Malicious use of procedures for research integrity, or a persistent attempt to rid the scientific literature of a fraudulent study?
In January 2019, Harald Merckelbach wrote in his monthly column in Dutch newspaper NRC about lobbyists’ attempts to muzzle scientists by harassing them with complaints to their university. It inspired the KNAW (the Royal Dutch Academy for Sciences) to organize a symposium on abuse of the procedures for research integrity that each university has put in place, and how to deal with it.
What is striking in the report of the symposium, which took place on 25 November, is that those present seem to know only one concrete example: a biologist who would stalk you with an ‘incoherent story’ about a study on the Basra Reed Warbler that should be retracted. Some even express their suspicion that he suffers from psychiatric problems.
The question whether this plaintiff is out to frustrate unwanted research – the subject of Merckelbach’s column – is not addressed. Nowhere does the reporter for Observant wonder whether this person might have a point after all. And apparently, no one thought about asking him to tell his side of the story.
I made up for that omission by interviewing the complainant, named Klaas van Dijk, by telephone on 3 January 2020. The interview below was published in Dutch on Kloptdatwel.nl on 10 January. [NB this translation was done rather quickly, I might polish it up somewhat later]
Little brown bird
If I understand correctly, the problem is one article from 2013 about a bird that occurs in Iraq. What exactly is this study about and what is its importance within the field of ornithology?
It is about a study of the breeding biology of the Basra Reed Warbler and also concerns a short reaction by the authors to a comment on it, two years later. In the Netherlands you have species such as the Blackbird, the Great Tit and the Mallard, on which many studies have been carried out into all possible aspects of their breeding biology. When do they start breeding? How many eggs do they lay? Are they monogamous or polygamous? In the Netherlands, extensive studies on the Great Tit are being carried out in nesting boxes since a hundred years ago. You can easily pass them by bike for inspection.
Elsewhere, such research is often much more limited. About this species, which only breeds in Iraq, nothing was actually known. Just some loose observations and a few pictures.
And then suddenly, out of the blue, a study appears that is based on a thousand nests, and a few hundred cases that are about polygamy. It would be fair to say that it is a groundbreaking study. Not that that species wasn’t known, but a lot of basic details of the breeding biology were unknown.
If we could translate this to a Dutch context, do we then have to think of a bird like the Great Tit, a well-known species?
No, it is also an unknown bird in Iraq, but the exceptional thing is that it breeds almost exclusively in the reed marshes of southern Iraq. Because it is a species that only breeds in one place and because there are not many of them, such a species can become extinct. There are international agreements that stipulate that you should use the scarce research time for species that are endangered. And that is the case for this Basra Reed Warbler, because it only occurs in a very limited number of places. It has a special status, actually from the moment it was well described.
We have similar bird species breeding in reed marches such as the Oostvaardersplassen or the Biesbosch. Think of the Eurasian Reed Warbler and the Great Reed Warbler, the Sedge Warbler and the Marsh Warbler. These are just very vague little brown birds, which are difficult to tell apart.
This particular study, of which the Iraqi Omar Al-Sheikhly is the first author, is fraudulent according to you and quite some supporters. What are the clues for that?
The main clue is that the raw data do not exist. Requests for access remain either unanswered, or threats are coming. Well, then you can be pretty sure something’s wrong.
The second clue you can find in the following sentence in the article where the authors talk about polgygamy: “Males are often polygynous (42.9%, n=317 observed males)”. It’s just one sentence, but in science it’s always about nuances and precision. You have to be able to substantiate such a percentage.
If you want to say something about polygamy, then of course you have to know who is male and who is female. Sounds very simple. You can see it in people, but what about birds? With Mallards it’s easy, because their plumage is different. That’s not the case with reed warblers. You also have species where you can tell them apart by size. Can’t help you here either. So how would you determine sex in the Basra Reed Warbler, such a little brown bird?
Then you almost have to catch them.
Exactly. So you have to catch the birds and ring them with different colored rings around the legs, an individual color code. Then you know this is the one with red-white-blue and with that you can follow it in the field, and based on the behaviour you can keep the males and females apart. Nowadays you can also easily determine sex after catching using DNA. But in that case there must be samples and those are not there either.
Tracking takes a lot of time, but it can be done. Then you see for example: ‘hey, this man is with this woman now, but earlier he was with a female with different colours’. Of such field studies there should be raw data. There aren’t any.
When the article appeared, a number of people thought that this study was too good to be true. So they asked for the data and these were not given. Did it go this way?
That creature breeds in the great swamps of Iraq. Sadam Hussein tried to drain much of it. That’s almost like trying to reclaim the Wadden Sea, a disastrous activity.
Well, at one point Hussein was driven out and Iraq became more accessible again. That’s when recovery projects got underway. The first step is then to determine what species are present. There have been projects in cooperation with Western countries for this work. British ornithologist Richard Porter was involved from Birdlife International. Omar Al-Sheikhly was one of the field staff he had trained.
That collaboration lasted a couple of years, approximately until 2008. And then suddenly, in 2013, this article appeared in a scientific journal. Porter immediately realised that something was wrong.
You might think that when a study appears written by people he trained, his first thought would be: ‘hey nice, wonder what they’ve done’.
Porter also put a lot of effort into getting those people to put their observations on paper. He also knew exactly what this Al-Sheikhly had in store, an excellent photographer. Back in those days he also published a small article, with the idea that almost nothing was known about this species and that at least now he could show some beautiful pictures of it. But then, years later, out of the blue, that new study of his appears.
Porter quickly sent a letter to the editors of the magazine explaining why it could not be true. And then the problems actually arose, because the editor in chief never, ever wanted to admit that he made a huge blunder.
After enormous efforts, about a year and a half later, a reaction was finally published by Richard Porter, who had gathered around him an impressive list of co-authors, you can say the crème de la crème of people familiar with the Basra Reed Warbler.
Clearly not just a bunch of ornithologists stand behind the criticism. Are there any scientists who would give this study the benefit of the doubt?
So we have had this discussion in the magazine, an Expression of Concern was stuck onto the article, but why did it end with these actions? Why wasn’t the article retracted?
That Expression of Concern only came after I published a report in 2016.
Oh, so it’s from later? After Porter’s reaction, Al-Sheikhly’s reply and Porter’s rejoinder? That’s not explained in it at all.
First, for a year and a half, there was an attempt to resolve the issue silently. It was extremely difficult. It was quite a struggle to finally get those three letters published. In the spring of 2015 the issue became widely known in the ornithological community. And at that time I got involved.
I thought we had to focus on getting the raw data. That’s what it’s always about. Until then, they hadn’t put so much effort into this. Based on the other problems we found regarding this article, a report was written on 1 July 2016, and that Expression of Concern was published as a result. Only the publisher just doesn’t want to admit that.
That Expression of Concern was eventually published by the editor in chief?
I don’t know. The problem is that quite soon the publisher, Taylor & Francis, said they would not communicate. And the editor in chief has said much earlier that he will not communicate anymore about this: “I take the view that we had a neat scientific discussion, via that comment and that rejoinder, and as far as I’m concerned the matter is now closed. Period.” And he doesn’t respond to our request for the raw data.
For writing that report, I had to put quite some pressure in my requests to the publisher for the raw data. I had to sound the alarm quite a bit. But finally I got an answer, which was essential for that report.
And that answer was?
Well, that they don’t have to give the data, because such requests are not covered in the journal’s rules. And furthermore: “We’ll never respond to you again. And the way you’ve approached us doesn’t make any sense at all.” It took me a little less than a year to get that short answer to my question: “Show me the data”, which in my view is the essence of science. As soon as there are doubts about studies, in whatever field, you have to go back to the data.
The problem could easily have been solved at that point if those data had been shown. There may still be something wrong with the data, but that does not necessarily indicate malicious intent.
No, of course it doesn’t.
Okay, so Taylor & Francis, one of the largest publishers, refuses your request rather bluntly and actually states that there’s nothing wrong with this research. We are simply asked to believe them on their blue eyes.
Then you and your fellow criticists knocked on the door of the Committee on Publication Ethics, COPE. Or were you alone in this from now on?
When it all came out in 2015, we started working together and we went different ways to get the data. We tried at the university of one of the authors, at the publisher and at the editor in chief. It soon became clear that the publisher did not wish to collaborate and then we approached COPE. Like “Dear COPE, in our opinion these are the arguments that it’s all wrong. And the publisher, a member of your club, refuses to cooperate with us. That’s why we’re filing a complaint.”
This went nowhere pretty quickly, because COPE refused to look at our criticism in a substantive way and refused to do anything with our request to help us getting access to the data. Instead, they started to send me threatening letters.
So far, I can still imagine your steps. First you expect the editor in chief to act appropriately, but he refuses to do so for some reason. Then you go up a level and knock on the publisher’s door. It turns out they are also unwilling to do something, and then finally there’s something like COPE. But when it appears that also COPE doesn’t want to do anything, a lot of people would leave it at that, I think. Why continue now anyway?
We’ve been acting along different paths for a quite some time. I focused on getting that raw data, while Richard Porter and others tried to talk to the publisher, behind the scenes. That didn’t work at all. Behind the scenes there were also conversations with Al-Sheikhly.
So far I don’t get the impression of someone who is on a crusade and goes beyond all limits. It’s already exceptional that someone takes so much trouble, but it’s still within what you could call the normal procedures.
I think correcting mistakes is one of the crucial things in the practice of science. You can go on endlessly publishing new things, making that pile higher and higher. But you also always have to make sure that that structure, where all those new building blocks come on top, is solid. And if there’s a rotten piece somewhere in there, you just have to get rid of it. You could say that it is not so annoying at all, it’s not important, but in the end your structure collapses.
That motivation doesn’t get any attention for instance in the report of the symposium. Nor that you aren’t the only person who worries about this study. So you make a strong case for cleaning up science. It has to be stripped of clearly wrong studies.
More people are working on that. Look at the issue of food scientist Brian Wansink. How much time did people put in that case? And you have Nicolas Guéguen in France. I’m now working behind the scenes on the Ad van Liempt case. How much time do you think Bart FM Droog put in? That just takes a lot of time. And most people don’t have that time, or they have other priorities.
Let’s go back to why this has become such a loaded issue. I myself received a number of your mails as cc, have also seen a single indignant response from an addressee. Without going into the details, it seems to me that this indignation lies mainly in the fact that you address people about their responsibilities which come with their position, for example a board membership of COPE. If they don’t react, or hardly react at all, you throw at them that they apparently agree with what that organisation does, or better, fails to do.
That’s right. I can do a lot, because I don’t depend on anyone. What this has to do with is abuse of power. I only refer to the Dutch code of conduct for science. It simply states in black and white that you have a duty towards science in general and the researchers in your vicinity. Admitting and concealing misconduct by colleagues is one of the eight examples of violations of scientific integrity. You may like it or you may not like it, but that’s what the code says.
The preamble to the code of conduct states that it is also intended to address each other on these issues. If some ordinary guy, a mister nobody, does that, they don’t like it. That’s the whole story.
COPE has a problem getting tough on Taylor & Francis. You can easily find out that this publisher has to pay about £50,000 a year to be a member. The more journals a publisher has, the more money it has to pay. So if COPE would say to Taylor & Francis “we’re gonna punish you” they’d just stop paying.
I can’t prove that this is how it works, but it is true that there are many more examples of people who have knocked on COPE’s door and just got stonewalled.
Such stonewalling could be seen as ‘looking away’ from the real problem.
The Netherlands Board on Research Integrity (LOWI) writes about one of your complaints: “Complainant shows a pattern of behaviour which resembles a snowball effect: Complainant submits a complaint about person or body X, after which the handling of that complaint by person or body Y regularly leads to a new complaint from the Complainant, but about person or body Y, and so on”.
What they’re talking about here is an old issue, which by the way has been completely resolved through mediation. But exactly the same thing happened there, namely referring questions to others or not responding at all. And I don’t take that.
Such a pattern of behaviour may be called annoying, but it is not a judgement on whether the complaint, and the subsequent complaints, have a ground or not. LOWI pretends that such a snowball of complaints in itself proves that the complainant is wrong, regardless of what the complaint is about.
That’s looking away. And that’s a very big problem. Not for me, but for them. As soon as they are confronted one-on-one with exactly what they are obliged to do according to those codes of conduct, they will have to turn around. You can’t walk away from that.
I can imagine that it is difficult to be held accountable for a responsibility that you yourself might see more as an abstraction. Now, if I have a direct colleague who I can see is involved with misconduct OK, I’ll have to deal with that. But if I’m a member of a board of a high level body that talks a bit about scientific integrity and someone complains about a study that already has an Expression of Concern attached to it … well, you can’t expect me to do something, can you?
In all those documents they’ve carefully avoided to talk about the content, and about our simple request to make a statement about the raw data. And they’ve carefully avoided even referring to the extensive report I’ve circulated. That did not happen for nothing.
Apart from that, I’ve also had conversations or e-mail contact with confidential counselors who know how it should work. Really honest people. Unfortunately, that’s not always the case. That’s because there is almost no public control in this whole scientific integrity business.
Look at a municipality, for example. As soon as they decide to cut down a tree, there will be angry residents to be reckoned with, they are outspoken and know how to find the press. That system doesn’t yet work well with scientific integrity, they’re not used to it yet. You still see a lot of abuse of power: ‘I am the big boss, and who are you? Away with the complaint!’
How can this issue be brought to a somewhat satisfactory conclusion? Is a retraction the only possibility? This only seems to happen if the article itself is manifestly wrong, or if the institution to which the fraudulent author is affiliated requests it.
There are three ways. There are of course editors who do take their responsibility, the Journal of Biological Chemistry is perhaps the best example. As soon as they notice that something is wrong, it’s “Dear authors come up with a valid explanation or the data.” And if it remains vague, it’s bye bye to the article, retraction.
Authors and co-authors can take the initiative. It may well be that a co-author did not know that the research was partly based on fraud. But then you expect someone like that to correct his mistakes as soon as he is aware of them.
An example of this has just made the news. Nobel Prize winner Frances Arnold withdrew a study from 2019 because the results were not reproducible and the experiments were not neatly recorded. Arnold has done exactly what you have to do in such a case. The research has to be in a lab book, if that is not there or if there are missing data, then you only have one option.
The Saudi second author of Al-Sheikhly (2013) has passed away. It is doubtful if he knew what happened. It is also striking that the exact name of his university as mentioned in the article, does not exist at all in Saudi Arabia. Richard Porter and others had already persuaded Omar Al-Sheikhly to have him retract the article. But then the Italian co-author put a stop to this, he doesn’t want to retract the article.
Also a university can request a retraction. In the Dutch situation, if there is an affiliation with a university mentioned for one of the authors of the article, that university bears responsibility for it. And in principle that resposibility lasts forever. We have tried to follow that path at the University of Pisa, to which the Italian researcher is affiliated. But even with colleagues as intermediaries, he didn’t want to react. I filed a complaint in 2017, very neatly, nicely formal. No response.
Right to Complain
The reason for this interview is the report of the symposium on misuse of procedures for research integrity. Reading this one could easily get the impression that your fight is an example of how you should not use the right to complain, that it is a malicious complaint.
That is abuse of power. You can’t just unilaterally decide to deprive someone of the right to complain. That’s what they did. I don’t mind. The last word has not been spoken on it yet.
The only example the workshop participants seem to be able to put forward as an example is your case. As if it were about obstructing research.
All those issues would have been solved if all those people had just read that report and reacted with ‘yes indeed, I don’t know much about it, I’m not an ornithologist, but that raw data is a problem and it looks very much like it is based on fraud’. Nothing would be the matter, but they chose not to act like that.
COPE, you say, turns out to be a sham. Of course, it’s fine that publishers set up a body to discuss integrity issues, but don’t we lack a truly independent body that can handle complaints?
It’s fine for them to make guidelines. They have principles, but they get angry when you want them to uphold those. That’s because of the big difference in power.
In the Netherlands we have the example of the nitrogen deposition dossier. Huge power games were played at a political level. It had to be pushed through in a certain way. It succeeded. But there was one person who simply made use of his legal possibilities, his right of complaint. That best man said nothing more than: “The law says A and I think you do B. Mr. Judge, can you take a look at it?”
That’s kind of similar. However, there we see an independent judge who could give a binding ruling on the nitrogen issue. In science there isn’t one.
I’ve thought about it for a long time and now I know why. Look at medical disciplinary law. If a doctor makes a serious mistake, it’s not covered up. It’s right there in the newspapers. Then there will be a complaint and if the hospital doesn’t pick it up properly, you go to the medical disciplinary board and there will be a public hearing. Everybody can attend.
One problem with the complaints system for research integrity is that it’s all secret. Secrecy. I’ve been getting nasy mails. “If you do this, we’re going to stop the investigation in this case immediately!” That doesn’t make any sense at all. You can only get out of this by doing it in public.
And there are good arguments for that, because in almost all cases it concerns scientists who are fully paid by public money. Surely everyone has the right to see how they defend themselves against reasonable arguments that they have not properly obeyed the rules?
The counterargument is always that this would damage someone’s career. But that’s exactly the same with doctors. And you still have the legislation on defamation and slander. If you really do make a malicious complaint, then that would fall under defamation and slander, doesn’t it?
The symposium report talks about “dragging through the mud” and even stalking, but I don’t think they actually sued you for that. Do they exaggerate? Because this almost looks like slander.
The journalist didn’t check anything with me! For example, that article contains a very concrete statement by Yvonne Erkens, the chairman of Leiden’s committee for scientific integrity, that I would have sent as many as fifty e-mails in one day. I asked the journalist to prove that, just show me those e-mails. No response.