Fallibility and Reason

February 14, 2010

Looks like there’s one more tail to pin on the IPCC donkey, and you can bet that the usual suspects will make a great to-do about this — that is, when they’re not busy seeding other storm clouds.

The IPCC has said in response that “errors in the 2007 report of about 3,000 pages do not affect the core conclusions that human activities, led by burning fossil fuels, are warming the globe.” (That’s a quote from the article, not directly from the IPCC.)

Their response, of course, is probably correct. That there are flaws in the document should come as no surprise. The conclusions of the report don’t hang on any single premise; but rather on the abundance of evidence and the strength of inferences between those bodies of evidence.

The evidence, I am told, is substantial, despite a few flaws here and there.

Oh, spare me the “but, but, buts…” Sure, I can even accept that the flaws weren’t just here and there, but that there have been multiple flaws, and some on the IPCC have taken political steps to downplay the importance of correcting those flaws. That’s a non-insignificant political matter, and maybe one to sort out with Roger. Addressing these concerns deftly and appropriately will help to support the body of research in the mind of those who don’t have much of the information.

No. One flaw will not undo the IPCC report. Why not? Because just as weather ain’t climate, so too is data or evidence not science.

What holds the abundance of existing evidence together is the strength of the reasoning. And, as it happens, the conclusions of the IPCC currently offer the best explanation of the body of evidence. We’re not looking here for mere alternative explanations of the evidence. Such theories would be useless. Plus, there are an infinite number of plausible explanations that can explain the evidence. Maybe the Great Hippo of Sydney is tinkering with our instruments, for instance. No, we’re looking for the best explanation of the evidence; and that is what the IPCC offers.

If the skeptical community truly wants to take down the IPCC bear, they’re going to have to do a lot more than tackle the evidence. They’re going to have tackle the reasoning, and they haven’t done that yet. They may be chinking the armor in the scientific community, but if they’re winning any battles they are winning only public and political battles; battles in the minds of the lumpen proles who, when aggregated around a voting booth, determine how we respond to the science.

Don’t get me wrong, that’s a big friggin’ deal. If we hope to make any progress with the policy, we’re going to have to handle those political issues as well. As a little note to the folks at RealClimate and elsewhere, I’m not at all persuaded that leaning into the science alone will do the trick. It’s only one part of a very big job.

Setting the right policy is a monumentally pluralistic undertaking. I think that’s gotta come through a very wide public discourse about not just the science, but also the nature of the science, the uncertainty associated with the science, and the reasoning behind the policy responses.

As a good little Habermasian, I’m a fallibilist at heart. I don’t mind a few lumps in my oatmeal. I just want to make sure that what I eat is nourishing. Acknowledging fallibilism, I’d think, would help the IPCC maintain the high ground.

Happy Valentine’s Day everyone.



  1. So the sea level error turns out to have little substance, leaving us with the Himalayan glacier flub as the only real mistake. Eh. When all is said and done about both the IPCC and the CRU, it doesn’t appear likely that there will be anything else to point to. Even the much-ballyhooed FOI transgression is looking pretty baseless these days. Hopefully the press will be suitably chastened, although the lack of introspection in the aftermath of coverage of the Iraq war build-up doesn’t give me much hope in that regard.

    • So would a true believer such as Bloom be an infallibilist?

  2. RC has the definitive round-up on the IPCC business.

    • Nice. Hadn’t seen that. Thanks for the tip.

  3. The Himalayan glacier issue was major because it was a. a mistake and b. could have affected policy. The Amazon one was not even minor because it was a. correct (not a mistake) and b. should effect policy.

    FWIW, Peter Cox has been going on for over fifteen years in the published lit about drying out of the amazon.

    • Eli:

      Another reason the Himalayan glacier issue was major was because when it was brought to the attention of the President of the IPCC RK Pachauri, he attacked the person raising the mistake as practicing “voodoo science”. This is a major problem for the credability of the IPCC, and therefore its reports.

      Personally, the data suggest to me that it has been warming, but the reason it is warming and whether it will continue are not as clear.

      Only time will tell. If the 30 year warming trend is broken in the next 10 or 15 years or so, it will really cloud the issue of where temperatures are heading.

  4. I guess my attitude is that we can’t expect any document to have no errors. These sorts of things have errors. Sometimes they’re big errors. Sometimes they’re little errors. But errors there are.

    How are we to know what to make of the errors? We reason past them.

  5. The more important questions raised are ones of process. Quoting here from Mike Hulme, professor of climate change in the School of Environmental Sciences at the University of East Anglia, Norwich, United Kingdom and the founding Director (2000-2007) of the Tyndall Centre for Climate Change Research:


    “The incorrect statement in the Fourth Assessment Report of the Intergovernmental Panel on Climate Change (IPCC) that the Himalayan glaciers could completely disappear by 2035 is remarkable in many ways.

    “First, how could such a physically implausible claim have entered an early draft of an assessment undertaken by ‘the world’s leading experts’, as IPCC authors are frequently described?

    “Second, how did the claim survive several rounds of peer review from other IPCC authors and outside experts?

    “Third, how did the claim, published in April 2007, remain unchallenged for more than two years before hitting the news headlines?”

  6. Hi, Ben: I think you’re right that the errors thus far identified in the 4th IPCC report are insufficient to lessen the confidence in the IPCC’s conclusions of anyone willing to undertake the burden of reasoning through. In a complex (a/k/a “conductive”) argument, knocking out a few lines may not significantly weaken confidence in the conclusion.

    However: The IPCC’s report for many of us is not an invitation to work through its 3000 pages of reasoning. Given limited resources of time and attention, we are apt to take it–and I think *invited* to take it–as an authoritative statement. We trust the report, without working through all its reasoning, because we *trust* the scientists and their assessment process. The errors in the report and the CRU emails seem to me to provide prima facie reasons for doubting the trustworthiness of some of the scientists and their process. And if these are untrustworthy, we can’t trust the report.

    A Habermasian might not like it, but trust in others’ reasoning allows the public sphere to operate under real-world conditions such as limited time and attention. Maintaining trustworthiness may require of scientists (and others) more than ordinary epistemic virtue.

  7. Yes, yes. I think I agree with you, Jean.

    I’m not suggesting that we’re required to read through the full 3000 pages… but that the IPCC thesis hangs primarily on the reasoning, not strictly on the evidence. In fact, I might even go so far as to suggest that the evidence is just blind data.

    And, absolutely, trust is a key issue with any authoritative and expert statement. This is why I think it’s important for IPCC authors to assume the stance of fallibility. I think Bloom, above, illustrates the contrary position — the position often taken by those defending the IPCC.

    Much as I often agree with and am thankful for Bloom’s input, this is a case where rigid insistence on infallibility sets a trap for the conclusions of the report.

  8. Well, neither Jean nor Ben (apparently, since he agrees with her!) can be bothered to read the full report, but are happy to jump to conclusions about the IPCC based on the “prima facie” evidence of IPCC and CRU errors. (Wasn’t Sen. McCarthy happy to go with prima facie evidence?) As I pointed out above, there is evidence of the non-prima facie variety for only a single real error by the IPCC and none by CRU.

    I would add that the IPCC has summaries written for non-experts like Jean, and even an overall summary to make it really painless. But Jean hasn’t read even the latter, preferring instead the prima-facie evidence proffered by wingnut blogs and the tabloid press. Were she ever to read either the WG2 summary or the overall one, she wouldn’t see the 2035 figure since it wasn’t carried over. That’s peculiar for such a striking conclusion, don’t you think?

    Ben, re assuming a stance of fallibility, to the extent that I’m clear on what that means I don’t think it would make a speck of difference to the AGW policy debate. I’d be interested to see your case for how it would, though.

  9. Looks like Robert Muir Wood Gate just closed and John Houghton Gate was shut earlier today. Eli expects to see a bunch of mea culpas. Eli is always disappointed.

    The amount of misleading nonsense posted without any apology by so called “serious” people makes it difficult to afford them any respect that they pretend to deserve. In the case of Muir Wood, it was an obvious bowl of spinach because Muir Wood was a contributing AUTHOR to the chapter he was supposed to have been misquoted in.

  10. […] few posts ago I raised the topic of fallibilism and proposed that the IPCC and its proxies (like the fine fellas at Real Climate) would do better […]

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: