2.4 The meaning of 'consensus': peer review and the IPCC process
At the time of writing (2006), debate about the 'hockey stick' reconstruction continues to rumble on. In this and other controversial areas, it is natural that scientists who are not part of the IPCC process should scrutinise its assessments and continue to ask probing questions about its conclusions. At the same time, however, it's important to keep claims that run counter to the mainstream view in perspective - and to bear in mind that there may well be a political agenda behind the selective promotion of such claims. In the US, for example, Congress had already refused to ratify the Kyoto Protocol before the Bush Administration took office. A sustained campaign, stressing the uncertainties in the science, by the notorious and now largely defunct Global Climate Coalition (a business NGO comprising several large multinational fossil fuel companies) is credited with having played an important role in that decision.
The IPCC's remit is to analyse and evaluate the existing peer-reviewed literature, pertinent to the many scientific, technical and socioeconomic aspects of human-induced climate change. This huge interdisciplinary task depends on the work of thousands of collaborating natural and social scientists - a significant proportion of the academic community engaged in climate change-related research. To put the sensitivity of the IPCC's role into sharper focus (Edwards and Schneider, 2001):
As a hybrid science-policy body, the IPCC must maintain credibility and trust vis-à-vis two rather different communities: the scientists who make up its primary membership, and the global climate policy community to which it provides input […] The IPCC's rules of procedure spell out a variety of methods designed to ensure its reports include the best available scientific knowledge and that they represent this knowledge fairly and accurately. Chief among these is the principle of peer review, traditionally one of the most important safeguards against bias and error in science.
As far as the peer review of scientific literature is concerned, scientists write articles (papers) and submit them to a journal. The journal editor sends the paper to several referees, all of them experts in the authors' field (i.e. their 'peers'). Referees can typically chose one of three recommendations: acceptance, rejection or acceptance after certain specified changes are made. The third option ('revise and resubmit') is by far the most common. The process usually goes back and forth a few times, with several rounds of revisions, until an acceptable compromise is achieved.
This highlights one of the perceived problems with peer review; different referees can come up with radically different conclusions about the merits of a particular piece of work. Some commentators see this as a fundamental weakness of the whole system. Others have concluded that most reviewer differences probably result from 'real and legitimate differences of opinion among experts about what good science is or should be'.
As we said earlier, disagreement is healthy; it moves science on. But, as Edwards and Schneider go on to say:
if expert judgement varies too widely to provide a quasi-mechanical means of winnowing out bad science from good, why is peer review important? […] We maintain that peer review ought to be regarded as a [sometimes fallible] human process whose primary functions are to improve the quality of scientific work, to maintain accountability both inside and outside the scientific community, and to build a scientific community that shares core principles and beliefs even when it does not agree in detail.
This perspective on what peer review is 'for' bears directly on its role in the IPCC process. Recall that IPCC reports are not primary science, but assessments of the state of the field based on a critical evaluation of existing work. Nevertheless, draft chapters and other IPCC documents are subjected to their own peer review process. This is more open, extensive and inclusive than most, involving non-specialists (government advisers, business lobby groups, etc.) as well as expert scientific reviewers. Typically, hundreds or even thousands of changes are made as each document goes through several drafts.
This exhaustive process has played a major role in building a broad-based scientific consensus on the causes and implications of recent climate change, and in establishing the credibility of IPCC reports for policy purposes. These days, most of the world's leading climate researchers are involved in one way or another - as authors or reviewers, or because their work is used and cited. Over the years, some of the more outspoken scientific sceptics have been drawn in as well, so their views are now represented in the process that produces eventual consensus on the 'current state of knowledge'. In effect, then, the IPCC has become the voice of the expert climate science community, and is now regarded as an authority by most (if not all!) governments around the world. Its assessments are a major driving force behind international climate policy.
Yet there are critics who charge that the very notion of 'consensus science' is a nonsense, commonly citing those giants of the past (e.g. Galileo, Einstein) who have challenged and revolutionised the scientific dogma of the day. If controversy and robust debate is the lifeblood of science, the argument goes, then 'consensus' must surely be its death knell, and deeply anti-scientific.
The counter argument is that the IPCC consensus is not some unassailable 'truth'; it is simply a fair representation of the expert scientific community's current general opinion, based on the available evidence and subject to revision. Behind the public, government-negotiated and carefully crafted face of this consensus (in the SPMs) is a lot of messy and uncertain science. The highly technical bulk of each report documents limitations of current understanding, areas of disagreement, caveats about uncertainties, etc. There is no point pretending this is not the case. As stated at the outset, in a field as complex as climate change, uncertainty is unavoidable. Moreover, ongoing research may help to reduce uncertainties in some areas while at the same time uncover new sources of uncertainty elsewhere. We have already encountered one example - growing awareness of the complicated climatic effects of tropospheric aerosols - and doubtless many more will come to light in the years ahead.
To return to the original focus of this section, few researchers base their underlying concern about the build up of atmospheric CO2 on the Earth's recent temperature history. Rather, it is rooted in what might be termed the 'relentless logic' of the physics of the greenhouse effect (Section 1.5), and fuelled by the dramatic rise in greenhouse gas concentrations over the past 200 years or so (Figure 16). On the other hand, there is little doubt that the record-breaking warmth of the 1980s and 1990s has lent warnings about the 'greenhouse problem' a popular credibility they previously lacked. Activity 1 invites you to ponder on that popular perception.
The theory that we were heading into another 'ice age' was quite topical and scientifically respectable in the 1970s. Indeed, this was one of the concerns on the agenda at the first World Climate Conference, along with the prospect of greenhouse warming. Here are a couple of quotes, which give you a feel for how the issue was presented at the time:
The threat of a new ice age must now stand alongside nuclear war as a likely source of wholesale death and misery for mankind.
(Nigel Calder, International Wildlife, July 1975)
This cooling has already killed hundreds of thousands of people. If it continues and no strong action is taken, it will cause world famine, world chaos and world war, and this could all come about before the year 2000.
(Lowell Ponte, The Cooling, 1976)
(a) Look back at Figure 24a. Can you suggest why the idea that the world was headed for a cooler regime might have gained credence at the time?
(b) What devices are used in the quotes above in order to communicate the implications of'global cooling'?
(c) In the 1970s, some scientists argued that the cooling was due to expanding industrial activity. What do you think was the basis for this suggestion?
(d) What salutary lessons can be drawn from this episode that are relevant to the current debate about global warming?
(a) If you cover up the last bit of the record in Figure 24a, it's fairly easy to see how the slight cooling in the post-war years (from the early 1940s to the mid1970s, say) could be interpreted as evidence that a long-term downward trend in global temperature might be underway.
[In fact, cooling during this period was stronger in the Northern Hemisphere, and particularly marked in the well-monitored regions around the North Atlantic. This northern cooling was offset to some extent by a slight warming in the Southern Hemisphere, but this only became apparent with the generation of the first reliable records of global temperature (i.e. GMST) in the mid-1980s.]
(b) The language used conjures images of wholesale (doom and gloom), made more potent still by reference to the deepest fear of the 'Cold War' years - the threat of nuclear war. Indeed, the devices used in these quotes are strikingly similar to those sometimes used today to communicate the implications of global warming. Recall, for example, the image of 'a world riven by water wars, famine and anarchy' in the recent Pentagon report, and references to the 'threat of terrorism' by Sir David King and the Prime Minister of Tuvalu.
(c) The most likely basis for a cooling influence from expanding industrial activity is the large amount of particulate matter (sulfate and carbonaceous aerosols) pumped out by burning fossil fuels, especially coal. [See Section 2.6 for the part this has played in 20th century climate change.]
(d) This episode exemplifies points flagged up at the beginning of Section 2.1. Given evidence of an apparent, but relatively short-term, trend in global temperature, we need to be wary of jumping to conclusions - both about the significance of that trend (i.e. it needs to be set in a long-term context, the message of Sections 2.2 and 2.3), and about the underlying cause or causes (the issue taken up in Section 2.6).
[It is worth noting that this episode would continue to haunt the climate science community. In the years that followed, it was often used to cast doubt on the credibility of climate science and the emerging consensus that greenhouse warming would, sooner or later, prove to be a major factor in the Earth's climate future.]