Danish high court clears Pedersen in misconduct case

Klarlund Pedersen

Lawyers one, scientists nil.

Danish judges have overruled scientists in that nation, concluding that a panel of experts erred in finding that physiologist Bente Klarlund Pedersen, of the University of Copenhagen, was guilty of misconduct.

Last September, Pedersen announced that she would fight the ruling of the Danish Committees on Scientific Dishonesty (DCSD, Danish acronym UVVU), which had said she had committed misconduct in four of 12 articles it had examined.

As we reported then, Pedersen’s case is tied to that of another Copenhagen scientist, Milena Penkowa, with whom she had collaborated and who also has been found guilty of scientific misdeeds. (The new ruling does not address Penkowa.)

According to Nature:

In a unanimous decision, the High Court of Eastern Denmark in Copenhagen cleared Klarlund Pedersen of the charges and ordered DCSD to pay 400,000 Danish kroner (US$61,000) in court and legal fees. The court found that Pedersen’s actions did not amount to scientific dishonesty.

Eigl Lego Andersen, Pedersen’s lawyer, says that the ruling “sends a very strong message” to the DCSD that not all imperfect research practices should be stamped as dishonesty. “They have to have clear definition of what is and what is not scientific dishonesty.” …

The court agreed with Klarlund Pedersen, determining that she did not act with “gross negligence” or intent to falsify or distort the scientific message of the articles in question, but it rejected the procedural argument [that the DCSD panel members had served too long].

In what may prove a prescient statement, Nature quotes Daniele Fanelli, whose work we admire, saying the Pedersen case is far from a one-off:

Daniele Fanelli, a research misconduct expert at Stanford University in California, expects courts to play a more prominent role in deciding such cases in the future. Prosecutions of scientists are on the rise, and many countries are drafting legislation specifically addressing misconduct. “I think we should expect that some of these individuals accused of scientific misconduct will fight back.”

Indeed, we’ve been seeing it happen already — Piero Anversa and a colleague are suing Harvard over the institution’s investigation into their work, while Mario Saad is suing the American Diabetes Association to prevent its flagship journal Diabetes from retracting four of his papers.

17 thoughts on “Danish high court clears Pedersen in misconduct case”

  1. The Copenhagen Defence!

    Translated via Google Chrome from the source link provided (http://www.domstol.dk/oestrelandsret/nyheder/domsresumeer/Pages/Pressemed_18_02_15.aspx):
    “The court finds that Bente Klarlund Pedersen’s explanation appears corroborated by the enlightened that image manipulation for a long time after the publication of Article 4 was not discovered by several other researchers in the field. ”

    So if we forge some Euro’s, and manage to spend them for many years without people noticing, even expert bank tellers, it’s OK?

    We should tell our friends in Athens – could solve a lot of problems being faced over the next few days.

    1. Darrel, it is important to be aware of the context. For the benefit of all a short timeline, but first a disclosure: I work at the same faculty as Bente Klarlund Pedersen (BKP) and have met her once, several years ago.

      BKP, a professor, led a research project that included taking some muscle biopsies to look at the effect of the different treatments. The visualization of those muscle biopsies was done by Milena Penkowa (MP), at that time also a professor. Those biopsies confirmed other observations in the study.

      Fast forward a few years, and MP is under heavy scrutiny; more and more examples are found where she manipulated various pictures, along with other apparent falsifications/forgeries. Thus, all publications on which MP is an author are being scrutinized, including the paper coming from the research project described above. It turns out that the muscle biopsies pictures are also falsified: pictures supposedly taken after 0, x, and y hrs are actually a single picture cut into three pieces.

      In comes the DCSD, declaring that BKP should have noticed this, because everyone can clearly see this misrepresentation, and therefore BKP is guilty of scientific dishonesty. BKP challenged this with the argument that if _she_ should have noticed it, surely others would also be expected to notice it. It is not like the paper was never cited, that no reviewers looked at it, etc. The court apparently accepted that counter-argument.

      The court may also have taken a favorable stance towards BKP because of the term “scientific dishonesty” and the way this is likely to be understood outside of Denmark. That is, if we for argument’s sake accept the DCSD claim BKP should have noticed the manipulation, can you then call her “scientifically dishonest” because she did not? Within the Danish framework of the DCSD, yes. There are currently a lot of discussions ongoing about this broader definition in Denmark of “scientific dishonesty” compared to what is understood as “scientific misconduct” in most of the rest of the world.

      1. So BKP is blameless because he didn’t notice the figure manipulation. Well, a lot of other people didn’t notice the manipulation either. So they’re saying that he shouldn’t be held to a higher standard than the average reviewer looking at figures–only a Claire Francis would notice, probably because she was looking carefully at all of them with a suspicious eye.
        It’s reasonable as far as we understand what we’re saying, which is that we don’t expect BKP to be suspicious enough of his own fellow researcher to go over everything with a magnifying loupe.

        1. The claim is not that BKP is blameless, the claim is that BKP’s actions do not constitute scientific dishonesty in a specific legal sense of the word.
          More in general the claim is that failure to detect that certain data was falsified, when that is not immediately obvious (otherwise the reviewers would have noticed presumably) is not scientific dishonesty. I find that argument reasonable. Now maybe we can argue that such a failure on the part of a supervisor would constitute negligence, but that is not the same thing as dishonesty.

  2. This is a single case, with very specific set of facts, in a single jurisdiction. You cannot generalise from that . Politically, it is a small setback for the role of scientist-experts to determine what is malpractice, in favour of lawyers, but there will be other cases in other countries.

  3. From Marco’s comment it is not clear if Pedersen knew of the manipulations or did not wish to know how images were made, which is still both misconduct. Maybe she was really lied to by Penkowa, but this was what the commission was supposed to determine, and apparently they found misconduct. But their chain of evidence may make perfect sense to scientists and laypeople, but does not necessarily hold in court of law.
    The danger: What is and what is not misconduct is now more and more being determined by lawyers. This is utterly wrong, simply for the difference of approach: lawyers professionally do not seek truths as scientists do, but follow the laws and legal procedure.
    For a lawyer, misconduct must be proven by legally valid facts of breach of specifically defined rules, proven by witness testimony, forensic or bureaucratic evidence, or self-confession. Negligence, or rather the desire not to know how data was produced, cannot be determined as misconduct in this sense, as long as Pedersen never was recorded (not just heard!) saying “I specifically accept if you wish to fake your data”. Even silent condoning of a kind “I don’t wish to know” is not a legal proof of misconduct.
    Thus, courts now determining misconduct instead of institutions or ORIs is good news for all research fraudsters. My two eurocents.

    1. “Maybe she was really lied to by Penkowa, but this was what the commission was supposed to determine”

      No, this is not what the commission was supposed to determine. As far as I can judge from the discussion in the final report, they actually assumed that BKP was not aware of the deliberate manipulation, and then determined she should have noticed in one case.

      Please see the discussion with examples on pages 95-101 (it is in English):

  4. Here are my two bits of info.
    A scientist that should notice something or should not is depending on one crucial factor which is the scientist in question. Let’s face facts here if you saw a picture or a graph and you would trust that individual why would you assume something to be wrong?
    The only time you start questioning is if controls look strange, positive/negative controls are not valid or if the work was repeated by someone else and the results showed something different.
    Most places would not conduct another experiment just to be sure it was done right the first time. Similarly most places would not question the work.
    I know of at least two occassions where work was questioned because the data was ‘too nice’. The one time the results turned out to be fabricated and the scientist in question was fired (after they were filmed in the act). Second case turned out the results were actually real and reproducible and the person in question was talked to about their work procedure (turned out they were just messy in the lab). I think that if misconduct was actually the case punishment is the only course of action. However, if data was used and a small mistake was made (that was identified and corrected), than this is a different story. Mistakes can happen we are human and we continue to do so. To assume we don’t is wrong and to assume that everything we do is a mistake is wrong too. The reality is somewhere in between.
    Now try to teach science to lawyers…

  5. “High Court” is a dubious translation. This is the second of three layers.

    This is what I would call “The Eastern Appeals Court” since it handles appeals from the
    lower level in the eastern part of the country. However, that translation isn’t great
    either since some cases start at this level. A quick scan of http://www.domstol.dk/oestrelandsret/nyheder/domsresumeer/Pages/Pressemed_18_02_15.aspx suggests that this one might have, but it isn’t conclusive.

      1. “High Court” is how the Danish judiciary system translates it.

        The New York Supreme Court, perversely, is third banana, with the Court of Appeals being the end of the line.

        1. I should say, though, that the hed is misleading in exactly the sense that M Welinder suggests; the “high” should have been 86’d. Initial caps wouldn’t salvage it.

  6. If we allow this human venture we call “science” to continue down this road, science will devolve into another huckster-laden venture, run by charlatans with big egos, access to money and powerful connections.

    That someone can be so careless as not to see the poor effort underway under her supervision speaks volumes as to the currently derailing scientific venture. Whatever the excuse, publish or perish pressures, understaffed and underfunded labs, poorly worded Acts on Research Consulting, the growing trail of retracted papers clearly shows the level of scientific effort herein represented.

    If, as elvenc says, “Most places would not conduct another experiment just to be sure it was done right the first time. Similarly most places would not question the work”, this also indicates the degree of derailment of the current human venture of scientific research. The hallmark of science is characterizing repeatable phenomenon, and to disbelieve one’s own lying eyes in favour of data and evidence. Failing to question, and failing to repeat experiments also constitutes gross scientific negligence.

    That a collection of legally and judicially trained individuals issue an opinion on scientifically trained individuals judging scientific conduct is chilling. With the power of the state now in favour of hucksters able to afford teams of lawyers, the quest for natural truths takes a back seat to the quest for power and money. We can look forward to more fake papers touting nonsensical health manoeuvres, because there’s money to be made; and with that money, access to clever lawyers. Incompetent lab PIs will fail upwards, as the cartoonification of science gains legal legitimacy. As long as you can photoshop your way past busy reviewers, or turn a blind eye to such shenanigans by those under your “supervision”, you’re good to go. (Shouldn’t a supervisor have vision super enough to spot fake presentations of images?)

    How many scientific expert witnesses were heard at these hearings, giving testimony about scientific integrity and the responsibilities therein entailed? If anyone can comment on this point, it would be valuable to know how much scientific input was heard before this judicial body offered this opinion. Science is certainly not helped by such rulings.

    The laws of man are made to be broken. Many take delight in doing so. The laws of nature are broken at one’s own peril. A judiciary ruling on how scientists judge scientific competence is setting up the stage for charlatans to profitably hawk dangerous medicines, build weak bridges and buildings, and set back real understanding of natural phenomenon for who knows how long.

    We can laugh now at the Indiana state House of Representatives writing a bill concerning pi, with such gems as “disclosing the fourth important fact, that the ratio of the diameter and circumference is as five-fourths to four;”


    but we can also rejoice that more sensible heads prevailed in the Indiana state Senate, which wisely tabled “Engrossed House Bill No. 246”.

    Let us hope that the UVVU has the money and the will to take this to a higher court in Denmark that will vacate this specious ruling, and remind the judiciary that proper scientific evaluation of natural phenomenon will never ensue from the judicial bench.

    1. “How many scientific expert witnesses were heard at these hearings, giving testimony about scientific integrity and the responsibilities therein entailed? If anyone can comment on this point, it would be valuable to know how much scientific input was heard before this judicial body offered this opinion. Science is certainly not helped by such rulings.”

      The court’s press release specifically refers to three scientists:
      Klaus Bock, chair of the Danish National Research Foundation
      Niels Borregaard, a colleague at the same university, but generally high up the food chain in various foundations
      Knud Juel, considered the “state’s health statistician”, who works at the Danish National Institute of Public Health.

  7. I (as perhaps most of the others expressing opinions here) was not present at Pedersen’s hearing. However, I do not think it inappropriate for judges to decide on the merits of a legal action brought against a deciplinary decision (which may have been unfair) by a university or similar establishment. Who else besides judges specialised in employment law can make such decisions? Of course, those judges need to base their decisions on expert opinion concerning who was responsible for, and who must have been aware of, any scientific misconduct. However, I see know reason to believe, in the present case, that the judges failed in their duty to take into account such expert opinion. For someone to be severely penalised by their employer, an alleged professional fault must be proven beyond reasonable doubt because, in civilised societies, it is considered better to let the guilty get away with something than to wrongly punish the innocent. Therefore, if the judges in the Pedersen case believe there is reasonable doubt that Pedersen was aware of the alleged data manipulation at the time of article publication, then they were right to exonerate her. Their argument that she could not have been expected to have spotted alleged data manipulation, committed by an dishonest colleague, if experts in the field also failed to notice this for several years after publication, seems to me perfectly valid.

  8. What law was violated when a committee of scientists issued a report documenting incompetent scientific oversight by the head of a scientific laboratory? What employment law was transgressed?

    It is one thing for a court to find against an employer for firing an employee for a reason for which a law exists that has been violated, such as firing an employee because the employee has a certain sexual orientation when there are anti-discrimination laws on the books prohibiting dismissal based on sexual orientation.

    Beyond that, if an employer finds an employee incompetent in their job performance, issuing a report on said employee should not be an action that lands the employer in a court, paying costs.

    Is there a law that prohibits issuing a report documenting job performance? I can not read Danish, so I can not read the court documents to find out if such a statute exists in Denmark.

    What I find problematic is a bunch of judicial employees (judges) telling scientific employees what constitutes scientific incompetence.

    Scientific incompetence in Stalin’s Russia involved disagreement with Stalin’s close ally Lysenko. Russian scientists had to pussyfoot around, saying nice things about Lysenko’s silly opinions, even when Lysenko’s opinions were clearly scientifically inappropriate. A group of geneticists who submitted a letter protesting the state of affairs ended up in prison.

    See “Stalinist Science” by Nikolai Krementsov, p79: “As had been the case with the Great Terror two years before, these arrests proved strategically damaging: they removed the most distinguished spokesmen for genetics, including Vavilov, Karpechenko, and Levitskii, and undermined genetics’ links to the control apparatus. Not surprisingly then, following these arrests, the genetics spokesmen were replaced by their competitors. Lysenko himself came to head Vavilov’s Institute of Genetics . . . ”

    Science was not well served by such political and judicial interference. Some of those scientists did not reappear outside of the prisons to which they were relegated. Vavilov died in prison.

    There is plenty of evidence that scientific efforts were handled extremely poorly in this Danish episode. That a group of judges think they know science better than the members of the committee on scientific dishonesty does no service to science.

    If anyone can read the Danish and report on what law was violated by the committee, that would be helpful information in this affair. When the judiciary starts pushing scientists around, we should all get involved.


    “Decision of 28 August 2014 on reuse of biopsy material, research protocol, image manipulation, etc. (reopening of decision of 18 December 2013)

    18 February 2015

    The Defendant has brought the case before the Danish courts and the Eastern High Court has issued its ruling on 18 February 2015. The High Court generally finds that based on the evidence, there is not sufficient grounds for concluding that the Defendant comitted scientific dishonesty. The main conclusions of the ruling are:

    The Defendant’s claim, that certain members of the DCSD was not permitted to take part in the case, is dismissed

    The DCSD is aquitted from the Defendant’s claim that the Committees decisions of 18 December 2013 and 28 August 2014 is invalid in so far as the decisions conclude that the Defendant comitted scientific dishonesty

    The Defendant’s claim, that the Committee’s decision of 28 August 2014 be repealed in so far as the decision concludes that the Defendat committed scientific dishonesty, is sustained

    The DCSD must pay partial costs of the case

    The Eastern High Court has published a summary of the ruling on their website

    Østre Landsret har givet [DEFENDANT] medhold i, at [DEFENDANT] ikke har handlet videnskabeligt uredeligt (in Danish)

    DCSD have issued a ruling in a case pertaining to an earlier ruling on 18 De-cember 2013, where scientific dishonesty, i.e. research misconduct, was confirmed in six articles on health science. The DCSD reopened the case based on new information.

    In the reopened case, the DCSD found the Defendant guilty of scientific dishonesty in 4 of the 12 articles covered by the complaint in the form of:

    Lack of information about biopsy material in 2 articles entailing that the interrelationship between results was concealed from the reader
    Lack of information about biopsy material in 2 articles entailing that a selection of test subjects was concealed from the reader
    Lack of information about biopsy material entailing concealment of the fact that a group of test subjects was subject to a different research protocol than the one described in the article
    Failure to respond to obvious image manipulation in an article

    DCSD found that the Defendant had acted intentionally in relation to the lack of information about biopsy material in 3 articles. In addition DCSD found that the Defendant had acted grossly negligently in relation to the lack of information about biopsy material and therefore the fact that a group of test subjects was subject to a different research protocol than the one described in the article. Furthermore DCSD found the Defendant joint responsible for the image manipulation as the Defendant had acted gross negligently as leading author of the article by failing to respond to the image manipulation.”

    1. “What I find problematic is a bunch of judicial employees (judges) telling scientific employees what constitutes scientific incompetence”

      Steven, they did not rule about scientific _incompetence_, but declared scientific _dishonesty_.

      Of direct relevance here is that this same DCSD in 2003 declared a book, but not its author, Bjørn Lomborg, to be scientifically dishonest. The argument used was that Lomborg was not competent enough in the areas he wrote about to determine that he must have deliberately misrepresented the data. In other words, the DCSD has earlier ruled a scientist (Lomborg) to not be scientifically dishonest _because_ he was scientifically incompetent. That decision was rescinded by the Ministry under which the DCSD resides.

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.