century; 0.1° C/decade over the last 30 years)

2.       People are causing this

3.       If GHG emissions continue, the warming will continue and indeed accelerate

4.       (This will be a problem and we ought to do something about it)”


Connolley also includes the following important rider:


“I've put those four points in rough order of certainty. The last one is in brackets because whilst many would agree, many others (who agree with 1-3) would not, at least without qualification. It's probably not a part of the core consensus in the way 1-3 are. Most…of us here on RealClimate are physical scientists—we can talk sensibly about past, present and future changes in climate, but potential impacts on ecosystems or human society are out of our field.”


This is a useful summary, because it enables us to see where the disagreements lie. Point 1 is generally accepted, although the fact remains that satellite temperature measurements show a smaller warming trend and the reasons for that remain a topic of genuine scientific debate. Nevertheless, there is general agreement that the world warmed slightly over the past century.


Point 2 is rather imprecisely worded as everyone agrees that temperature changes over the last century have been affected by a variety of human and natural effects, both warming and cooling. The idea that man has not affected the climate in any way has virtually no supporters. Roger Pielke, Jr.—no skeptic—of the <?xml:namespace prefix = st1 ns = “urn:schemas-microsoft-com:office:smarttags” />University of Colorado compiled a list where he demonstrates that all the so-called skeptics, including Fred Singer, Pat Michaels and even President Bush, have accepted that there is an anthropogenic influence on climate. The claim that opinion-formers deny this is a classic straw man.


Yet all scientists agree that there is more than just one form of human influence. As well as greenhouse gases, land-use changes, aerosol concentrations and other “forcings” have a role to play. At the time of the last IPCC report, we knew a lot only about the role of greenhouse gases (see figure 9 here), but we have invested a lot of time, money and energy into finding out more about the other forcings. They have enabled scientists to declare that such factors as land-use changes and black carbon (soot) concentrations may account for large portions of the recent warming. Moreover, we now know more about natural forcings such as the oceanic phenomenon known as the Pacific Decadal Oscillation which some researchers think may account for half of the recent warming trend. This is an area of genuine ongoing scientific discovery.


Point 3 is more contentious, as it relies on theories that assume that there is a so-called “positive feedback mechanism” in the atmosphere that will accelerate any warming trend. This is where the so-called skeptical scientists part company with the consensus. MIT Professor of Meteorology Richard Lindzen, for instance, is well-known for having advanced a credible, peer-reviewed theory that the Earth has an infrared “iris effect” that will produce negative feedbacks. Recent NASA research indicates that feedback mechanisms are not as pronounced as climate models suggest. This is again an area of ongoing scientific discovery, yet the genuine disagreement here would not have shown up as dissent in Oreskes' research as she actually defines the consensus as “that Earth's climate is heating up and human activities are part of the reason” — in other words, she defines the consensus as points 1 and 2 of Connolley's definition, which, as we have seen, are not really in question.


Yet the reason for Orsekes' principal complaint—that we are not doing anything about global warming—can only stem from point 4, which as Connolley says does not really form part of the core consensus and in fact lies in many aspects outside the realm of science. Indeed, one of the commentators on Connolley's post points out that there may well be a fifth, economic component to the consensus, “that global warming may be bad—but it is NOT as bad as what it would take to prevent it.” Connolley accepted this as perfectly valid, and it is backed by economic analysis exercises such as the Copenhagen Consensus which found currently proposed mitigation measures like Kyoto to be poor investments of the world's resources.


Orsekes has therefore cheerfully elided a genuine consensus on points 1 and 2 of Connolly's definition into an assertion that this mandates policy action. It can do no such thing. Science only alerts us to possible problems and potential solutions; it is the job of economics, within the political process, to determine whether action should be taken and if so, which of the potential solutions science has identified should be chosen (and even then, we may choose not to adopt the whole solution).


So if Oreskes' work is based on a false premise, as it seems to be, does it have any other worth? It may be said that it is useful that she has demonstrated a consensus exists. This is made problematic by the fact that Orsekes has since admitted that she looked at only about 1,000 scientific abstracts out of 11,000 relevant articles (and the question of whether analyzing abstracts gives a true reflection of the nuances of the full article remains open).


Yet even if we take her result at face value, it is really only to be expected. We have known since Thomas Kuhn's masterpiece, The Structure of Scientific Revolutions, that at any one time in any science there exists a consensus—the paradigm, as Kuhn termed it. That one should exist even in a relatively new discipline such as climatology is unsurprising. In the end, Oreskes is presenting a truism as evidence against a straw man. That's no way for scientific debate to advance.