Tuesday, February 1, 2011

Don't trust this blog

This blog, I think, is well written and edited. I think that each post I produce is based on solid evidence and often, it is based on one or more peer-reviewed papers which are particularly newsworthy. Nevertheless, you can't trust it.

Try as I might to be totally balanced, this is just my representation of scientific news stories and peer-reviewed work.

I suppose what I'm saying is that my blog, like all other blogs, is not a replacement for peer-reviewed literature and the blogosphere is not an alternative to peer-reviewed scientific journals.

As a scientist, I recognise that the peer-review system works. Of course, there are problems with it and a few high-profile cases where it seems to have failed us, but overall it has stood the test of time. The British Government has recently announced a review of peer-review, but I would be confident that it will be shown to be the best method at our disposal to verify scientific findings.

Unfortunately, it would seem that some commentators feel that peer-review is definitely old hat and that a new system is needed. Of course, I'm talking about James Delingpole.

In a recent interview with Paul Nurse, for the BBC programme Horizon, Delingpole argued for the use of what he called 'peer-to-peer' review.

As you'll see from the video clip of the interview below, Nurse first made an analogy between accepting the consensus scientific viewpoint on the treatment of cancer and accepting the consensus scientific view of climate change. Writing in The Telegraph, Delingpole, a well known climate change sceptic, subsequently described this analogy as "shabby, dishonest and patently false".

"The consensus on climate change; and the consensus on medical care", says Delingpole, "bear no similarity whatsoever". You can judge for yourself whether the analogy  makes sense or not.

Having been flummoxed by Nurse's astute line of reasoning, he was subsequently asked about his use (or not) of peer-reviewed climate change literature.

"One of the main things to have emerged from the climategate emails was that the peer-review process has been, perhaps irredeemably corrupted" replied Delingpole.

"What I believe in now...is a process called peer-to-peer review. The internet is changing everything. What it means is that ideas which were previously only able to be circultated in the seats of academe, in papers read by a few people can now be instantly read on the internet and assessed by thousands and thousands of other scientists; people of scientific backgrounds and people like me who haven't got scientific backgrounds but are interested."

What he's talking about of course is the blogosphere. Now, as keen as I am on this whole blogging lark, I do not believe that the ability to switch on a computer and type entitles anyone (including myself) to begin to interpret scientific data for which we are wholly unqualified. Sure, we can have opinions and ideas about the findings but, as the old adage goes,"we are not entitled to our own facts".

Delingpole however seems to have no time to even begin to interpret the data correctly:

"It is not my job to sit down and read peer-reviewed papers because  I simply haven't got the time, I haven't  the scientific expertise. What I rely on is people who have got the time and the expertise to do it and write about it  and interpret it. I am an interpreter of interpretations."

All well and good, and it's to his credit I suppose, that he admits that his opinions are not based on the real, original data. However, if he is to interpret the data (or the interpretations of the data... you know what I mean) then he must interpret all of the data and that means the overwhelming volume of research that points  to a global warming phenomenon caused by human interventions.

I may blog, but blogs are not real science. You can trust me on that!

3 comments:

Bishop Hill February 1, 2011 at 7:27 PM  

It's not obvious to me that you can trust either blogs or peer reviewed science. They both have their problems, and as the Royal Society's motto makes clear, science is not about trusting anyone.

Delingpole was quite right that the analogy between climate science and medicine is not a good one. The efficacy of a medical treatment has been tested and tested again. The AGW hypothesis hasn't been and can't be. The failure of the Earth to warm since the last IPCC forecasts were issued suggests that the hypothesis of 2deg/century may be overstated.

You quote Delingpole's point about the peer review process in climatology having been corrupted. Do you accept this point?

Humphrey Jones February 1, 2011 at 11:14 PM  

Great post Eoin! As a Kerryman, I generally find it difficult trusting any Corkonian! But I make an exception for your blog!

Colm Ryan February 2, 2011 at 11:28 AM  

After reading your post last night I watched the programme on YouTube.

I have to say, Delingpole's point about replacing peer review with a review by everybody seems quite odd, to say the least. I am not an expert in climate science. I don't have a strong foundation in statistical analysis or research methods. I am not aware of the vast body of literature pertaining to climatology. Is he saying that I have as much right to comment on an academic paper as a climate scientist? Just because I might have a strong opinion, should I expect to be taken just as seriously as the experts? How is more noise meant to help peer review process? How do you stop every crank on the planet piling in to obstruct publication of papers they have an idealogical bent against?

Peer review may be a flawed process, but what precisely is a better process? It's my understanding that there *are* further controls in place. Say a bad paper gets published - peer review for whatever reason is half-arsed, and it gets into a recognised journal. Does the science end there? Of course not. What happens then is that the paper becomes available to a much larger audience, and if someone sees a flaw in the methodology, more experiments may be carried out, more papers written. So sure, peer review may fail for one paper or a small number of papers, but over time, there is an error correction process in place and the original papers may be discarded. You can't easily use the excuse that peer review is flawed when a large body of evidence has been submitted from institutions across the world over a significant period of time. It simply becomes much harder to hide methological errors.

  © Communicate Science; Blogger template 'Isolation' by Ourblogtemplates.com 2012

Back to TOP