The ETH Case – Part 1

ETH Zürich: The scene of Machiavellianism and entrenched power battles.

Systematic Failure

Systematic Failure

An astronomy professor at Zürich’s ETH university has been accused of harassing doctoral students for years. She’s about to be fired, but the question of her purported guilt has never been clarified. How a world-class institution betrayed everything: the professor, the presumption of innocence – and even itself. A reconstruction of the ETH case, part 1.

By Silvan Aeschlimann, Dennis Bühler, Dominik Osswald (Text), Dominic Nahr (Photos) and Paul Cohen (Translation), 27.03.2019

Teilen0 Beiträge

The ETH has demanded a counterstatement («Gegendarstellung»). It is situated at the bottom of this article.

Sunday morning, October 22, 2017. The mild late summer came to an abrupt end overnight. Still lying in her bed, Marcella Carollo reached for her tablet, checked the online news portal – and felt her heart leap into her throat.

«My first thought?» Roughly a year later, in December 2018, Carollo is sitting in her living room and the atmosphere is bleak. The landscape outside her window is firmly in the grip of the fog, and the neighbors’ house is barely visible. The shutters of her home in the countryside outside of Zürich are usually kept closed anyway. The house is richly furnished with heavy carpets, antique furniture and black-and-white photos.

«At first there was just emptiness», she recalls, «then a stabbing pain. I had trouble breathing and it was as if, in a matter of seconds, I was being sucked into a deep hole. My husband held me in his arms, I don’t know for how long. At some point, my first thoughts formed: How can we go on like this? Can we even go on at all?»

What Carollo read on that Sunday morning changed her life forever: «Scandal at ETH: Professor Harasses Students» was the headline in big bold letters on the front page of the NZZ am Sonntag. The newspaper gave the professor a fictitious name to render her anonymous, but it only took a couple of clicks online for anyone to figure out who «Gabriela M.» really was: Marcella Carollo, 54 at the time, a leading member of the faculty of the Institute for Astrophysics at ETH in Zürich since 2002 – was reportedly a bully.

The article stated that for more than a decade, she allegedly harassed her subordinates, referred to women as weaklings who should spend less time on their make-up and more on their research, and micromanaged everything under the sun while demanding superhuman commitment. «I consider this a dark episode in my life», the article quoted one of the former doctoral candidates as saying.

The news made headlines around the world. «Der Spiegel», the «Süddeutsche Zeitung», «BuzzFeed» and «The Wire» all reported on the allegations of professional misconduct at ETH.

The evidence against Carollo appears overwhelming. Between December 2016 and May 2017, current and former doctoral students and postdocs wrote a total of 12 secret testimonials outlining the ordeals that they experienced at ETH. They all shared similar stories. Astronomy professor Marcella Carollo reportedly harassed her subordinates, placed excessive expectations on their availability and engaged in unfruitful discussions until late in the night.

«Several of us at the time were talking of ourselves sarcastically as ‹their slaves›, which described our feelings pretty much to the point», according to one of the testimonials. In another account: «This woman steals your work, your dreams and your life. She plays with you like a puppet in a theater, and then she breaks you down.»

When fragments of the testimonials were leaked to the public – despite strict secrecy – on that Sunday in October 2017, ETH suddenly became the scene of «harassment» and «abuse of power» in the upper echelons of academia.

The institution still hasn’t been able to shake this reputation, but now, presumably, it has acted to clear the air. On March 14, ETH requested Carollo’s dismissal. It would be the first time in the 164-year history of this institution of higher education that a professor would have to take leave of their position in this way.

Does that close the matter, as ETH officials hope?

Republik has been investigating this case for half a year, has spoken with the individuals involved, read the minutes of hearings and pored over emails and letters – roughly 3000 pages of documents in all, including 10 of the 12 secret testimonials that are only accessible to the inner-most ETH circle. Carollo is not a member of this select group. And to this day, she has been denied access to the original wording of the allegations against her.

Is the professor guilty?

Or is she the victim of a «vindictive doctoral candidate», as she says?

Or does the truth lie somewhere in between these two extremes?

Nobody knows for sure because no one has mediated the conflict or investigated the allegations: neither the head of department, the office of the vice rector, the college administration nor the external administrative investigation. Professor Carollo’s guilt appeared to be a foregone conclusion right from the start.

The investigation by Republik has exposed systematic failure at ETH. An examination of the process leading to Carollo’s dismissal reveals serious administrative and procedural deficiencies for an institution that is proud to rank among the world’s top 10 universities.

The Case of the Abusive Professor – an ETH Case.

Over the past two years, Carollo has not once benefited from the principle of innocent until proven guilty. The professor has never had a genuine opportunity to explain herself and her work, to concede mistakes and make improvements. The conflict between a doctoral student and the professor degenerated into chaos as players at all levels of the institutional hierarchy began to pursue their own interests – and even ended up toppling the university president.

Ultimately, they could only agree on one thing: Carollo, one of the most cited scientists at ETH Zürich, had to go. Now, she is to be dismissed – without official prior warning – although there are no provisions for such action in the university’s personnel statutes.

But Carollo is fighting back. The final report from the administrative investigation states that Carollo showed little understanding for the procedure and never recognized that she was part of the problem. And to underscore her position, Carollo speaks for hours and days on end, in an endless volley of words. The 56-year-old insists that she has done nothing wrong. «All I want is for the truth to come to light», she says.

0:00 / –:––

How could it have come to this point? The following reconstruction describes in six acts how one of the most renowned institutions in Switzerland, subsidized to the tune of 1.3 billion Swiss francs (1.15 billion euros) of taxpayers’ money every year, maneuvered itself into one of the most humiliating situations in its history.

The leading roles are played by:

  • Marcella Carollo, professor of astronomy: allegedly harassed doctoral students, leading to calls for her dismissal

  • Elisabetta Marignano (not her real name), doctoral student: complains to the ombudsman about her adviser, Carollo, and sets everything in motion

  • Wilfred van Gunsteren, ombudsman: is determined to push for ethical standards at ETH and plays a highly active role in the so-called «Carollo case»

  • Antonio Togni, vice rector: is a staunch advocate of better supervision for doctoral students and sides with the ombudsman, van Gunsteren

  • Rainer Wallny, deputy head of the physics department: reorganizes the physics department and emerges as one of the few winners in the Carollo case

  • Kevin Schawinski, assistant professor of astronomy: is striving for a full professorship and levels serious allegations against Carollo

  • Simon Lilly, professor of astronomy and Carollo’s husband: is never accused of any misconduct, but is nevertheless dragged down with his wife

  • Lino Guzzella, ETH president (2015–2018): amid allegations that he tried to «sweep things under the carpet», is forced to forgo a second term in office in the wake of the scandal

  • Joël Mesot, ETH president (since 2019): serves as Guzzella’s successor and has to quickly reach a decision in the Carollo case

By the end of this story, ETH’s reputation will be severely tarnished. Some of its doctoral students will have lost their enthusiasm for academia, and one of its professors will be out of a job. And the president – shocked by the revelations made by Republik – will call a press conference at short notice on Thursday, March 14, 2019, and apologize to all the doctoral students who have been treated without respect.

The venerable ETH, which ranks among the best research institutions in the world, will be shaken to its very foundations.

Act One: A Doctoral Student Rebels

This story begins with two women, both of whom are Italians. In addition to their origins, they share a great passion for astronomy. Exactly one year later, one of the two, Marcella Carollo, will rise to dubious notoriety as the ETH professor who allegedly harasses her students. The other woman, who we’ll call Elisabetta Marignano, will reap praise as the doctoral student whose courage was instrumental in bringing to light a case of professional misconduct and abuse of power.

During the summer of 2016, everything seemed perfectly fine. Marignano was a doctoral candidate at the Institute for Astronomy and apparently had a cordial relationship with her adviser, Marcella Carollo. Emails between the two often ended on a friendly note with abbraccio, the Italian word for hug. On Marignano's birthday, the professor organized a cake and sang along with the other researchers at the institute.

In order to complete her doctorate within the prescribed four-year time limit, Marignano was tasked with evaluating digital images of 40 million galaxies as part of an international collaboration called the Dark Energy Survey. But she only had a vague grasp of the programs and lacked the astrophysical knowledge to develop her own analysis software, at least according to her adviser, Carollo.

In Carollo’s eyes, Marignano had been out of her depth ever since her hiring one-and-a-half years earlier. In a bid to help her, Carollo sat down with the student and analyzed her strengths and weaknesses and recommended what books to read and which lectures to attend.

Marignano saw things differently and felt that Carollo should have been more supportive: «She was very aggressive and treated me very badly saying that my brain doesn’t have any logic (...) and that I should ‹go and do philosophy instead of physics›», as she recounted in her testimonial, in which she would later reveal to the relevant ETH authorities what she viewed as the scandalous situation at the institute.

Speaking through her lawyer, Marignano declined to be interviewed by Republik, and she did not respond to a list of questions submitted by email.

In October 2016, the conflict between the doctoral student and her adviser came to a head. Carollo no longer believed that Marignano would complete her doctorate. The professor said that she was progressing too slowly to finish within four years, and perhaps six years might not even be enough, which is the maximum amount of time that university regulations allow students to pursue a doctorate.

Carollo was distressed over the situation. Although Marignano had not transferred to ETH from a top-ranking university, she had received excellent grades while studying for her master’s degree. Carollo saw in her a young woman who deserved a chance – and the student reminded her of herself in her younger days: After studying biophysics, Carollo’s financial situation had forced her to work for a number of years as a high school teacher, and then as a software developer. It was only then that she was able to pursue an academic career – and only because a professor of astronomy, whom she had met by chance, offered her an opportunity to pursue a doctorate.

The Situation Escalates
Carollo decided to use an employee performance review to confront her doctoral student about her lack of progress and, working together, establish new objectives and deadlines: a last chance for Marignano, but also a formal step that would allow her to terminate the young woman’s employment if she failed to produce results. In doing so, she was acting upon the recommendation of Rainer Wallny, the deputy head of the physics department and former director of studies.

In 2013, one of Carollo’s doctoral students had exceeded the maximum study period, requiring six-and-a-half years instead of four years, which was only possible thanks to a special exception granted by the rector. According to Carollo, Wallny wanted to avoid a similar situation, so he coolly instructed Carollo to decide at an earlier date whether a candidate was capable of completing a doctorate within the prescribed period of time.

The two women’s assessments of each other during the performance review couldn’t have been more widely divergent. The doctoral student praised the professor across the board, saying that Carollo did a great job of looking after her students, and that she motivated them and was always willing to find solutions to problems. The professor, though, stuck to her guns and said that the doctoral student’s performance was inadequate. Marignano was totally crushed.

The student did her best to provide the results for an initial publication, but Carollo remained unconvinced by them. Shortly before Christmas 2016, the professor informed her that she would not renew her contract beyond February. The situation escalated. The furious doctoral student accused her adviser of unprofessional supervisory conduct.

Up until this point in time, the conflict between Marignano and Carollo had been like many conflicts between doctoral candidates and professors. Not all doctoral students successfully complete their work and, according to the latest statistics, 86 out of 100 successfully complete the doctorate program at ETH.

In fact, ETH has established no quantitative criteria, such as a specific number of publications or research results, that would qualify a candidate for a doctoral degree – nor does it rely on committees or even a second adviser for doctoral students, as is prescribed for instance at the Ecole polytechnique fédérale (EPFL) in Lausanne. The disparity in power between professors and students is huge, and advisers serve as both the judges and executioners of their doctoral candidates.

This state of affairs is dangerous for doctoral students and professors alike. Without clearly defined criteria, when irreconcilable differences arise it remains impossible to unequivocally decide whether the students lack the necessary academic skills and diligence – or whether professors have placed unrealistic demands on their candidates and failed to advise and supervise them properly.

But what differentiates the conflict between Marignano and Carollo from other academic clashes is that this candidate refused to accept failure and decided instead to take a stand against her adviser.

On Jan. 9, 2017, Marignano entered the office of Wilfred van Gunsteren, the ETH ombudsman. Together with his colleague Maryvonne Landolt, van Gunsteren serves as a point of contact for internal conflicts that cannot be resolved with direct dialogue, as well as suspicions of improper conduct. This is precisely the type of conflict that Marignano reported.

Van Gunsteren found her accounts shocking, so he asked the doctoral student to put her allegations in writing. In her testimonial, Marignano writes: «I realized I am victim of psychological abuse by my own supervisor. The amount of offenses I try to overcome everyday is suffocating my work and my own life, to the point that I decided to ask for help and search for another supervisor.»

The list of allegations that Marignano brought to bear against her professor covers six pages. But this time around, there was no mention of the professor’s abilities to motivate or her willingness to take a keen interest in her doctoral students, which had only recently been much-lauded by Marignano. Instead, the student described in anecdote-rich detail how Carollo had allegedly harassed her.

Once, she said, the professor admonished her to waste less time on make-up and nail polish and, instead of wearing fine clothing, to don a jogging suit, like her officemate.

On another occasion, she wrote that Carollo refused to bring forward a meeting because Marignano was suffering from menstrual cramps – and had waived aside her request by quipping that men didn’t have such problems, so women couldn’t afford to have them either.

The ombudsman was appalled. He backed the doctoral student’s plan to look for a new adviser and offered his support.

The first act had been launched.

Over the next four weeks, the ombudsman received nine additional reports on Carollo, a total of 24 pages that paint a consistently disastrous image of the professor.

The testimonials are confidential, and Carollo has still not been able to see them. Republik has acquired copies of 10 of the 12 testimonials. The authors are former and current staff members of the Institute for Astronomy, and some of the accounts go back a number of years. The ombudsman later wrote to the school administration that «under no circumstances should the reports be shown to MC (Marcella Carollo), as the people who wrote them would otherwise have to contend with professional repercussions and experience difficulties. The world of astronomers is not large.»

Ever since her appointment to ETH in 2002, Carollo had maintained a spotless personnel file, without a single entry. But now, within a very short period of time, nine complaints had landed on the ombudsman’s desk. This was no coincidence. Republik’s investigation shows that both the doctoral student and the ombudsman contacted former doctoral candidates and postdocs and asked them to write testimonials. When confronted about this, neither van Gunsteren nor Marignano were willing to comment.

In their search for damaging material, the two received support from the vice rector for doctoral studies, Antonio Togni, who met with the doctoral student on Jan. 10, 2017, one day after she first contacted the ombudsman.

With his round glasses and gray beard, Togni, 60 at the time, is a fatherly figure. In addition to his core activity as a professor of chemistry, for the past nine months he’s been responsible for the doctorate program at ETH, from admissions all the way to the granting of degrees. He has a reputation as a good listener, as someone who takes students seriously.

There are no written records of his conversation with Marignano, and all we know is that he was shocked by what she told him. He reached the same conclusion as the ombudsman, van Gunsteren – namely that the doctoral student should apply for a change of adviser – and indicated that he would support such a move in his role as vice rector.

ETH Has No Memory
ETH actually has a sophisticated, multi-tiered conflict resolution process that is anchored in the regulations for doctoral studies. If significant differences of opinion between professors and doctoral students can’t be resolved, the department head is to mediate the conflict. If need be, the vice rector assumes the role of mediator. If that fails to produce results, the vice rector appoints an arbitration committee, whose membership is clearly defined. If this committee fails, the next step is decided by the rector, who may opt to assign a new adviser.

If this process in the conflict between Marignano and Professor Carollo had been followed to the letter, the situation might not have escalated the way it has. But ETH has been flouting its own rules.

Vice Rector Togni has declined to comment on the situation to Republik, stating that as long as the process of dismissing Carollo has not been completed, he views it as too sensitive a matter to discuss with representatives of the media.

Instead of first trying to mediate the conflict, Ombudsman van Gunsteren and Vice Rector Togni advised the doctoral student to switch advisers. Neither of them thought it was necessary to investigate the accuracy of the descriptions provided by the doctoral student.

Although nothing official had been brought against her, Marcella Carollo had a reputation for being «difficult person» – an ambitious and dogmatic control freak. According to the rumor mill, she had only been appointed to her position at ETH thanks to her husband, Simon Lilly, an outstanding British astronomer. When ETH negotiated a contract with Lilly in 2002, he reportedly demanded that his wife also be given a professorship, and the university agreed to sign them both on as part of its dual-career program. Carollo and her husband built up the institute together, but some gossip holds that he was actually the brains behind it.

The testimonials that the ombudsman received appeared to confirm the rumors that Carollo overcame her presumed inferiority complex by taking out her frustration on her subordinates. Van Gunsteren concluded that Carollo had been systematically harassing her doctoral students for 15 years.

At this point in the chronology of events, two additional serious flaws appear in the ETH system. First, at the end of an ombudsman’s tenure, they don’t pass cases on to their successor – there is no managerial administrative system. In other words, ETH has no memory. Second, since the role of ombudsman is always assumed by professors emeritus of ETH, they are all biased. They are familiar with all the rumors and stories that have been circulating and have either a positive or a negative relationship with the professors.

«Do We Have a ‹Battle Plan›?»
While a storm was brewing around Marcella Carollo, her husband, Simon Lilly, 58 at the time, was celebrating a success. The physics professors at ETH had encouraged him to seek a second term as department head. This lanky man with glasses and a touch of red in his hair is a quiet, hard-working man who doesn’t make a lot of waves. His unpretentious manner made him popular among his colleagues, and one-and-a-half years earlier they had elected him to be their boss for two years. Now, he had been asked to consider adding another two years. At this point in time, he had no way of knowing that the conflict between his wife and her doctoral student was also about to turn his career on its head. In order to avoid conflicts of interest, Lilly is not responsible for his wife. Instead, this role falls to his deputy, Rainer Wallny.

In early February 2017, Carollo got summoned by Vice Rector Togni to his office to explain why she was no longer willing to work with her doctoral candidate. The professor handed the administrator a dossier with a comprehensive explanation for why, in her view, there had been too little scientific progress to merit continuing in the pursuit of a doctorate. The meeting lasted three hours. There is no record of what was said, only a file entry by Togni.

During their discussion, Carollo said she suspected Marignano had changed the password, preventing all access to her research data in an effort to block her dismissal. In fact, the data belongs to ETH. Togni promised to look into the matter.

But instead of mediating between the professor and the doctoral student, as prescribed by the regulations for doctoral studies, Vice Rector Togni met two days later with two doctoral students who, until then, had not complained about Carollo. Once again, there is no record of what was said. A few weeks later, though, ETH’s administration informed Carollo that the two doctoral students had been removed from her tutelage because they reportedly wanted to change advisers.

The ETH receives 1.3 billion Swiss francs of taxpayers’ money every year: main entrance.

After Togni had met with the two doctoral students, he organized another meeting with Professor Carollo and wrote an email to Rainer Wallny, the deputy head of the physics department. The situation in Carollo’s group had become intolerable, Togni wrote, «because it is unworthy of ETH», adding: «I’m therefore prepared to pursue the objective of allowing no more doctoral candidates to be assigned to MC.» He closed his email with the question: «Do we have a ‹battle plan›?»

Wallny responded that he would be happy to attend an upcoming meeting to elaborate a «battle plan», although he expressed little hope that Carollo could be persuaded to change her attitude. He also sent a copy of this email to ETH Rector Sarah Springman.

In view of the unmistakable tone of the exchange between Togni and Wallny, the rector must have assumed that there was a serious problem, and yet the facts of the matter still had not been clarified. Nobody knew what, if anything, the professor had actually done wrong, or when it had allegedly happened.

On Feb. 9, 2017, Vice Rector Togni, department deputy head Wallny and Professor Carollo held a meeting. The two men read excerpts from the ETH Compliance Guide and, based on the testimonials, reproached her for violating university regulations, alleging that she had not sent her doctoral students to conferences, had not provided them with helpful supervision, had not clearly communicated with them and had forced some of them to sign compromising documents. Carollo was not permitted to view the testimonials and was forced to comment on the allegations without any knowledge of their context. The professor felt like she had been convicted without a hearing.

In an email that Wallny sent to Rector Springman the next day, he portrayed Carollo as intransigent. Although he admitted that «none of the truly successful doctoral candidates» had written a testimonial, and that the complaints had moreover come «from the (female) postdocs whose careers were over after Marcella», he nevertheless was convinced that Carollo was a person who subjected herself and her staff to «extreme quality criteria». In his eyes, she was a «tortured soul who tortures other souls with the intention of wanting the best for them».

After the meeting with Togni and Wallny, Carollo went to speak with the ombudsman. She shared with him her concerns that officials at ETH intended to proceed against her without investigating the accuracy of the allegations that had been lodged against her. She added that they were concealing the concrete allegations against her, thus depriving her of the possibility of defending herself.

Although he had been working on the case for a month, the ombudsman acted as if this was all news to him. To make matters worse, only a few hours earlier he had spoken with the top school administrators about Carollo in a conversation that exposed a difference of opinion with ETH President Lino Guzzella – one that would soon serve as a catalyst for the affair.

Carollo handed the ombudsman a document with a written complaint against Togni and Wallny. She assumed that he would treat it confidentially – a fatal mistake, as it turned out. Van Gunsteren, the ombudsman, was himself a professor emeritus of chemistry, and he knew Togni from his time in the department. Toward the end of the meeting, he said that he would speak with Togni about the matter, but Carollo recalls asking him to refrain from doing so. Carollo says that van Gunsteren responded that he wouldn’t allow anyone to tell him who he could and couldn’t speak with, adding that he didn’t take orders from anyone.

Act Two: An Ombudsman Tastes Blood

Wilfred van Gunsteren, born in 1947 in Wassenaar, The Netherlands, is a man who takes the office of ombudsman very seriously. Every year, he writes a detailed report on the work of the office of the ombudsman, a role which he shares with his colleague Maryvonne Landolt. He has the feeling that ETH President Guzzella doesn’t read it. It angers him that the school administration does not recognize that the office of the ombudsman contributes to a better ETH by listening, acting as a liaison and mediating. Van Gunsteren has consistently swallowed his irritation.

But on the morning of February 14, 2017 – the day he later met with Carollo – he decided to change course.

Together with Landolt, van Gunsteren presented the 2016 annual report to ETH President Guzzella and the other top-ranking members of the school administration. Once again, it seemed Guzzella didn’t really care about what they had to say. Van Gunsteren later complained about the president in a letter addressed to another member of the school administration: «Maryvonne and I found the reception (...) to be not particularly friendly. The president wanted to spend as little time as possible and not really listen to us, made no allowances to discuss in detail the problems that we had mentioned, and treated us like subordinates who could object to nothing that the boss says.»

In his presentation, van Gunsteren wanted to warn the school administration of the Carollo case, which had brought to light an important question to which ETH had no response: «What is to be done if a professor appears to be grossly lacking in management and supervision skills?»

This question exposes a fundamental problem in the supervision of doctoral candidates: There are few rules or quality standards. What is expected of a doctoral adviser? What criteria must be fulfilled and how is it possible to evaluate whether the supervision is good or bad? What if a doctoral student claims to have failed because they have been insufficiently supervised? How is this reviewed? At what point do we speak of a lack of supervising skill? ETH has no way of dealing with such questions.

Growth at ETH, from 2006 to 2017

In percent

Students0+58 % Doctoral students0+47 % Professorships0+31 %

Source: ETH Zürich; number of students and doctoral students, full-time professorships.

This problem has been exacerbated by the rapid growth at ETH in recent years, especially with respect to the student body and non-professorial teaching staff. The number of students rose from 2006 to 2017 by 58 percent from 13’007 to 20’607, and the number of doctoral candidates increased by 47 percent from 2792 to 4092. Meanwhile, the number of full-time positions for tenured professors rose during the same period of time from 312 to 408 (an increase of 31 percent).

This uneven growth has resulted in a progressively poorer ratio of advisers to candidates: In 2008, a professor at ETH in Zürich or the Ecole polytechnique fédérale in Lausanne still served as an adviser to an average of 34 Students and doctoral candidates, whereas in 2018 that number had risen to 39.8, according to ETH. Other elite universities that have better international rankings than ETH are proud of their adviser-to-doctoral-candidate ratio of roughly 20 students per professor.

A Document «That Would Stand Up in Court»
Despite an urgent need for action, on February 14, Guzzella refused to deal with the questions raised by van Gunsteren. The ETH president demanded that the ombudsman revise his presentation before revealing it to other high-ranking members of the school administration that same day – and he specifically admonished van Gunsteren not to cite the possibility of misconduct by professors in their interactions with doctoral students. For instance, Guzzella wanted the issue of «shouting at or willfully insulting staff members» to be removed from the presentation as an example of professorial misconduct. Guzzella has declined a request by Republik to comment on this incident or on his role in the Carollo case.

The meeting that morning once again revealed Guzzella’s low opinion of the office of the ombudsman: He saw it as a necessary evil that caused more problems than it solved. In his eyes, scientific research is not a nine-to-five job, especially not in the upper echelons of academia, where ETH is a major player.

Indeed, if ETH intends to keep pace with American, British and Chinese universities, it will have to redouble its efforts. Just recently, Yale University outclassed ETH, forcing it off Times Higher Education magazine’s annually published list of the world’s top 10 universities. Scientific publications and Nobel prizes are the decisive criteria for a ranking, not the work-life-balance of the staff.

Guzzella smelled trouble if they launched an in-depth debate on the working conditions of doctoral students, despite the fact that these students – along with the ombudsman – wanted to address these issues. According to a study commissioned in 2017 by the association of non-professorial teaching staff, Aveth, the average workload of an ETH doctoral candidate is 50.5 hours a week. Another Aveth survey from 2018 revealed that 24 percent of respondents had experienced incidents of abuse of power, and specifically complained of countless overtime hours, high pressure and poor supervision.

On Feb. 14, 2017, van Gunsteren gave in to Guzzella’s pressure and adapted his presentation to accommodate the president’s requests. Van Gunsteren had decided to grin and bear it – but only one last time.

From then on, he would complain at every available opportunity that, in his words, Guzzella wanted to «sweep things under the carpet». In a long letter to a member of the school administration, he poured out his soul and wrote that he was dumbfounded by Guzzella’s ignorance: «We didn’t feel personally insulted by the president last Tuesday, merely treated somewhat disrespectfully. But the office of the ombudsman has been insulted», he wrote. In the same letter, van Gunsteren repeated his earlier warning: The Carollo case posed a serious threat to ETH’s reputation because the university had no standardized procedures in the event that professors failed to conduct themselves properly. He went on to say that the case was likely to lead to a legal battle, hence the need to come up with a document that would «stand up in court».

This demand is symptomatic of the stage that the Carollo case had reached at this point. After ignoring every single rule in the book at the outset of the conflict between Marignano and Carollo, now they were concerned about the legality of their own actions.

On March 1, 2017, Carollo was summoned to a meeting with ETH Vice President Ulrich Weidmann, who had taken over the case for the school administration now that Rector Sarah Springman was leaving her position. Before the scandal erupted, Springman had been a friend of Carollo’s and a weekly dinner guest at her house. But now the outgoing rector didn’t want to have anything to do with an alleged academic bully, neither privately nor professionally. Carollo was told to come to the meeting with Weidmann accompanied by a «trusted individual», so she brought along a lawyer in the hope that she would finally have an opportunity to respond to concrete allegations. She was confident that she could refute many of the accusations and clear up any remaining misunderstandings.

But Weidmann didn’t want to hear any explanations. He immediately launched into the topic of the conditions that Carollo would have to agree to if she were ever to hire doctoral candidates again. First, she would have to undergo coaching. Second, in the future each of her doctoral students would be assigned a second adviser.

But that was not the end of the meeting. Carollo was informed that her two remaining doctoral candidates had meanwhile said that they also wanted to leave her. Weidmann shared no reasons with Carollo, and it would not be until three months later that these two doctoral students provided their testimonials. They, too, ended up sharply criticizing Carollo. «In years under Prof. Carollo, I was repeatedly offended as a woman in front of my male colleagues», as it says in one of the documents. Adding: «I was told that if I had time to put on make-up then I should have produced more plots instead.»

As with Marignano, ETH disregarded its own rules once again and there were no attempts at mediation. Is it possible that ETH recommended to these students that they also leave their adviser? Neither the administrator in charge at the time, Vice President Ulrich Weidmann, nor Vice Rector Antonio Togni were willing to comment on this to Republik. But it is clear that the university’s habit of ignoring its own rules had become increasingly systematic with Carollo’s case.

A Chain Reaction
For the time being, Ombudsman van Gunsteren remained in the dark about the agreed coaching and co-supervision measures. Three weeks earlier, Weidmann had asked the ombudsman to provide an overview of the case. Van Gunsteren did so in writing and the ETH vice president responded: «The situation is so clear that it requires immediate action.»

But in response to van Gunsteren’s request to keep him up to date on the matter, Weidmann merely responded that the process would be discussed among top-ranking school officials. Afterward, the ombudsman heard nothing more about the case.

The ombudsman decided to take matters into his own hands. He was unwilling to see the matter put to rest without visible consequences. His determination to see things through certainly appeared admirable, but it was actually cause for grave concern.

According to the basic ethical principles of the International Ombudsman Association, an ombudsman should always remain neutral and independent, and handle all matters confidentially and informally. They should take an impartial approach to contentious issues – and under no circumstances are they to conduct their own investigations.

And yet van Gunsteren, a man who has sworn to uphold ETH’s ethical standards, would end up violating every basic principle in the Carollo case. He essentially morphed into the legal counsel for everyone interested in lodging a complaint against Carollo and began by helping Marignano collect testimonials from other disgruntled doctoral students. He even went one step further and summarized the charges from the testimonials under 14 points and sent the list to Carollo on March 6, 2017, with a request for her to comment on them. Afterward, he informed Weidmann of his approach.

Weidmann responded tersely: «I’m extraordinarily unhappy with your uncoordinated actions. We have initiated diverse processes, some of which have already been implemented. What you have apparently undertaken is of little help here.» Weidmann sent a copy of his email to every top-ranking official at ETH. The message to van Gunsteren was patently clear: Leave it to us – the school administration will take care of it.

But van Gunsteren had no intention of withdrawing and, obviously unimpressed, wrote back to Weidmann: «I don’t see any obstruction of the school administration’s work on my part. Instead, I’m helping to identify possible grievances at ETH. So, I’m merely fulfilling my duty as an ombudsman.» He then referred to Article 15 of the organization’s code of conduct, which states that an ombudsman is not subject to instructions and can demand information and the presentation of files from any department at ETH. This passage would come back to haunt top-ranking school officials. Indeed, van Gunsteren had tasted blood.

The second act had been launched.

This was a decisive moment. The school administration and office of the ombudsman were no longer working together. From that moment on, they would both insist that they alone had jurisdiction. This development in the Carollo case sparked a chain reaction that would soon lead to a state of chaos that far overshadowed the original conflict between a professor and her fired doctoral student. Yet despite their differences, they all seemed to agree Carollo must be guilty – and that her guilt didn’t need to be investigated.

Carollo could only guess which former and current doctoral students and postdocs were behind the 14 allegations that she was expected to respond to. She feared she would get tangled up in contradictions as long as she was unaware of the exact wording of the allegations, so she decided to wait. Since Vice President Weidmann had scheduled another meeting for March 24, she thought that perhaps now she would finally have an opportunity to clarify her perspective on things.

As she took a seat across from him, Carollo brought up the summary compiled by the ombudsman, van Gunsteren, and pointed out that it contained several untenable allegations. Under point 12, there was even a quote that suggested scientific misconduct, the ultimate deadly sin among scientists. Someone had maintained that she had manipulated a chart to make the data look better. Carollo said she was certain that she had never done anything of the kind. She went on to explain that she could easily refute the allegation because she saved all of her data for precisely this reason – it was her insurance that no one would ever be able to accuse her of ever producing a scientific fabrication.

Carollo demanded that this allegation be investigated. She thought it could be a starting point to confirm her suspicion that a smear campaign had been launched against her, presumably initiated by the fired doctoral student, Elisabetta Marignano. But Weidmann dismissed the idea out of hand and assured the professor that the ombudsman had been acting alone and that his allegations, at least those related to scientific misconduct, were without relevance. And that was the end of the issue.

When Carollo realized that the school administration was also not interested in investigating the allegations, she responded to Ombudsman van Gunsteren’s 14-point summary. Her response is 10 pages long and includes a list of all her doctoral students, many of whom have gone on to pursue highly successful careers after leaving ETH.

On April 13, 2017, she sent her statement to the ombudsman – and heard nothing more from him, which is not to say that nothing happened. It was not until half a year later that Carollo found out what the ombudsman had done with her statement.

But by then the whole institution had been engulfed in scandal.

Counterstatement by the ETH

  • The Republik article states: «An astronomy professor at Zürich's ETH university has been accused of harassing doctoral students for years. She's about to be fired, but the question of her purported guilt has never been clarified.» This is not true.

The truth is: The independent external investigator has conducted a thorough investigation to clarify the allegations made against the professor. In view of the professor's misconduct, ETH Zurich has initiated dismissal proceedings and has now submitted a formal request for dismissal to the ETH Board.

  • The Republik article states: «Nobody knows for sure because no one has mediated the conflict or investigated the allegations: neither the head of department, the office of the vice rector, the college administration nor the external administrative investigation. Professor Carollo's guilt appeared to be a foregone conclusion right from the start.» This is not true.

The truth is: The Vice Rector of Doctoral Studies, the Department Management, the Ombudsperson and the ETH Executive Board responded to the allegations by holding a number of discussions both with Professor Carollo and with those making the allegations. The independent external investigator then conducted an inquiry into the allegations made against Professor Carollo. The investigatory procedure was unbiased in reaching its conclusions.

  • The Republik article states: «Now, she is to be dismissed – without official prior warning – although there are no provisions for such action in the university's personnel statutes.» This is not true.

The truth is: In the spring/summer of 2017 the professor was given several verbal warnings and one written warning. Dismissal even without an official warning is possible under certain circumstances.

  • The Republik article states: «And the president – shocked by the revelations made by Republik – will call a press conference at short notice on Thursday, March 14, 2019, and apologize to all the doctoral students who have been treated without respect.» This is not true.

The truth is: The decision to arrange a press conference was made before the Republik's questions have been received.

  • The Republik article states: «In their search for damaging material, the two [Ombudsman and doctoral student] received support from the vice rector for doctoral studies, Antonio Togni, [...].» This is not true.

The truth is: Both the Ombudsperson and Vice Rector Togni reviewed the allegations in accordance with their areas of authority. They did not seek out material with a negative bias.

  • The Republik article states: «As with Marignano, ETH disregarded its own rules once again and there were no attempts at mediation. [...]» This is not true.

The truth is: ETH has observed its own rules. The entities involved have adhered to the applicable laws, ordinances and regulations.

ETH Zurich

Republik stands by its reporting.

The ETH Case: Coming Up in the Next Installment

Former ETH President Guzzella is determined to put an end to the Carollo affair as quickly as possible, but he sparks an internal power struggle at the university that soon reaches a new level of escalation: The allegations of harassment are leaked to the public. And on the international academic stage, Marcella Carollo will be largely portrayed as the sole guilty party. The allegations of harassment quickly become a smear campaign that threatens to put an end to her career.

About the Authors

Silvan Aeschlimann is an author and journalist who lives in Zürich and Barcelona. His novels «Glück ist teuer» (2017) and «Ungehört» (2013) deal with issues like the pressure to perform, economic growth and materialism.

Dominik Osswald studied geology, is an enthusiastic mountaineer and a freelance journalist. He reports for magazines and TV broadcasters on efforts to reform Switzerland’s pension system and on the #MeToo debate, extreme mountaineering expeditions and climate change. He has worked for the Basler Zeitung, Tages-Anzeiger and the TV news programs «10 vor 10» and «Rundschau» for Swiss public broadcaster SRF.

Dennis Bühler is editor at Republik.

The ETH Case

Sie lesen: Part 1

Systematic Failure

Part 3

Rigged Game