On Bjorn Lomborg and climate change
Bjorn Lomborg’s chapter on global climate change is a clever polemic; it seems like a sober and well-researched presentation of balanced information, whereas in fact it makes use of selective inattention to inconvenient literature and overemphasis of work that supports his lopsided views. The Intergovernmental Panel on Climate Change reports and other honest assessments don’t have the luxury of using such tactics, given the hundreds of external reviewers and dozens of review editors.
It would take several pages to document how Lomborg lines up his citations to diminish the seriousness of climate effects while ignoring most literature that would stress the seriousness. (For that kind of documentation, see a review by my colleagues and me in the forthcoming January 2002 issue of Scientific American or Stuart Pimm and Jeff Harvey’s review in the Nov. 8, 2001 issue of Nature.) Lomborg does acknowledge an aggregate $5 trillion benefit of controlling and minimizing climate change, but then contrasts this to an estimated cost of controlling global warming of “from $3 to $33 trillion.”
Note that Lomborg offers a wide-ranging estimate for how much it would cost to control climate change but only one figure for how much the climate change itself would cost us. In reality, the cost of climate change itself is generally considered — by the very economists whom Lomborg quotes for costs of control — to be much more uncertain than the cost of controlling climate change. In other words, this putative statistician quotes a range of costs when convenient but not a range of benefits when inconvenient. Neither does he tell us, as any assessor should — let alone a statistician writing a popular book! — that these are very crude estimates grounded in subjective assumptions at every stage. To imply that the costs are empirically determined is to completely misunderstand the situation, or misrepresent Bayesian statistics (subjective) as frequentist probabilities (objective).
Moreover, one of the reasons the IPCC Working Group 2 did not put much credence in the aggregate climate damage numbers — a move Lomborg decries as politically motivated — is because the benefits of avoiding climate damage must be measured in more than commodities traded in markets (that is, money per ton of carbon abated). There are also human lives lost, species lost, distributional effects, and so forth. That makes putting a price on climate change very difficult, so to represent this by one number is a gross misunderstanding of the assessment process — or a deliberate polemic. You decide which.
Meanwhile, Lomborg’s analysis does not even address the cost that climate change will exact on nature, despite the media clamor that ensued after the Working Group 2 concluded that the monitoring data show a “discernible” impact on plants and animals due to recent temperature increases. Plants and animals breeding earlier or altering their ranges didn’t make the Lomborg climate discussion, but fertilization of crops with carbon dioxide (considered a “positive” outcome of global warming) certainly did — a good example of the one-way selective filter that Lomborg uses throughout the book.
Finally, Lomborg asserts that most of the six representative emissions scenarios from the IPCC Special Report on Emissions Scenarios (SRES) are implausible and that only their lowest emissions scenario is “likely.” The scenarios Lomborg attacks were prepared in three drafts over the course of three years by the SRES team, and were subjected to three rounds of reviewers. So on what grounds does this statistician who does not conduct environmental research, who does not even explain subjective versus objective assessment, and who uses ranges of estimates inconsistently and in a biased manner declare specific SRES scenarios with high emissions “fairly implausible” but those with low emissions “much more likely”? Note that the dozens of authors and hundreds of IPCC reviewers who assessed the emissions scenarios recognized the tremendous uncertainties involved in constructing such scenarios and judged all scenarios to be “equally sound.”
Now, as you can see in a piece I authored in Nature (May 3, 2001) — and in the Wigley and Raper or Reilly, et al. pieces in Science in the summer of 2001– we’re all for trying to estimate subjective probabilities for scenarios so that a risk tradeoff can be made. But first, analysis please! From Lomborg we get selective quotes from a tiny fraction of the literature and no balanced discussion of the wide range of available studies. How dare this guy announce the likelihood of various scenarios to which the SRES authors are very uneasy about assigning probabilities? Worse, how can he do so on the thinnest reed of literature citation and superficial analysis? Moreover, it is my understanding that most of these criticisms were leveled at Lomborg when the shorter Danish version of his book was published — yet the same problems haunt the English edition.
On the Media
The real travesty is that the mainstream media have quoted The Skeptical Environmentalist as if it contained something new — some original analysis the rest of the community had missed, or some more balanced assessment. The sooner Lomborg’s own unbalanced and incomplete “analysis” is exposed, the better we will all be. But will the mainstream media notice corrective coverage in scientific journals such as Scientific American and Nature, or in alternative venues such as Grist?
The Skeptical Environmentalist also raises questions about the role and the responsibility of the publisher. Why did Cambridge University Press, a publisher with so excellent a reputation in the natural sciences (it even published the IPCC reports), publish a polemic under its imprimatur? Did CUP have the book competently reviewed? Why did only the social science side of the shop handle this book, given that it is mostly grounded in badly garbled natural science assertions? Did the opinion of economists or the bottom line trump all other considerations? (The book has been a big seller on Amazon.com.) I hope there will be some soul-searching at CUP, as well as an internal investigation of its assigning this multidisciplinary topic to only one disciplinary branch.
What a monumental waste of busy people’s time countering the scores upon scores of strawmen, misquotes, unbalanced statements, and selective inattention to the full literature. If Lomborg at least had spent time in meetings with the people who really do debate these issues, this book might have been a useful contribution to the field. But for a non-participant like Lomborg to drop in flaunting a flimsy Greenpeace connection (the group denies he was a significant member, incidentally) and using sheer volume of citations — most to secondary sources and many to the same pieces over and over again — as a way to feign serious scholarship and thus get serious attention almost defies imagination.
It is incumbent upon those of us who remain committed to sound science and good policy to point out the large amount of disinformation in Lomborg’s pretentious claim to be “measuring the Real State of the World.” Those of us who have spent decades grappling with the numbing uncertainties involved in environmental protection would never claim to know the real state of the world, let alone pretend that selective citation to fuzzy historica
l data would tell us.