There is no need to highlight the irony of Francesca Gino, a Harvard Business School researcher of dishonesty, among other topics, being accused of data falsification in a number of her publications.
It does perhaps add to the irony that, in writing about a recent episode of the HBO series Succession, she praises resort to “motivated reasoning”, defined as “a cognitive process in which people use their pre-existing beliefs to interpret information in a way that supports those beliefs”. Similar to “confirmation bias”, this is indubitably a prominent ingredient in the recipe for research misconduct.
Equally, we should be wary of confirmation bias ourselves. Psychological sciences are notorious for shady statistical practices and irreproducible results, and it has been speculated recently that some researchers of dishonesty might have a predisposition to cheating.
The three-person Data Colada team that claims to have uncovered anomalies in the data underlying several of her publications were evidently very sure of their conclusions – and Harvard agreed, punishing Gino with a two-year unpaid suspension. Equally, Gino is not taking this “unwarranted and excessive” punishment lying down, arguing in a lawsuit against both Harvard and Data Colada that there is no evidence she personally manipulated data and accusing Data Colada of engaging in “a vicious, defamatory smear campaign” against her.
But if Gino is feeling like her life is on the line, so must the Data Colada team. I should know because I am one of the few individuals who have been subjected to the ordeal of a defamation lawsuit for alleging violations of academic norms in publications.
Nearly a decade ago, duplication of data within an image was brought to my attention. I investigated many articles by the researchers associated with the laboratories where the experiments in the article were conducted. There were numerous examples of image duplications and plagiarism.
I reported my concerns, in my own name, to the journals where the articles had been published, indicating the problems without making assertions about who was directly responsible. When the publications took either no or inadequate action, I contacted The New York Times. It published a front-page article incorporating my findings and focused on the prominent scientist who was an author on many of the articles.
He sued the newspaper, his own university (which had removed him as departmental chair) and me. In each case, the lawsuits were dismissed because of insufficient evidence at summary judgment, but the defamation lawsuit against me was active for five years.
Indeed, the ramifications are ongoing. The plaintiff did not pay his initial two sets of lawyers their entire fees and the first have successfully sued him, winning more than $1 million; his Baroque art collection has reportedly been seized to pay the judgment against him.
So the Data Colada team need to be prepared for a long process. One of the consequences of defamation cases is to interrupt communication between defendants and others with whom they might want to discuss what they have unearthed because it will be subject to disclosure to the plaintiff. There is no doubt that this makes defendants less willing and able to exercise their free-speech rights, in part because of their necessary reliance upon the cautious approach of their attorneys.
On the other hand, the process of legal discovery can be a wonderful tool. In my case, it revealed that the researchers had acknowledged among themselves that there was misconduct in the generation of that very first image brought to my attention, which was featured in The New York Times, but which they had denied to the journal was problematic. I also learned of efforts, which involved paid consultants and have not ceased, to discredit me and others who tried to hold the authors of the flawed articles to account.
This last disclosure highlights a central consequence of being sued and why these types of lawsuits are so damaging to the academic enterprise. Instead of the focus being placed on the validity of the data in dispute, the controversy becomes personal, and there are threats to the careers of those who were trying to ensure the integrity of the scientific literature. It is this outcome that has led to other researchers setting up a GoFundMe account that has raised more than $300,000 for the defence of the Data Colada trio. Time will tell if that defence succeeds or not.
In my case, my university agreed to pay my expenses at least as far as summary judgment. But even high-quality representation is not sufficient protection against the collateral damage of being a defendant.
Modern universities regard researchers’ purposes primarily to be to produce research and fill the coffers through grant-getting. They do not regard good-faith whistleblowing about anomalies in the literature to be an important part of a scientific career. And, for me, it was this discouragement of my activities and lack of recognition of their value that was the most damaging fallout from the legal action.
This attitude needs to change. Whatever the rights and wrongs in particular cases, seeking the correction or removal of apparently faulty research is surely just as significant a contribution to the literature as making new additions to it. But even if lawyers will now only act for defamation plaintiffs who offer to pay up front, universities must offer better support for such work – both financial and, especially, moral. Otherwise, good-faith whistleblowers will remain few in number and somewhat faint of heart.
David Sanders is associate professor in biological sciences at Purdue University.
