Climate truth and transparency
I can understand why the people at the IPCC and the UEA climate group did what they did, but it doesn’t stop it being stupid.
Peer-reviewed science has been captured by the spin people. The spin people think they can manipulate news by feeding press releases; and conventional journalism has long since been reduced to a process of press-release rehash. Science, however, is too important to be spun.
Spinning science is something the drug companies have been doing for ages. They have suppressed reports of side-effects and they have boosted publication of positive effects. Since they fund the clinical trials, they get to decide whether or not results should be published, so the result in the literature is that there are many more reports of positive trials than of failed or negative trials. Belatedly, the industry is now demanding that trials be registered before they take place, so drug companies can’t rig results. Still, a disturbing number of journals publish reports of unregistered clinical trials. If such results are published, publication should be accompanied by a big red health warning.
In climate change, the big corporate baddies aren’t the pharmacos but the petrocos, in particular Exxon which has secretly funded a number of climate denial pressure groups. These groups are very successful in getting their spun stories into the mainstream press, particularly those which are sympathetic to the denialist agenda. That their tactics are successful is not a reason to copy them, if they are wrong. Which they are; but the IPCC and UEA seem to have been seduced by the same ethos, putting spin over substance. It won’t do. We need to trust science and be able to trust science, which means that we must be able to trust peer-review. Peer-review isn’t perfect – by any means – but it’s the best we’ve got. There are certainly adjustments that can be made to the peer-review process; it needs to be transparent, but also to preserve anonymity. There’s also some evidence of institutional bias, including racial and gender bias, where reviewers are less likely to review positively papers from authors with foreign-sounding, or female, names; or from institutions outside the top rank. Even so, a peer-reviewed paper in a top journal is much more likely to be right than a press-release.
Press releases, brochures and other non-reviewed papers, and doctoral theses, are part of a category of literature called “grey literature”. The glacier-melting mistake in the IPCC’s proceedings stemmed from the use of grey literature. There is a lot of useful information in grey literature, and researchers should be able to use it. But citations to grey literature should be tagged, so reviewers and summarisers can pick them up and filter them out. The same sort of tagging could filter out citations to unregistered clinical trials, to papers published in journals with a known weakness in their peer-review policy, or to papers published without publication of the associated raw data. This last is a particular bugbear of mine; with one very strong exception, I don’t think that any paper should ever be published unless the raw data, from the experiment being reported, is published at the same time. The one very strong exception is in the case of medical papers involving human patients; here, the raw data is sensitive personal information subject to the data protection act, and patient confidentiality has to trump scholarly transparency. It doesn’t apply in climate science.