The Climategate Emails: a critical reading

This post summarises my take on the The Climategate Emails by John Costella. For someone like me, I felt it was important to look at it. I'm learning about climate change, sympathetic to the IPCC warnings, but responding to doubts from apparently well-informed people who view the IPCC warnings as alarmist.  The Climategate Emails is frequently cited as 'proving' the untrustworthy motives and dark political machinations of climate scientists who accept and/or contribute to the IPCC reports. Its basic premise is that the emails hacked from a UEA server back in 2009, exchanged between a fairly small number of players in the UK and US, expose the murky and unethical practices at the heart of a dark cabal. A long series of investigations since then have found that the emails do not expose scientific malpractice. Rather, it is said, the emails have been selectively quoted out of context and misrepresented as well. I have to say at the outset that is also my own conclusion after reading Costella. No truly sceptical believer in science could accept Costella's conclusions without mountains of further investigation and evidence. Read it yourself if you doubt my conclusions.

The author of The Climategate Emails, John Costella, describes himself as an Australian with a PhD in theoretical physics, a former researcher and schoolteacher who now works as a data manager and researcher for an investment firm.

The first thing I want to say is that Costella doesn't state the methodology he followed in analysing the climategate emails. It's not even clear he has one. Social scientists over many decades have developed techniques and methodologies for analysing individual attitudes and affiliations, social interactions, political organisation, and a whole range of other aspects of communication behaviour. For a task like this, I might have expected some kind of content or interaction analysis, or perhaps a grounded theory approach. Costella does none of this and includes no references to indicate any literature search in the field. In fact, other than the leaked emails themselves, he seems not to have read anything at all. I'm sure that's not true, but he makes no effort to show the reader where his ideas and influences have come from.

Instead, Costella simply quotes extracts from various emails in the hacked collection. For each extract, he 'interprets' the narrative for us. He explains what the author sought to use the email to achieve, and imputes motives to them for which we have no evidence other than the text itself. He attributes to the authors further actions and communications for which we have no evidence at all. In the circumstances, I thought it only fair to adopt the same approach in relation to Costella's work.

The document begins with 2 pages of introductory material, followed by about 170 pages of 'analysis'. Before this begins, we are given a "cast of colourful characters." Just like a play or a novel, apart from that 'colourful'.  Given that these are all working scientists, some very senior in their field, this already begins to feel a bit like a sneer. And it gives a flavour of what will follow. Keith Briffa is an "older player whose blunders lead the others to all but abandon him." (Sadly, Keith Briffa died just a few weeks ago, aged just 64). Ben Santer is "dangerously arrogant and naive." Other (at this point unnamed) players are "of varying degrees of complicity and integrity." Nevertheless Costella seems to present himself as an objectively minded scientist, neutral in his approach but reluctantly and sceptically driven to some unhappy conclusions. I don't think that stacks up at all...

In the first 5 pages, there are 7 email extracts. I read each email extract carefully and Costella's interpetation of it, and made detailed notes of my reactions. My notes run to three A4 pages; I'll try to summarize here.

The first extract deals with a request from a Russian collaborator to make payments in small amounts to avoid 'big taxes'. Costella characterises this as "financial impropriety" and implies that it is typical of other cases of "misappropriation sprinkled through the... emails." But he admits he has no other evidence about this particular incident, and can't even confirm whether the small transfers happened as requested. I don't know anything about the financial transfers - what they were for, who funded them, what agreements surrounded them, what legal implications there might be. And nor does Costella. In my view this is just trial by innuendo.

The next 4 extracts are from an exchange about Piers Corbyn. The email author calls him an 'utter prat' and mentions he is getting a lot of air time. Costella's interpretation is that, in taking "special note of the 'air time,'" the email shows that "politics and 'spin doctoring' already outweigh the scientific issues" for the author (this is Professor Philip Jones, Research Director of the CRU at UEA). Hm. Well, maybe. My own reading is that this was a private chat between colleagues about some of the personalities and politics around their work. I can't see how this in any way relates to their published scientific work, or their integrity as scientists. And I'm rather inclined to agree about Corbyn. More trial by innuendo.

The next extract is from an exchange about some data that a US researcher offered a UK colleague (Keith Briffa, then a Professor at the CRU). Briffa is interested in the data, but cautious about presentation. He says "the board would want the larger implications of the statistics clearly phrased in general and widely understandable (by the ignorant masses) terms before they would consider it not too specialised." What does Costella make of this? "Briffa makes it clear that he is only interested in the data if they can be used to 'sell' the climate change message to the general public." I don't think Briffa does that at all! A scientist involved in the public presentation of their work has a separate problem over and above the scientific work: how to make it understandable to a lay audience. Why would you put all the hard maths in a general press release bare of intepretation? I think that's all Briffa meant. I admit this text alone doesn't conclusively support my reading. But nor does it Costella's. Trial by innuendo again.

The next extract is a follow up to the previous email. A US colleague (Funkhauser) says he has "pulled every trick" to get some results out of the data, but "they just are what they are... let this little story go as it is." Costella goes to town on this. Funkhauser is trying to hide a null result; he's like a gambler who brags about wins and hides losses; this is a "fundamental scientific flaw." Worse, Funkhauser is yielding to "immense power wielded" by Briffa, who in the earlier email (above)  sublty reminded him that results need to be politically saleable - and that Funkhauser's future funding, publications, career even, depend on doing as he is told and suppressing this data. My view: the emails of both Briffa and Funkhauser can easily be interpreted in perfectly legitimate ways.  All Funkhauser says is he can't extract meaning from the data. He is warning Briffa that the data don't say anything useful, so don't use it. He says nothing about null results or suppressing anything. Nor does Briffa say anything that can be remotely said to exercise any leverage over Funkhauser. This is nothing short of blatant character assassination on the flimsiest of pretexts.

I stopped my analysis there (honestly, I'd had enough). But further on (p28) Costella makes much of the famous "hide the decline" email. Costella calls this "one of the most serious scientific frauds in the history of Western science." Hyperbolic, or what? This blatant misrepresentation of what Jones' email actually said, and actually meant, has been discredited over and over again in the years since. A random sample of rebuttals: Arizona Daily Star, Dec 7, 2009; article on skepticalscience.com; Mike Mann and Phil Jones video on denial101.

Conclusions... Costella appears to have looked at the UEA email material with preconceived conclusions already firmly held in his mind. His 'analysis' consists of no more than searching the text for anything that can be read, at any stretch, as confirming his biased preconceptions. His language is hyperbolic, and his attributions are consistently ad hominem. He claims scientific expertise, and says "these mathematical and statistical methods are completely within my own domain of expertise."Really? I'm not convinced he has any understanding of the domain of climate science. But if he does, what should we think then? And remember: this man accuses others, on no sound basis, of one of the most serious scientific frauds ever.