Academic Rights Watch is stunned by an investigation into alleged abuses at the Department of Astronomy and Theoretical Physics in Lund, where two prominent professors are accused by a co-worker of this and that, and where a shoddy investigation leads to their reputations being destroyed forever after an article in Nature. The post is the first in an ARW series about the affair.
The Department of Astronomy at the relatively newly formed Department of Astronomy and Theoretical Physics in Lund is something of a scientific success story. The latest evaluation of research at Lund University, RQ20, states that it is “recognised internationally as a strong research group.” Recently, it has hosted an ERC grant and heavyweight Wallenberg projects. In Lund’s publication database, LUCRIS, 1400 publications are registered in the department.
The department owes much of its success to a number of distinguished professors, several of whom are internationally renowned, not least Sofia Feltzing and Melvyn Davies. Both are highly cited researchers who have attracted large amounts of external funding over the years, which has also benefited many junior researchers in the department. Feltzing, who is a member of the Royal Swedish Academy of Sciences, and Davies are, incidentally, an academic couple.
At the same time, RQ20 points out that astronomy, like other physics subjects, must contend with a high dependence on external funding, while the grant rate of various funders is declining. Perhaps the stress this might cause is a piece of the puzzle in the unlikely story we will tell in this series.
That there were cracks in the wall became clear after a health and safety survey conducted in spring 2020, in which several employees expressed dissatisfaction with the working environment. However, the protests reached a new dimension when in September 2020 a researcher, Florent Renaud, formally reported both Davies and Feltzing for “harassment, bullying and abusive behaviour”, i.e. essentially unfair treatment and discrimination.
In the complaint, Renaud counts many alleged violations, 32 to be precise. The tone is affected and there is no documentation. The allegations range from being criticized at a seminar to being asked to attend a departmental meeting. In several cases, the accusations are slightly bizarre, as when Davies is accused of never making coffee:
On countless occasions (almost everyday when we are both present in the office), when Davies arrives in the morning, he asks me why I haven’t prepared his coffee. When coffee is already made, he systematically asks who prepared it and takes a cup. lf coffee is not prepared, he will wait until someone (usually a student) makes some, and then have a cup. I have seen him drinking coffee every single day I have seen him at work, and I have never seen him preparing coffee.
Renaud has clearly taken it as a personal attack that Davies is not more involved in the department’s coffee-making. (Davies claims in a comment that he contributes to the department in this respect as well as his scientific activities.)
Renaud also believes that Feltzing stalks him in the corridors. To escape her, he writes, he hides in one of the toilets ” they are the only room on our floar she won’t be able to open with her key or access card.”
The impression the reader gets is that the report is not entirely coherent. In the complaint, Renauld himself talks about severe psychological stress, which he believes was caused by the professors, but which, of course, could have other reasons.
In any case, the complaint led to an investigation by two organizational consultants at the company LifeWise: Martina Johansson and Mia Gruvstad. Johansson is a psychotherapist and Gruvstad a psychologist. It does not seem to occur to them that Renaud’s problems may have other causes, but the investigation focuses entirely on Davies and Feltzing.
Despite Johansson’s and Gruvstad’s laudable intention to conduct a “fact-based investigation”, questions arise early in the investigation about the methodology used.
One question concerns the persons heard as “witnesses”, as it is not clear how they were selected. Some in the field of astronomy are on the list, e.g. Oscar Agertz, others have apparently not been heard, e.g. David Hobbs. In a fair dispute, both parties call their own witnesses. This has clearly not happened here. So the question remains: why have these particular witnesses been selected and who was responsible for their selection? How has it been ensured that it is not just Feltzing’s academic enemies or competitors, or friends of the complainant for that matter, who have been allowed to testify?
In the following, the investigators do not assess the credibility of the witnesses, but consider it sufficient for a single witness to support the story in order for the accusation to be considered “proven.” It is sufficient, the investigators seem to think, that the witness says something vaguely reminiscent of Renaud’s description, even if it is not quite the same thing.
A consequence of the method is that if the accusations are sufficiently numerous, as in this case, the probability becomes close to 1 that some witness will at some point say something vaguely reminiscent of the accusation so that the accusation can be considered “proven”. This need not be because Davies or Feltzing have done anything other than their duty as academic leaders, for example by setting high standards for themselves and their staff, but simply because the standard of proof is so low.
Another serious error made by the investigators is to describe Davies and Feltzing as “the accused” in the investigation, implying that they would stand trial and that, if “convicted,” they could be considered criminals. The environmental legislation to which the investigators themselves refer explicitly states that it cannot be used to impose individual guilt: “The purpose of work environment legislation is to prevent ill health and accidents, but it does not regulate compensation or guilt issues.” (AFS 2015:4, p. 15).
If anyone is “accused” in this case, it is Lund University, which has a duty to prevent work environment problems. As the Work Environment Act (AML 2 §) states: “The employer must take all necessary measures to prevent the employee from being exposed to illness or accidents.”
The investigators have thus misunderstood the purpose of the work environment legislation, which means that the whole thing is degenerating into a kind of extrajudicial show trial. This does not bode well.
The first issue under investigation is whether Feltzing committed victimisation, i.e. “[a]ctions in an abusive manner at one or more employees that may lead to ill health or to exclusion from the workplace community” (AFS 2015:4, p. 6).
The first incident concerns Feltzing’s questions during a presentation given by Renaud at a seminar (GalForm Meeting). As it was a seminar, there should have been many witnesses in the audience. However, only two witnesses, Eric Andersson and Anders Johansson, are mentioned in the investigation. One wonders what happened to all the other witnesses to the event. Why were these two chosen?
Feltzing explains in her commentary that she was in a bad mood that day and that this may have contributed to a bad atmosphere. She confirms that the purpose of her questions during the lecture was to understand the details. She adds: ” So no, I never intended to diminish or minimize his performance. Just to understand.” However, the investigators attach no weight to Feltzinger’s comment, but conclude that the allegation “can be proven” (p. 11):
The accused confirms this statement made by the claimant. The witnesses confirm that the situations implying victimisation have occurred. The alleged claim of victimisation can be proven according to an evaluation af the facts.
Already here it gets wrong: That Feltzing would have confirmed Renaud’s claim of discriminatory treatment is simply wrong. After all, Feltzing has explicitly stated that she did not intend to offend him.
A number of questions arise here: Is it enough to have a bad day to be “convicted” of discriminatory treatment in a work environment investigation? Hand on heart: how many of us have had a bad day and taken it out on someone else? Why should Feltzing be “convicted” of this? Isn’t that what selective justice looks like?
And it goes on and on. Take Renaud’s accusation that he was “excluded” from a research project. He seems to think that it is almost a human right to be included in a research project. This is not the case; who is included is normally decided by the research director or other senior researchers on the basis of an assessment of what skills are needed.
The two witnesses who, for whatever reason, are allowed to testify recognize this. They are the same Eric Andersson, but now also Oscar Agertz. Both believe that Renaud should have been involved on the basis of scientific judgements they make about his relevance to the project. The problem is that Andersson is a PhD student and Agertz a research assistant, while Feltzing is a professor and may be considered much more competent to decide that question. The investigators, who do not seem to understand how academia works, nevertheless again conclude that Feltzing is guilty of discriminatory treatment, this time through “exclusion.” The bottom line is that professors are no longer allowed to decide who participates in research projects. This is a major intervention in academic practice by two consultants with no understanding of academic norms.
Feltzing is also said to have “controlled” Renaud by asking him to attend a departmental meeting. Feltzing comments that she may have knocked on the door and said it would be good if Renaud came, but denies that this would constitute discriminatory treatment. Two witnesses comment on this. Loke Lönnblad says that he does not remember anything. The other witness, Maxime Delorme, says that it was much as Renaud had described it.
Again, Feltzing is “convicted” against her will after a witness states that she experienced the situation in the same way as Renaud. Word against word. Why this person’s testimony outweighs Feltzing’s testimony to the contrary and is sufficient for it to be “evidence” is not clear.
Not even the investigators believe twelve of the allegations, despite the extremely low standard of proof. But Feltzing is still “convicted” in six cases. Davies, for his part, is cleared of all charges, including that of failure to make coffee.
The investigation’s incorrect choice of method and terminology makes it appear to colleagues that Feltzing, “the accused,” has been convicted of a crime and is now considered a criminal. This would, however, be a gross misunderstanding of the law.
In plain language, this means that the investigators themselves are engaging in discriminatory treatment of Feltzing, in a way that comes close to defamation. Feltzing, if anyone, is here subjected by the investigators to “[a]ctions in an abusive manner at one or more employees that may lead to ill health or to exclusion from the workplace community
The fact that investigations of this type can be directly harmful is warned against in the “Guide for implementation of the Swedish Work Environment Authority’s provisions concerning the organisational and social work environment, AFS 2015:4,” p. 52:
An inadequate investigation of victimisation may be harmful to both the work environment and employee health. The situation should therefore be investigated by someone who has adequate expertise, is able to act impartially, and is trusted by the involved parties.
We do not know to what extent the investigators were able to act impartially. It seems likely that they initially enjoyed the confidence of those concerned, but it is also clear from Davies’ and Feltzing’s comments that the process disrupted that confidence. That the investigators lacked competence for the task and conducted a “substandard investigation procedure” can hardly be doubted.
In the next section: new complaint leads to new substandard investigation in which Davies is also “convicted”, article in Nature, the media trial against Davies and Feltzing starts on Twitter.