As promised at the end of part 1 (here ), I shall now run over the evidence for the failings of epistemic character among climate scientists. I shall be breaking this up into loosely related groups: objectivity, the conduct of enquiry, what is owed to other enquirers and virtue in testimony.
impartiality
The emails show that these particular climate scientists are neither objective nor impartial. One even harbours violent urges towards a critic: ‘Next time I see Pat Michaels at a scientific meeting, I'll be tempted to beat the crap out of him. Very tempted.’[1]
Rather, they are heavily committed to one side of the issue, to the extent of being infuriated by even moderate disagreement and criticism. Some are willing to sign letters drafted for them by a political partisan, Greenpeace.[1] That being said, whilst both of these virtues are highly desirable, good science has been done by scientists who lack them. That success, however, depended on a surrounding context of scientific institutions that protect the epistemic integrity of science from the lack of objectivity and impartiality of individuals and groups of scientists. Hence it is especially troubling that the emails show the scientists clearly intending to suppress critical scrutiny, to subvert the institution of peer review, to interfere with the editorial independence of journals (‘If you think that Saiers is in the greenhouse skeptics camp, then, if we can find documentary evidence of this, we could go through official AGU channels to get him ousted.’[2]) and to retaliate against journals that publish dissenting papers. (‘I will… tell [the journal] I’m having nothing more to do with it until they rid themselves of this troublesome editor.’[3]). We don’t know the extent to which their activities have subverted the epistemic integrity of climate science, but it seems unlikely to have remained unsullied from an assault by scientists as powerful as those engaged in these email exchanges. Furthermore, these scientists have for many years played a major role in the UN IPCC process, and in other similar processes, and consequently our confidence in the epistemic integrity of these processes must be reduced.
disinterestedness
The hawks routinely accuse skeptics of being in the pay of evil oil companies whilst present themselves as being disinterested enquirers. For some time, however, there has been little reason to see hawks as disinterested. The question of global warming has been a bonanza for climate science and the interest of climate scientists is for the stream of research money to continue. Note that I do not say this is a bad thing: but the quantities of money involved (governments have spent billions of pounds on climate research) undermine the claim of disinterestedness. If in truth they had to announce tomorrow that it had all been a big mistake they would look like idiots, the money would stop and many would be out of a job. ‘If anything, I would like to see the climate change happen, so the science could be proved right, regardless of the consequences. This isn’t being political, it is being selfish’.[4] A frank, if rather disturbing, admission of interest.
[1] http://www.eastangliaemails.com/emails.php?eid=1045&filename=1255100876.txt
[1] http://www.eastangliaemails.com/emails.php?eid=32&filename=872202064.txt
[2] http://www.eastangliaemails.com/emails.php?eid=484&filename=1106322460.txt
[3] http://www.eastangliaemails.com/emails.php?eid=296&filename=1047390562.txt
[4] http://www.eastangliaemails.com/emails.php?eid=544&filename=1120593115.txt this version of the email seems to have got truncated before this quotation
[1] http://www.eastangliaemails.com/emails.php?eid=1045&filename=1255100876.txt
Different scientific questions also demand different levels of epistemic virtue. A simple, clear experiment is more robust against lacking research virtue than a complex model or messy statistical analysis. Expectation bias – getting the results one is expecting to get – is common in science and complex models are more easily affected. This means that impartiality and disinterestedness become increasingly important the more complex the issue is.
I have spent the last week playing around with the climate data recently released from the Met Office.
(the results can be seen at http://www.aleph.se/andart/archives/2009/12/significant_anomalies.html ) It was enjoyable to actually calculate warming trends based on this data but it also showed the complexity of doing it from real-world data. Many levels of selection, filtering, averaging across time and space, choices of reference periods and statistical models go into the final result. There is ample room for minor choices to unconsciously bias the result if one is not careful. I had several bugs that were not immediately obvious because they produced the “right” kind of answers. In order to be trustworthy, this kind of data analysis really needs evidence for a high level of epistemic integrity.
At the very least I think one should have a publicly available documentation (ideally in a runnable form) of how the data was processed.
Thanks for that. Yes, there’s something like a robustness function which is directly proportional to simplicity, publicity and checking of results, and inversely proportional to epistemic vice–simplicity dominates vice but vice dominates complexity. You’ll see in the next section, on the conduct of enquiry, that I’ve discussed the issue of complex data manipulation and bias. I agree that having this documentation publicly available is a necessary check, and CRU’s long standing refusal to share such information is a fault I discuss in the 4th part.
Comments are closed.