Details of investigative report into Sarkar released by ACLU

Fazlul Sarkar
Fazlul Sarkar

We knew that Wayne State University had investigated allegations of misconduct against Fazlul Sarkar, the scientist who is suing PubPeer commenters over criticisms of his work. We knew The Scientist had obtained a copy of the report, which concluded he had engaged in widespread misconduct, and he should retract more than 40 papers.

And now, thanks to the American Civil Liberties Union, which is representing PubPeer in court and has filed a motion to include the report as evidence in the case, we have a copy.

It’s a long read, but here are some highlights:

Several poor and/or irresponsible research practices appear entrenched in Dr. Sarkar’s laboratory … these include tailoring results toward specific conclusions, image manipulation, viewing figures and images as merely representative or irrelevant to experimental outcomes, and a reckless disregard of a meaningful use of control groups or control conditions … The Investigation Committee concludes that copying and re-using and manipulating images were accepted and practiced by many if not all lab members.


Even if there were no intent to deceive on Dr. Sarkar’s part — which, based on the evidence, the Investigation Committee finds difficult to believe — the level of carelessness needed to account for all the copying, manipulation, and re-labeling of images is so great as to define reckless on its own.

Apparently, Sarkar didn’t make it easy for the committee to investigateas the report notes:

The Investigation Committee was able to find some original data — albeit by brute-force searching — that Dr. Sarkar and his team claimed they could not find. There was even one occasion where Dr. Sarkar submitted as replicated data, data that were actually original. This suggests that he and his team apparently do not know at this point what data are real and what are fabricated, and have no means to find out given that their lab notebooks are often functionally useless.

The process was protracted, as the report notes; the Investigation Committee met for between three and five hours nearly every week for two years, totaling approximately 90 sessions.

For more information on this case, check out our timeline.

Like Retraction Watch? Consider making a tax-deductible contribution to support our growth. You can also follow us on Twitter, like us on Facebook, add us to your RSS reader, sign up on our homepage for an email every time there’s a new post, or subscribe to our daily digest. Click here to review our Comments Policy. For a sneak peek at what we’re working on, click here.

10 thoughts on “Details of investigative report into Sarkar released by ACLU”

  1. P429 describes the “siloing” of members of the Sarkar laboratory (ie nobody knew what anyone else was doing). This echoes the situation in the Anversa laboratory (and some other examples of laboratories that have been found to harbor environments that encourage misconduct). Could institutional policies that encourage open or joint lab meetings be a way to curtain these kinds of toxic laboratory environments?

    1. Only problem with joint lab meetings is if both PIs have the same practice. The only solution is to make science more efficient. Beyond a fairly modest funding (for people) level, returns decline per unit investment rather rapidly. Unfortunately, the reward system drives inefficient investment. This is not sustainable.

    2. The current reward model in the sciences may also play a role. When each manuscript is considered only an individual accomplishment and participating in team efforts doesn’t count for funding/hiring/promotion decisions, what is the incentive for scrutinising other people’s experiments? Being too critical of other people’s work in your lab can also earn you a reputation for being difficult or obstructing the publication/grant mill. Once this kind of culture is in place, it’s easy to see how misconduct can eventually appear out of what starts as mere sloppiness.

      1. That is all true although I do think some more enlightened institutions are starting to recognize and reward collaborative research.
        In addition to having a personal research program I run an institution wide facility core so I interact with a lot of different PIs and their lab members. I am struck by the large differences in the level of openness in different labs at my mid level institution. Some of the labs I deal with are definitely run in ways that mirror the worst traits of the Sarkar lab and I often wonder about what is really going on in them.

      2. @TL: your comment is perfectly reflected in the sections “Collegiality” & “Accepted Practices” of the Final Report released by the WSU. The consequences on the reward model you mention are astonishing; for example (on p. 429):

        “Dr. Sarkar said, ‘… many times, a co-author may not have a significant contribution, but I put the name of the co-author so that we can build our portfolio towards the objectives of the program project grant’ “.

        This sentence echoes weirdly in the pressure put on those involved in Science. In my country, the mantra is not at all “work hard!” but rather “Collaborate! Collaborate! Collaborate!”. However, at the end, 80% of the incomes of researchers are based on an individual evaluation considering these collaborations only in a very tangential manner, or even as a negative point (some referees follow equation “too many collaborations = dispersion”). Ultimately, we all know that the only thing that holds for the record of an individual is the number of publications, as mentioned by Dr. Sarkar.

        Thus, we come full circle. And Science is outside the circle.

  2. The detailed report of the WSU investigation into Sarkar is just incredible! How could this overt culture of research fraud go on for so long and aquire so many publications and EXTENSIVE research funding??? I am just floored! The investigation by WSU into this matter was thorough, diligent and very professional – I now understand why this took so long to come out.

  3. Where did these graduate students come from? At my alma mater, you’d never pass General Chemistry or one of the other hard science lab courses without the ability to keep an orderly and readable lab notebook.

  4. This is the one upside to “scientists” dragging everyone into a court room to deal with accusations of data and image mishandling – documents that would otherwise remain confidential are obtained and revealed through FOIA requests.

    PubPeer commenters criticisms look minor and tame compared to what’s in this report.

    Many thanks to Retraction Watch for making the report available. Time to donate to the ACLU again.

  5. Thanks, Retraction Watch, for publishing this report. I have been waiting for this report for many years.

    I was surprised that so few articles will be retracted. 42 papers out of 541 listed in PubMed. However, the investigation committee examined only articles “during the 6-years prior to the date of the initial investigation of research misconduct”. This means that they only examined articles published in the period Oct 2006 – Oct 2012. Approx 226 articles where published in this time frame, which of approx 19% will be retracted.

    How about the more than 200 articles published in the period 1977-2006 and the 120 articles published 2012-today? Based on the investigation report shouldn’t there be an expression of concern of every article published from this group?

    What strikes me is that WSU did not made the investigation report public available when it was finished, in August 2015? This is very important information thinking about all the resources other researchers have wasted based on this fraud.

  6. After reading much of the linked PDF, I sure would not want Dr. Sarkar to run a lab on my campus.

    Most of his explanations sound like excuses. “The wrong figure was inadvertently used” seems to come up a lot.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.