Madrid 1995: The Last Day of Climate Science (Part II)


MADRID 1995: Tipping Point?The Quest (Part II)—The Last Day (Part II)


Continuing from the Last Day The Last Day (Part I)

It is not over yet. We pick up the story again on the last day in Madrid. Yes, Al Sabban has lost the battle to base the D&A section of the Summary on the conclusions of Chapter 8. But now the approval process begins on the re-write of this section. Santer’s draft will now be debated line-by-line, word-by-word, and this debate continues through to the afternoon and into the night. Once again it is the new pattern studies giving the human ‘fingerprint’ that will be most resisted.

Ceramic banner created by Artigas from a design by Miró above the entrance to the Palacio Municipal de Congresos de Madrid

A shambolic Victory of the Virtuous

The Side Group’s redraft of the D&A section of the Executive Summary lists three key areas where recent results contributed to positive attribution.1 The first is the proxy data giving that the 20th century is the warmest century in the last six. The second is the statistical significance of the warming trend in the global mean temperature suggesting that it is unlikely to be entirely natural in origin [see SAR Fig.8.3]. This evidence on global mean temperature is well-known to be weak, continues to be expressed in the negative, had already been challenged by Australia, and it had previously been considered by the IPCC (in the 18Apr95 drafts and in the First Assessment) unable to provide positive support to the human attribution claim [see here]. As if to emphasise this, the third and final area of research, the CO2+ Sulphate pattern correlation studies, is introduced with the words ‘More convincing evidence:’

More convincing evidence for the attribution of a human effect on climate is emerging from pattern-based studies…1

These studies showed ‘pattern correspondences increase with time,’ as would be expected with increasing emissions, and there is a very low probability that ‘these correspondences could occur by chance as a result of natural variability.’ As we noted previously, the exclusion of ‘chance’ or ‘accidental’ variability implicitly leaves open the possibility of the standard century-old candidates for natural external forcing.  Perhaps it was to allay concerns about natural forcing (previously expressed in the commentaries, and so they are likely to have re-emerged in the Side Group) that the next sentence makes a curious reference to the ‘vertical’ pattern studies—as per our ‘Mirror in the Sky’—as also ‘inconsistent with the possible effects of known solar and volcanic forcing.’2

For this claim to be proposed for such a peak summary is curious because exclusion of such natural external forcing is not a major claim of Santer’s studies, nor of the other pattern studies, and no such conclusion is drawn in the Chapter itself. On the contrary, the Chapter repeatedly makes reference to the problem that we really don’t know what the pattern of nature forcing looks like. The best it can say right at the end of the section titled ‘Progress since IPCC 1990′ is that ‘we have now started to see pattern-based studies’ (which are not the flagship ones by Santer) that ‘try to rule out various non-anthropogenic forcing mechanisms.’ Thus, once again we have a situation where a claim is introduced during the inter-governmental negotiations that is contrary to the underlying scientific report, and nor is it derived from the ‘new evidence’ introduced to those negotiations in Santer’s extraordinary presentation.

Anyway, the more general problem of fidelity with Chapter 8 is that it gives CO2+Sulphate pattern evidence not as ‘more convincing’ but as ‘preliminary:’

Although these studies have very large caveats, they have yielded preliminary evidence for the existence of an anthropogenic effect on climate’2

In the end, after Madrid, to lessen the inconsistency between Chapter and Summary, Santer would change the Chapter from ‘preliminary’ to ‘initial’ [p416b]. But it is interesting to note that when the Third Assessment came to review the findings of the Second Assessment, it assessed pattern studies to have been then ‘still in their infancy.’ [TAR p701]  This is interesting to consider along with the Barnett et al paper, with the Concluding Summary, and with the fact that none of the studies providing this evidence had been published. All considered, it is hard not to be persuaded that the position taken by Saudi Arabia and Kuwait is the right way to go—namely, that the Summary should follow the Chapter in giving this evidence as ‘preliminary.’

Back in the review of the 18April95 draft the US Government was right with the Arabs on this point. While already pushing for a positive attribution statement (on both global mean temperature and CO2+Sulphate), the US Government repeatedly expressed concerns about the use of yet-to-be-published sources, it called for the CO2+Sulphate results to be presented in a manner consistent with their ‘preliminary nature’ and it said that ‘it is clear that we are only at the early stages of understanding’ their consequences.3  Again, this points to the importance for the attribution push in Madrid of the idea that Santer is presenting ‘new evidence’—but not just ‘preliminary’ evidence, it must be ‘more convincing’ new evidence if it is to provide sufficient grounds for a positive attribution claim.

‘Preliminary’ versus ‘more convincing’—neither side would budge. So the conference once again reaches a stalemate. In his short memoir of the meeting, Houghton passes over the importance to this debate of the new evidence when he tells how…

For an hour and a half, the meeting debated the appropriateness of “preliminary” instead of “convincing”. All other delegates who spoke argued that the evidence in Chapter 8 did not warrant the description “preliminary”, and a sentence was finally drafted that began “More convincing recent evidence…”

The stalemate is only broken with agreement on an extraordinary measure to indicate a failure of consensus. It is decided that a footnote will be inserted explaining the dissent of the Arabs, but lingering bitter feeling about this outcome would re-emerge in a ‘sharp exchange’ as the meeting was finally closed.

Now, at last, the discussion can move on to the ‘bottom line’ attribution claim:

Nevertheless, the balance of evidence now suggests that the global climate system is being affected by human activities.4

The proposed title of the D&A section condenses this statement with the removal of ‘suggests’ giving it more force:

The balance of evidence points towards a human influence.

It is now evening on the final day. The scheduled meeting close time of 6pm had slipped by with progress in the approval process nothing short of pitiful. Much of the Executive Summary remains to be approved before discussion can begin on the main body of the Summary for PolicyMakers. Moreover, the underlying Report itself still has to be ‘accepted’ by the conference before the night is out. Meanwhile, as this fight over the D&A section drags on and on, delegates are already starting to leave so they can make their flights home. This invites calls to close the conference: ‘at various stages the Saudis went close to forcing the abandonment of the meeting for lack of a quorum.’  With the combined IPCC meeting in Rome less than two weeks away, there is no realistic prospect of re-convening the Working Group 1 Plenary. The fight for positive attribution is pushing the conference to the brink of complete failure with approval of a Summary for PolicyMakers in any form starting to slip beyond reach. But still the all-important attribution bottom line has not yet made it over the line. Here, the Australian Delegation Report sets the scene:

Most delegations supported the adoption of the proposed text with several (especially the US, UK and Canada) arguing for stronger language. Dr [Robert] Watson wanted a statement to the effect that the “preponderance” or “weight” (rather than “balance”) of evidence “indicated” (rather than “suggested” or “pointed to”) a human influence on global climate. Dr Stone from Canada felt the evidence was “stunning”. Another smaller group of delegations lead by Saudi Arabia (with informal support from US industry NGO’s) sought to weaken the statement and increase the emphasis on the uncertainties involved in attributing the observed changes to human activities.

Eventually agreement is reached on the following wording:

(Nevertheless) the balance of evidence (now) suggests an appreciable human influence on global climate.

With a weak attribution statement at last in the bag, the conference can move on to the next section of the Executive Summary. However, the situation remains precarious, for even after achieving this prize, it could still be lost with the failure to complete the review and approve the whole Summary. With the clock passing 9pm and the numbers dwindling, most of the Summary for PolicyMakers remains untouched. Pausing to discuss how to proceed, the situation seems hopeless, until there is a breakthrough.

It is decided to take the Executive Summary for the Summary for PolicyMakers and to recast the rest of this Summary as a ‘Technical Summary.’ Now as the ‘Technical Summary’ the main body of the Summary for PolicyMakers would only need ‘acceptance’ by the Plenary (as with the chapters) and not the line-by-line ‘approval’ for which there is now no time. This strategy is accepted and the conference proceeded with approval of the rest of the Executive Summary, now at a pace and with little or no discussion.

But…Oh, No! The question of attribution had dominated the entire conference and well into extra time, yet it still would not go away. There are rumblings across the floor as a new crisis emerges. In fact, the ‘bottom line’ attribution claim is not in the bag.

Dissatisfaction with the outcome had arisen on both sides of the debate. The Australian Report continues:

Unfortunately in achieving a consensus, the Chairman did not read out the bracketed words (which he had taken as “given” from the earlier text) and a number of delegations soon made clear that the text the Chairman said had been approved was not what they thought they had agreed to. Dissatisfaction also began to mount over the term “appreciable” which had been proposed by one of the Lead Authors (Trenberth) and strongly supported by the US delegation.

Apparently the bulls were not happy with the inclusion of the bracketed words, and, for the bears, ‘appreciable’ was too strong. Bert Bolin had been moving around the room consulting with various delegations in an attempt to find a resolution. Finally, at 10.30pm he interrupts proceedings, ‘took over the meeting’ and declares that…

…he had decided, as an extraordinary measure, to overrule the agreed text because of the extreme importance of the wording to the way the IPCC findings would be interpreted. He said he did not wish there to be any discussion but he believed the meeting would accept, as the ‘bottom line’ on detection and attribution:

‘Nevertheless the balance of evidence suggests that there is a
discernible human influence on global climate.’

No one dissented from this ruling and so this is how this famous line came into being. For Zillman this is more than another admission of the failure to reach accord. He is surprised and taken aback by the decisive way Bolin resolves a matter of no small import to the whole assessment process.

When the attribution bottom line is shoved back in the bag it is fast approaching midnight, and while the interpreting team and hall management staff could keep the conference going until then, the building would need to be vacated soon after. It is now a real scramble for the finish. The Executive Summary approval is hurriedly wrapped up during the final hour while delegations pack up and leave. Past midnight, with minutes to go, the only thing left to do is to ‘accept’ the underlying chapters. This would not have been of much concern if it weren’t for the (now even more glaring) inconsistencies between (the body and conclusion) of Chapter 8 and (the newly drafted) D&A section of the Summary.

The story goes that this Working Group 1 Plenary did indeed give consent for the Lead Author to revise Chapter 8 according to the consensus it had finally achieved. For example, here is Houghton again in a ‘Justification of Chapter 8′:

The plenary meeting finally ‘accepted’ the draft chapters (including Chapter 8) subject to their revision by the lead authors to take into account the guidance provided by the meeting and in particular the need for overall consistency. [Nature 382, 22Aug96].

How explicit was this acceptance and how specific was this guidance is hard to establish. The matter would not be raised at Rome (where only the 9Oct95 draft appeared) and precious little has been obtained recording or discussing the problem before we come to the justifications proffered months later when the controversy broke. But anyway, perhaps it doesn’t matter what was actually said or not said, agreed or not; for, by all accounts, at this stage the meeting had degenerated into a shambles.

The Australian Delegation Report was completed and available before the full IPCC conference in Rome two weeks later (Zillman recalls drafting it on the plane home) and this does give a brief account. It suggests that while on the night it was evident that Chapter 8 would need changing, both the authority and process remained unclear:

In the closing minutes of the meeting, Sir John Houghton asked for formal acceptance of the complete text of the underlying report. This was agreed, although it was not clear what licence the Lead Authors were given to revise the individual chapters in the light of the discussion at the Session.

In fact, as we saw earlier, Santer and Houghton would later explain how licence had already arrived with the collated comments from the US State Department. Changes were requested and changes were made.

Ceramic banner created by Artigas from a design by Miró above the entrance to the Palacio Municipal de Congresos de Madrid


The former Meteorologial Office in Bracknell, England

The Met Office in Bracknell, England (now demolished)

After Madrid, Santer flew directly to the Working Group 1 Technical Support Unit in the Met Office, Bracknell, England. There Santer worked on revising Chapter 8 with the head of the Unit, Bruce Callendar. When the changes were complete he left the new version with Callendar, as it was Callendar’s job to prepare the Summary and Chapters for publication. Nothing is then heard of this version of Chapter 8 until the following May.

Meanwhile, because of the lateness of the Working Group 1 Plenary, before it even began in Madrid the draft Summary for PolicyMakers and the ‘final’ [9Oct95] version of the underlying chapters had already been circulated to the delegations for the full IPCC meeting in Rome. It was expected that a new SPM would come out of the approval process in Madrid and this was duly and very quickly made ready for Rome. Speed was vital because the approved Summary would be used along with the Working Group 2 & 3 Summaries to inform the approval of the already drafted ‘Synthesis Report.’

No new version of Chapter 8 appeared in Rome when Houghton proposed the acceptance of the full Working Group 1 Report. Nor is any new version of (or amendments to) Chapter 8 mentioned in the minutes of Rome. What is mentioned is that Saudi Arabia and Kuwait volunteered the removal of the ‘preliminary evidence’ footnote from the Summary, thus allowing a complete consensus to be achieved.

It was in May 1996, after obtaining a copy of the new Chapter 8, that the Global Climate Coalition initiated the controversy over the changes. Demands to withdraw the new version from publication and revert to the old version went unheeded. The entire report was published in June.

In 2008, when the Chapter 8 Controversy had long ago subsided and was mostly forgotten, Sir John Houghton re-lives it with the short memoir in Nature.  Here there is no sense of regret and no small sense of pride about his role in that meeting that changed the world. The article opens with: ‘Climate change is increasingly recognized as one of the most serious threats to humankind. But it was not always so.’ The IPCC, claims Houghton, was vital in changing this attitude. The article concludes with:

In Madrid in 1995, the IPCC scientific assessment process, based on the findings of the latest research, was sorely tested. Had the science not come through unscathed, the integrity of the panel would have been seriously questioned, and governments would have faltered on taking urgent action on climate change, such as the signing in 1997 of the Kyoto Protocol.

Without success at Madrid there would have been no Kyoto. For Houghton, it was a tough job in a tough meeting, but it was a job well done.




Many people have generously assisted with research on Madrid 1995 and the Chapter 8 controversy. The author is especially grateful to the following: Dr Ben Santer for confirming facts and offering referrals; Sir John Houghton for careful responses to many question; Dr Tim Barnett for assisting with understanding the science of the controversy; Dr Mike MacCracken for providing unpublished documents and responding to many queries; Dr Vincent Grey for assisting with the search for documents; Dr Fred Singer for sending his writings on the topic; Dr John Zillman for his recollections and analysis; Lucy Skywalker and Andrew Montford for advice on this narrative; and Dr Aynsley Kellow for political background.


Any errors of fact in the posts are by the author.  Notification of corrections are encourage and they will be applied (using strike through) and acknowledged as soon as they are verified.

Related Posts

This post continues The Last Day of Climate Science Part I.

The order of the narrative is as follows:

  1. Madrid 1995: Was this the Tipping Point in the Corruption of Climate Science?
  2. Madrid 1995 and The Quest for the Mirror in the Sky (Part II)
  3. Madrid 1995: The Last Day of Climate Science (Part II)



1.  The D&A Side Group’s redraft of the D&A section of the SMP Executive Summary is Appendix I of the Australian Delegation Report. The document is titled Revision of Executive Summary of SPM:11/28/95 : Executive Summary of Summary for Policymakers: Detection Section.
2.  Note that the Plenary ended up agreeing to remove the ‘possible’ from possible effects, thus strengthening the claim. We presume that ‘possible’ was in the original claim because the real data on natural forcing is unavailable and so this claim relies on effects anticipated by the models. ‘Possible characteristics‘ is used in this way in the second paragraph of the Chapter 8′s introductory ‘Summary’ with reference to internal variability.  Even in the published version of this Summary there is no equivalent claim on solar or volcanic forcing [9Oct95 p8.1 & SAR p411a]. The quotes in the text are drawn from the two concluding sentences of section the 8.1.3 ‘Progress since IPCC 1990′ in the 9Oct95 draft, which reads: ‘Although these [pattern] studies have very large caveats, they have yielded preliminary evidence for the existence of an anthropogenic effect on climate. Furthermore, we have now started to see pattern-based studies that directly address the attribution question, and try to rule out various non-anthropogenic forcing as explanations for some observed pattern of climate change (e.g. Hansen et al, 1995b; Karoly et al, 1994).’ [page 8.4, compare with SAR p416b]
3.  Source: US Comments on the IPCC WG I Second Scientific Assessment—Summary for policy makers, United States Department of State, July 10, 1995.  See page 2 of the cover letter and page 2 of the General Comments.

4.  The 9Oct95 SPM draft had ‘Taken together, these results point towards a detectable human influence on global climate‘ and the US Government comments [15 Nov, by Robert Watson] suggested this change to: ‘Taken together these results indicate a detectable….



7 thoughts on “Madrid 1995: The Last Day of Climate Science (Part II)

  1. I just found your blog through Bishop Hill. Your account of Madrid is fascinating and highly readable. I will have a dig through your older posts as well.

    Do you have a policy regarding reblogging of posts / translation and reposting?

  2. A fascinating bit in the whole story is how deputies within the Clinton administration were in a position to conduct the entire story : a vice president, a deputy secretary of state.

    With the president not a strong supporter of the global warming alarm, and fully aware that no majority @ congress would ever vote any Kyoto Convention…

    Did you investigate this side of the story ?

  3. Thanks Daniel and no I did not.

    Politics is not my stong point and I know less about US politics. While they seem to have been very influential, I am yet to reach the conclusion that these deputies ‘conducted the entire story.’ (But aren’t similar situations common, like with Rumsfeld pushing on Iraq?)

    One political aspect that is curious, and I might investigate, is some sort of (senate?) inquiry into the Chapter 8 Controversy. I know this happen. Mike MacCracken has sent me some docs prepared for it. But I have yet to come across any account of it.

Join the discussion

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

You are commenting using your 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