Guest Post by Steven Mosher
All processes are subject to hacking, both from external and internal entities. The IPCC process is no exception, neither is the process of writing for science journals exempt. As recent reports show Pachauri hacked the IPCC processes to make them work against themselves and bend the system to his financial ends. An inside hacker, like Pachauri, can be especially dangerous since he plays a role in structuring the very system that they hack. Not only do they exploit a system they were meant to protect, but they also act in ways to prevent detection of their hacks.
What the recent disclosures about Pachauri show is that the IPCC process is now totally compromised, compromised from the inside. Like a system exploited by a destructive hack, that system may be beyond repair. It’s time to reformat the hard drive and start from scratch, as climate scientist Hulme, spotted 101 times in the Climategate files and closely tied to Pachauri, suggested–perhaps in a moment of clarity after the discovery of the emails:
“It is also possible that the institutional innovation that has been the I.P.C.C. has run its course. Yes, there will be an AR5 but for what purpose? The I.P.C.C. itself, through its structural tendency to politicize climate change science, has perhaps helped to foster a more authoritarian and exclusive form of knowledge production.”
Just how was the CRU system hacked? And can such hacks be prevented or are they a very part of the nature of authoritarian systems? As discussed in our book “Climategate: The Crutape Letters , now available on Kindle and in Ebook format, the “hacks” were focused on the publication process.
The IPCC reports were intended to be summaries of the science, both what we know and what we don’t know. These summaries were written for policy makers who would use the science and its findings to take action: action to prevent climate change, mitigate it, adapt to it, and to fund new studies where knowledge was uncertain. And action is where the money is. Every hack of the system cashes out into some form of compensation to the hackers: more money for their organizations or more prestige for themselves.
As the mails show the hacking of the CRU process and the scientific process itself was exposed primarily because of the “pressure” put on the hackers by the FIOA process. And as the mails also show, the hackers moved to thwart the FOIA process by corrupting FOIA officers. In fact, the most egregious hack of the system, Jones’s request that people delete mails, came as the direct result of trying to cover up a hack of the process, the hack surrounding the ‘Jesus Paper.’
During the course of writing the account of Climategate, we noticed that Pachauri, was named directly in the files 11 times and he’s mentioned by title in others. Before canvassing those mails it’s key to understand how the hackers worked to subvert the IPCC process and the scientific publishing process. And finally, it’s important to understand that they are planning counter measures that will allow them to continue hacking the system with the upcoming fifth assessment report due in 2013. As Jones indicates in late 2009 ,Pachauri and Thomas [Stocker] an IPCC Co Chair took up the issue of FOIA with the full IPCC. One can’t imagine that they argued for full disclosure or full compliance with disclosure laws. As detailed in this mail Jones’ alerts Stocker, who works at Bern in Switzerland, to measures Holland suggests for the next IPCC report. Jones alerts Stocker to specific comments on Climate Audit, arming him for the next battle.
Tricks of the Trade.
The high level hacks of the IPCC system include the following: changing chronologies, altering the appearance of graphics to tell a different story making up science out of whole cloth , and citing non peered reviewed literature when they could not keep contrarian papers out of the IPCC documents.
The IPCC reports are supposed to be objective summaries of the state of the science and are supposed to reference only works that are published in the peer reviewed science that met specific time deadlines. The deadlines are critical to have a fair and open process, ensuring that papers referenced have, in fact, been published and reviewed by others. Jones, however, finds such procedures to ensure fairness an impediment to his determining what is important.
Hacking the deadlines to get the ‘Jesus paper’ into Chapter 6 of the 4th assessment report of the IPCC was easiest hack to uncover. As the hackers left a paper trail of their activities that exists outside of the climategate mails. The mails, merely confirm what Steve McIntyre had already figured out. The mails of course add some background that strengthens the case, including the hacker’s knowledge that McIntyre was on to their game as well as Jones suggestion to cover up the hack by changing the “received date” on the paper.
By hacking the deadlines of the process, and shoe horning in a paper that would not be fully reviewed and published for over a year, a paper that contains references to another paper published after the fact, the hackers could ensure that lead author Briffa had the ammunition he needed to write the chapter the way Overpeck, his overlord, wanted it written: with a punch line more compelling than the ‘Hockey Stick’, that iconic figure of the global warming religion which “shows” that the current warming is “unprecedented in human history.” This hack is akin to check kiting or akin to planting older fossils with younger fossils as occurred in the Piltdown Man hoax.
To further the promotion of the ‘Hockey Stick’ message the Chapter 6 team also manipulated graphics, “hiding the decline,” even when one reviewer explicitly demanded that they show the graphic as it appeared in the original science. And finally, after keeping one skeptical paper out of the first two drafts of the report, as they had threatened to, the scientists finally resorted to making up facts on the fly.
Those are the user level hacks, but they go deeper. At the IPCC level the hacking is open to scrutiny, and as we see, the journalists following the references in the IPCC document are now finding these user level exploits. The deeper hacks, like a rootkit hack, involve the operating system of science, the science journals themselves. Manipulation of the science at the IPCC level is easy to document, but corrupting the journal process, the process that is supposed to feed the IPCC process with “trusted” information, is tougher to document.
Without the mails which detail how these hacks work, one could imagine that the IPCC could be made “hack proof” merely by adopting more controls and a more open process. The mails, however, indicate that the science publishing process has also been hacked. Editors have been compromised, and the system of peer review has been corrupted. Very simply, one can make the IPCC process as open and transparent as one likes, but as long as it is fed by a corrupt journal process, you will still get garbage science out of the IPCC process. And further, you could reform the journals all you like and the process can still be corrupted by the individual influential researcher who hides his data and his code.
The 11 mails mentioning Pachauri by name, only give hints at how the system was worked to his advantage. Reading them now in hindsight is instructive. The first from Rob Stewart to Pachauri in 1998, before Pachuari’s tenure begins, details the process of communicating about IPCC matters. As it notes, all communication should happen within the structures set up so that a traceable chain of evidence can be created. As the Climategate files show, this policy was violated, violated precisely so that the 4th assessment could be shaped in a way not reflective of the underlying science.
In the second mail Pachauri invites Hulme to a conference, initiating a relationship with Hulme that will culminate in Pachauri and Hulme working together to on UEA’s effort to win a government bid on the creation of a climate change center in England, as detailed in the third mail in the stack. That effort results in the founding of the Tyndall Centre, a center named no less than 11 times in the mails.
The interests of Hulme and Pachauri are clear. Use the science arm at CRU to drive conclusions in the IPCC that will drive funding into Tyndall and drive money into TERI, Pachauri’s organization, and CRU. The importance of funding should not be underestimated. Money works to corrupt science, not by changing the answer, but by changing the questions that get asked.
The fourth mail shows some of the TERI organization’s interests and one can see from the articles promoted in their newsletter that the agenda is clearly one of using the threat of climate change to drive a particular developing nation’s agenda.
The fifth Pachauri mail is interesting for a variety of reasons. The mail contains other mails, notably mails from Tom Wigley and Phil Jones who question the appointment of Pachauri, as they apparently believed that Pachauri was put in place by political forces, namely Bush, to serve corporate interests. This typical kneejerk reaction of the Union of Concerned Scientists, who alerted Wigley to the issue, on reflection, is somewhat laughable now. Hulme, as the mail show, defends Pachauri, his Indian partner, in the bid to get the Tyndall center, on rather odd grounds. Simply: Wigley points out that the “bushies” are behind this and Hulme defends Pachauri by arguing that perhaps Watson ,a Brit and the previous IPCC secretariat, has gone too far; and Hulme wonders why an Indian and an engineer should not be considered. Hulme strangely turns a political attack on Pachauri into an attack on his back ground. The defense seems entirely misplaced, unless of course there has been some other communications where these concerns were raised.
So why does Hulme go “racial?” Whatever the reason it’s apparent that discussions about Pachauri’s appointment don’t continue. Hulme played the race card and issue ends there. Perhaps Hulme played the race card so he would not have to explain how having Pachauri “on board” would be beneficial. Hulme’s partner is now in charge of the IPCC; and as Hulme writes, recalling the language of the conference Pachauri invited him to back in 1998, it will be a good thing to have someone in charge who understands that climate change is a North/South issue. It will also be good to have someone in charge who can funnel benefits your way.
The sixth Pachauri mail is from Tom Wigley in 2003. The topic of the mail is the strategy for responding to a single paper by Soon and Baliunas. It’s hard to fathom why a scientist would include the head of the IPCC as well as many other scientists on a mail that lays out the strategy for responding to lone contrarian paper. The Soon and Balinunas paper, did have the notoriety of being a skeptical paper accepted by the peer review science, but it was hardly a crushing blow to “the science.” Still, Wigley lays out for Pachauri, what the goal is. Rebut the paper to make the job of writing the next IPCC report, due years down the road, easier. So, the practices of just letting bad science die, or of letting individual scientists rebut the bad paper as they see fit are brushed aside. Wigley believes that there must be some sort of orchestrated effort to provide the future writers of the IPCC reports with specifically targeted ammunition. Science gets turned to producing papers to make the job of the science summarizers easier.
In the seventh mail in the stack Michael Mann appears to weigh in with his opinion. Included in the mail, is Stephan Schneider. Schneider, whose name appears at least 71 times in the mails, will later prove instrumental in getting the Jesus Paper into Chapter 6 of the 4th Assessment. He is a Trojan horse inside the journal system. But for Schneider, who “knows the drill” , the Jesus paper would never make it into the report. What Schneider and Pachauri see, of course, is that Mann is making the contrarian paper a political issue and David Halpern of the OSFT has been contacted. In the eighth mail, Jim Salinger will add his voice to the choir of those suggesting that the “bad” science be refuted rather than merely ignored. Again, one has to ask what is the head of the IPCC doing on these mails? Why do Mann and Salinger think that he will be remotely interested in the fate of a single paper?
In the ninth Parchuari mail, we see Hulme and Pachauri starting to take advantage of their connection through the Tyndall Centre, and Hulme suggests a joint publishing project. So we have Hulme, who brushed off Wigley and Jones’s political concerns about his partner Pachauri, actively engaging with the head of the IPCC to work on projects that both of their organizations will benefit from.
The 10th mail is interesting for one reason. It provides some rich irony and it shows what men of character do when a publishing process is subverted. The file contains an email from Mann and from Salinger, and both reference the turmoil cause by the Soon paper. Notably Hans Von Storch is reckoned a hero for resigning his editorship at Climate Research over the incident. The journal that he worked for had published the suspect paper, and his response and the response of other editors was resignation. Simply put, those who worked on the IPCC reports took the same stance as Von Storch, they would take their names off the IPCC report. Just as Climate Research erred in publishing the flawed Soon paper, the IPCC process has also published bogus science, even invented science where there was none.
In 2003 that one simple error was enough to get men of character to resign. Authors of the IPCC report would do well to follow their example. The incident bears some striking similarities to what we see today. The Wall Street Journal editorial by Antonio Regalado, is included in the mail from Mann.
“This week, three editors of Climate Research resigned in protest over
the journal’s handling of the review process that approved the study;
among them is Hans von Storch, the journal’s recently appointed
editor in chief. “It was flawed and it shouldn’t have been
published,” he said”
And most importantly, the flawed science found its way into an EPA document much as flawed science has made its way into the IPCC report:
“A reference to Dr. Soon’s paper previously found its way into
revisions suggested by the White House to an EPA report on
environmental quality. According to an internal EPA memorandum
disclosed in June, agency scientists were concerned the version
containing the White House edits “no longer accurately represents
scientific consensus on climate change.” Dr. Mann’s data showing the
hockey-stick temperature curve was deleted. In its place,
administration officials added a reference to Dr. Soon’s paper, which
the EPA memo called “a limited analysis that supports the
administration’s favored message.”
In 2003 the journal Climate Research let through a bad piece of science that challenged the “hockey stick.” That bad science made its way into an EPA document because it favored the administrations view. Clearly this was wrong. And Von Storch took the right action. And now, with the evidence that bad science has made its way into the IPCC report, its time for the responsible parties to follow the fine example of Von Storch.
The 11th mail which mentions Pachauri is important for understanding the events leading up to Phil Jones’ refusal to give data to Warwick Hughes, but its relation to Pachauri is tangential, except for the hint it contains about the kind of rhetoric Pachauri would use to further his cause, an early clue into his character. Buried in a series of editorials about Michael Mann forward by Briffa to Jones, we find the following 2005 gem from Steven Hayward of the American Enterprise Institute:
“Not long ago the IPCC’s chairman, Dr. Rajendra Pachauri, compared
eco-skeptic Bjorn Lomborg to Hitler. “What is the difference between Lomborg’s view of humanity and Hitler’s?” Pachauri asked in a Danish newspaper. “If you were to accept Lomborg’s way of thinking, then maybe what Hitler did was the right thing” Lomborg’s sin was merely to follow the consensus practice of economists in applying a discount to present costs for future benefits, and comparing the range of outcomes with other world problems alongside climate change. It is hard to judge what is worse: Pachauri’s appalling judgment in resorting to reductio ad Hitlerum, or his abysmal ignorance of basic economics. In either case, it is hard to have much confidence in the policy advice the IPCC might have.”
Looking at the Pachauri mails in the Climategate files doesn’t provide any new examples of wrongdoing. We don’t need any. What it does do is give a sense of how the men who hacked the climate science operated. How they reasoned, how they strategized and what they viewed as threats and opportunities. If we want to improve the science in climate science and build a trusted system, we need to understand the “black hats.”
Building a trusted system for climate science
With the IPCC’s reputation in tatters, it’s hard to avoid the conclusion of Pachauri’s man at CRU, Hulme. Perhaps the IPCC has run its course. If trust is to be rebuilt in climate science it must start from the bottom up, with the fundamental data and code of climate science. The IPCC process will always be hackable, and so to will the climate science journals. Cleaning up the mess must start at the foundation of science with open access to the data and the code used in climate science.
The notion that science can move forward while individual climate scientists hide data from their critics is antithetical to the dictates of reason. CRU and others have no more excuses. All and any data used in climate science should be published under a Creative Commons like license, free for anyone to view and use. Confidential data should not be used; it is not necessary to the science. Phil Jones should be removed as an advisor to NOAA on data archiving and access. That’s having a black hat hacker in charge of the hardware. The code of climate science should likewise be freely available. In particular, we should press climate science to adopt a GPL license, one that enforces sharing of code. In part GPL is required because on more than one occasion some climate scientists have used the public code of others without re-sharing it. For example, they have used public code, modified it in undocumented ways, and refused to share the derivative work.
At the level of the journals trust can only be rebuilt by changes in people and principles. The list of people who need to go is easy to draw up. More importantly the Journals need to adopt principles of reproducible research. If a reviewer or official “replication” expert cannot recompile the science from source, both data and code, the science needs to be rejected.
Above the journal level at the IPCC level of course Pachauri needs to go. But the risk of him being replaced with a less crude hacker is high. The issue is that trust cannot be blindly placed in people. The entire IPCC process must be opened. We want to watch that process first hand and be eyewitness to every word.