Tuesday, June 2, 2009

Sarkozy and Peer-review...

Those of you who follow a bit French politics (actually probably not many of you, but who knows...), about six months ago Sarkozy initiated a big reform of the research system in France. The French research system is indeed in deep need for changes, although many academics were not really enthusiastic about the direction taken. But the point of this post is not to criticize Sarkozy’s views on the role the private sector should play in universities or whether or not the CNRS (more or less the equivalent of VR in Sweden and NSF in the US) represents a good research system. What I wanted to discuss is what he said in a speech a few months ago to different research council’s representatives, university directors and politicians.

The speech was video recorded and so (un)popular that it was immediately translated and criticized on web platforms such as Youtube (here is a link for those interested). One of his most criticized statements in this speech was his open and extreme criticisms towards the way papers are published or more so the way grant applications are reviewed and research evaluations are conducted. Basically he literally criticized peer-evaluations and peer-reviews, finding it remarkable that “the one who acts is at the same time the one who is evaluating”. Well, this is rather inaccurate of course as we all know, but beyond this lack of basic knowledge on the way research works, there is a deeper problem in this way of reasoning.

Of course, there can be bias in the peer-review system as we all might have experienced, sometimes positive bias and sometimes negative bias, but most of the time reviewers do a great job for free and with great professional consciousness. But most important is that Sarkozy is actually implying that, for instance, research evaluations or grant applications reviewing should be conducted by people outside research, like politicians. The administrative side of academia is already one part of the big problem in my opinion, but adding a few bureaucrats to the equation is not only going to slow things done (including reforms) but also, it is going to result simply into bad research, because politically biased and of course incompetently judged/financed. As the video I suggested you to watch ironically says, let’s have all the speeches of the French president peer-reviewed before he pronounces them, even by his own team if he chooses to, just for fun…

8 comments:

  1. Interesting insight in to the ugly thoughts of a politician! He is unfortunately not as unique as one might have hoped...

    I have made some layout-changes, separated in to three different sections to improve readability and made link in your post, you are still somewhat inexperienced in the finer points of blogging :)

    ReplyDelete
  2. thanks a lot erik, i really tried to set-up a link, but it kept not appearing when published... you really have to show me...
    i agree he is certainly not unique, but still, i find it remarkable that he dares to express his thoughts so openly, as if it was common-sense... well, maybe it is, and maybe it is also our responsibility as researchers to make the peer-review system, as other specificities of our worls, more understandable to a general audience, and blogs such as this oneor other types of web medias might help in this process... from a personal experience, my parents were amazed the first time i explained to them that it can take up to a year or more to get a paper published, and that usually, we dont get paid for either publishing or peer-reviewing, but rather the opposite (for publishing i mean)...

    ReplyDelete
  3. This is interesting. The surprising thing about peer review, to me, is that is works so well!! Yes, you get the occasional junk review, but most of the time the reviewers improve my papers and raise good points. We tend to create a lot of work for each other, too, because it would be easier on everyone to just say, "Yup, looks good." But we all have way too much pride for that! Lazy people don't last long in science. Perhaps you have to experience external reviews to believe how well they work; perhaps it isn't obvious to those not in the game how all of this works. My gut feeling, however, is that the goals of Sarkozy are different. What he proposes is a power grab that would put science firmly in the hands of politicians and other power hungry folks. Frankly, I'm surprised this wasn't done by Bush et al. first!

    I should add that perhaps there is a better system of reviewing out there (for example, perhaps we shouldn't know who the authors of the paper are?), but putting the review process in the hands of politics will do great damage, if for no other reason than the reviewers are unlikely to understand what is being said.

    ReplyDelete
  4. As I read this post I came to laugh at the thought of having Swedish politicians reviewing our articles, I guess there would be some hilarious comments... ;-) But well, in reality, Sarkozy's proposal is not at at all fun.

    I'm also wondering if it is a general worldwide trend that politicians are gaining more power in directing research funding etc? Isn't that the case, alhough not so extreme, also here e.g. in the way funding is being divided between universities or how applied research is promoted? I don't know, I'm still quite new in the game so I don't know how it "used to be", probably research has always been directed like this to some degree. But if it is a general trend, how far will it go? And how much power has Swedish politicians in science?

    ReplyDelete
  5. Great post Fab.

    I have been following a podcast called TEDtalks. They have video podcasts of talks by leading scientists from over the past 6-7 years for TED (Technology Engineering Design). They have some great talks and the topics range from a massive spectrum of areas. It is worth checking out.

    The reason I mention it here is that there is a talk given by Kary Mulis (won the nobel prize for developing the PCR) about celebrating the scientific experiment. He talks about how the churches interference in science (such as Robert Boyles vacuum pump, which the church didn't like, as how could we create something where God couldn't exist), was one of the main reasons for the founding of the royal society in London. A place where anyone could come and present their ideas for scrutiny from their piers separate from church and state. Funny that 350 years on and after all that has been achieved a Frenchman wants to change the ideals of the English ;-).

    If Sarkozy gets his way there will be more French scientists outside France than in, which will bring the suffering to us all, we must fight this tooth and nail. Maybe the ones who remain will re-open the files on Lamarckism (how could an Englishman be right!!!!).

    ReplyDelete
  6. Oh I should mention that Kary Mullis talk ends up being about how clever he is for coming up with PCR and what a crock of poop climate change is. It is not the best talk, just slightly relevant to this post.
    There are a lot of talks listed, and I have not seen that many, but here are some of the talks I have enjoyed:

    Secrets of movement, from geckos and roaches - Robert Full
    Talking about the evolution of feet and how it can help in design.

    The paradox of choice - Barry Schwartz
    Interesting talk about the downsides to freedom of choice.

    How juries are fooled by statistics - Peter Donnelly
    Brilliant demonstration of the dangers of misunderstanding statistics and the probability of uncertainty

    Measuring the fastest animal on earth - Sheila Patek
    A nice look at the speed of the mantis shrimp striking appendage, and also shows a real life spandrel (14:29 the patterns on the springs!)

    A brief history of violence - Steven Pinker
    On the misconception that we have become more violet than our ancestors.

    ReplyDelete
  7. this TEDtalks seems indeed interesting, i will try to check it out whenever i can...
    i did not know about the royal society, but it seems that people who might have thought this "independency" argument might be obsolete might now have to revise their judgement... i agree with all of oyu that science should be conducted independently form politics and other things, although the progress made by science should be published for everyone to access, like in PLoS ONE for instance

    ReplyDelete
  8. Tina:

    You are absolutely correct that there is an increasing tendency, also in Sweden, for politicians to want to gain control over science. They simply do not seem to trust peer-review and the researchers themselves.

    For instance, the Swedish Research Council (VR) has sofar had a majority of scientists in the board, but our current "Utbildningminister" (Lars Leijonborg, Folkpartiet), has suggested that this should be changed. Now, there are representatives from the industry and politicians in the board, but they are a minority. Leijonborg wants to have a majority of non-scientists in the board, which is likely to have some bad consequences.

    More generally:

    Peer-review does have problems, in terms of nepotism and unfairness (as we have experienced). But this is no argument against it, no system is perfect and as long as we have not come up with some better, I think we will have to rely on peer-review also in the future.

    However, one should not forget that the most important peer-review might take place AFTER publication, when a paper is critically evaluated, cited and discussed. Thus, there will be an "a posteriori" peer-review process, similar to the process of natural selection. The "fittest" papers will survive and become cited, while the bad papers will (hopefully) die. Those bad papers include those which by mistake (or due to friendship corruption from positively biased referees) made a paper become accepted even though it should not have been.

    It is in this context the new commenting function at PLoS ONE, where one could comment and rate paper, hopefully will have a future and increase in popularity. I would strongly encourage all of you to sign up at PLoS ONE and rate and comment on articles to contribute to this process. This same model has now also been applied to the other journals in the PLoS-group, such as PLoS Biology.

    ReplyDelete