Friday, March 07, 2008

The Kundu case: A new twist


I have just been alerted about this letter (pdf) in the March 10 issue of Current Science:

A greater shock awaited me when I entered a name that was discussed widely in the pages of Current Science last year, primarily as a test of the veracity of the database. Two pairs of citations were picked up. The first pair appeared to be legitimate, consisting of an original paper in one journal and a review article by the same group in another journal on the same theme. The second pair of publications was more alarming since it appeared to be a clear case of duplicate publications by the same group. The first paper ... was published in 2004 and the second3 in 2006. What was not apparent from the site without further browsing was the fact that the second paper had already been withdrawn from the journal precisely due to prior publication. The journal that published the 2006 paper had a bland erratum in fine print in 2007, stating ‘This article was withdrawn due to prior publication in an alternate publication by the authors’. It does not state who withdrew the paper, the authors or the journal. Shockingly, however, the erratum gives a link to the 2006 paper and one is able to view the full text, without any indication that the paper had been withdrawn. This information can be found only by chance. It appears that many journals also do not wish to take the responsibility for making a mistake. Incidentally, the 2004 paper2 was part of the controversy discussed at length in Current Science last year. The group was accused of misconduct and later exonerated of using part of the data in that paper to represent different sets of conclusions in another publication4 in 2005, which was subsequently withdrawn by the journal. Interestingly, the withdrawal of the 2006 paper happened when the controversy was raging in the pages of Current Science and found no mention anywhere at that time. As on 21 February 2008, the official NCCS website continues to include both withdrawn papers among its list of publications.

Wow. Just wow! This bit of smart sleuthing is by Prof. S. Mahadevan, a colleague, who goes on to echo sentiments similar to my own:

we in India have no formal mechanism to address such issues. What is to be done if we discover an unethical act? Whom do we approach? Whistle-blowers are exposed, their confidentiality violated. Independent watchdog organizations are accused of ulterior motives and maligned. Our science academies are mute spectators during most of these discussions. There appears to be a great desire for damage control rather than getting to the bottom of the issues and resolving them fairly. This is in stark contrast to what happens elsewhere, including our Asian neighbour Korea. The ability of the Indian scientific community to institute fair and transparent mechanisms to handle allegations of scientific misconduct will be a measure of its maturity. Where are the Luke Skywalkers, Princess Leias and Han Solos of Indian science?

Some of you may remember that the Kundu case was covered quite extensively in this blog. Let me link to these posts: here, here, here. The clinching evidence -- and a very good summary -- is available at Rahul's site: here.

6 Comments:

  1. Unknown said...

    That's fascinating! Sad, also. I think part of the problem is due to journals' long review times. Things have improved now that many journals have online systems that send out automated reminders to reviewers, but editors at the journals that don't have that in place just have too many things to do to track papers and twist reviewers' arms when a review is overdue. While that kind of incident reflects extremely badly on the authors in small scientific communities where many researchers know each other (it's the first time I hear about something like that), I could imagine people trying to cut down time to publication by submitting to several journals. That would hurt someone's career if that was uncovered, though. What I don't understand is why the authors didn't withdraw the second publication earlier. Papers are rarely put online by journals before a copyright agreement has been signed. Did the authors sign the agreement twice? That is a legally binding document. Signing it twice would be very stupid. Maybe one author signed the first one and the other author signed the second one and they didn't keep each other updated. An odd story all around. Thanks for sharing it!

  2. Anonymous said...

    Well, I thought many journals required all the authors to sign the copyright agreement - a more reasonable thing to do, if you ask me.

  3. Rahul Siddharthan said...

    Signing it twice would be very stupid.

    Aurelie - publishing the same research twice is very stupid. Now that it has come to light, I hope the authors have more to worry about than copyright infringement liability.

  4. Anonymous said...

    Abi,
    You've often waved the banner of open access publishing-I wonder if you are able to see the connection between ethics and professional publishing. I am not a dispassionate observer-I've worked in scientific publishing for very many years now-but I bet you have no idea how many Kundu's and Chiranjeevi's are stopped in their tracks by the efforts of professional editors and copy-editors. The system is by no means perfect and our misses make headlines, but far more often than not the system does work and cheats are identified and blacklisted. Regretfully, the blacklists are never publicized for fear of legal repercussions. Let me throw out another provocative thought--and I do not mean to offend by rather to provide food for thought: in my humble opinion, India and China should be the last places to advocate open source publishing because
    a) have you ever noticed the quality of writing/english in manuscripts submitted for publication from India? I have no doubt that your submissions are impeccable but surely you've had a chance to look at what your colleagues send our way. Exactly who do you think is going to pay to clean up the garbled pidgin submitted from these countries?
    b) ethical standards remain highly suspect, I argue that there is a difference between ethical standards in the west and India/China, ask anyone who has worked in scientific publishing as to how much of a battle it is to deal with Chiranjeevi's of the world and exactly how many of them do exist (disproportionately from India). Is laxer policing that is inevitable with open access going to make things better you think?
    3) Research published in open access journals is certainly not paid for by the Indian taxpayer so there is really no argument of this being a right of any sort

    -well that is my 0.02$

    --Sahil

  5. Rahul Siddharthan said...

    Sahil - I don't know what you mean by "open source publishing" -- making the latex manuscript available?

    Open access is refereed just as stringently as closed access, or more so. Take a look at PLoS and BMC, the two most prominent open-access publishers. In addition many publishers -- OUP, etc -- have optional open-access systems in place.

    And I don't know what conclusions you draw from Kundu and Chiranjeevi's cases. Kundu's 2005 paper was published in a respected, non-open-access journal (though his paper seems to be accessible from my home machine too -- perhaps JBC has country-wide access for India, as PNAS does). His 2006 paper is published in an entirely obscure non-open-access journal that epitomises everything that is wrong with closed-access (namely, very few people can read the paper -- few institutions subscribe to the journal, because of its obscurity -- so the duplication is less likely to be caught). Chiranjeevi's papers were mostly in tier 2 or tier 3 closed-access journals.

    If you really want to make a point, please highlight a case of fraud that has happened in an open-access journal and then say why it couldn't happen with a closed-access journal.

    Finally, if you want to be taken seriously and not assumed to be an astroturfer from the publishing industry, please use your full, real name.

  6. Suresh said...

    Sahil,

    You make no sense. Fraud is going to be there whether or not publishing is open source or closed source. The people who support open source publishing incidentally do not do so in the belief that it will somehow lower the incidence of fraud. It has more to do with the belief that scientific publishing - as a public good - should not be subject to copyright restrictions.

    Regarding fraud, the problem, at least in India, is that the redressal mechanism is very poor. This has many bad effects. One, it encourages the unscrupulous guys. Secondly, it lowers the incentives for the good guys to come forward to expose cases of fraud. Thirdly, it unfairly tars the names of even the good guys. Can anyone doubt that a side effect of cases like the Chiranjeevi one is that *all* Indian academics are now looked at suspiciously?

    You are entitled to support closed source publishing which is fine. But don't link it with fraud, please.