Sunday 3 January 2016

Harassment and that powerful male PI

After decades of sexual harassment astronomer Geoffrey Marcy has been fired. What surprised me reading many comments was how invincible he was thought to be. I am not nearly as powerful, but can maybe still offer a view from the other side. My impression is naturally subjective and I work in a different field and in another country, but I wonder if Geoffrey Marcy was really that invincible. Unfortunately, people having that impression is partially a self-fulfilling prophesy. Making this picture more realistic may thus help reduce harassment, which is why I wanted to write about this.

Your scientific network is crucial

This impression of invincibility may partially stem from the lone genius syndrome in the media, which is not even right for Einstein. To make a story more pleasant to read journalists like to personalize everything and simplify their story enormously, which leaves little space for history and multiple contributors.

More realistic is that it is very hard to do science in isolation. Feedback is very important. If you only get feedback after publication, you will progress very slowly. Collaboration is also important to get new skills into a study. Understanding new methods and datasets takes a long time. Without collaboration you would have to invest a lot of time and still have a high risk of making rookie mistakes. Without collaboration your productivity and quality will be much lower.

Also the peer review of research proposal and scientific articles makes it important to keep friendly relationships with most colleagues. A research proposal always has weaknesses; if everything was clear already it would not be a science project. Even if not, a reviewer can always claim a proposal is too ambitious or not ambitious enough. Judging a proposal to be “very good” rather than “excellent” is always possible and reduces the funding probability a lot given the low percentage that is funded.

Negotiating conflicts

Due to the importance of collaboration and good relationships open conflicts are rare. There are open debates about specific scientific questions; questions that have an answer and enhance your scientific reputation if you are right. Vaguer questions that do not have an answer are not worth an open conflict that may damage relationships.

That normally most scientists are on talking terms with each other may give outsiders and young scientists the feeling that the seniors are one solid block. This is certainly not the case. Inside you can see the cracks. Most conflicts between groups are for long-forgotten historical reasons. Most conflicts between people because their personalities do not match.

(Hardly ever do you see conflicts for political reasons. I just send an email to Iran and got a friendly answer. My colleague from Serbia still talks to me after we bombed his country based on flimsy evidence. As far as I can judge the climate “debate” is nearly never a source of conflict; the quality of someone work naturally does influence collaboration decisions.)

Conflicts in science are not resolved by a strong man and presented at a press conference, rather it is a continual collective negotiation. That makes the conflicts and the influence of evidence of harassment less visible.

Two examples: If I had a preliminary result I was not sure about yet and I presented it at a conference to get feedback and thus had added a no-tweet sign to my talk. In this hypothetical case, our friend of the freedom to tweet, Gavin Schmidt, could naturally tweet about it anyway. I would then see that as a breach of trust and would only collaborate with Gavin if really no other scientist had the competencies I would need for a specific project. Gavin likely could not care less about a dwarf from an adjacent field. More important is how the scientific community sees the situation. If they see my request as reasonable, the tweet would hurt Gavin’s reputation. For something as small as this he will not lose his post as director of GISS, but people might be a little more reluctant to collaborate and less willing to share preliminary results with him lest he may tweet it. His closest friends may communicate that to Gavin in private; most will just draw their conclusions and not say anything. That makes those close friends so important, because it is better to know.

That Gavin has tweeted that he does not agree with a conference presentation of Peter Wadhams may cool their relationship. However, there was no breach of trust because Wadhams had said similar things in public before. Thus Wadhams’ unreasonable complaints about Gavin’s tweet probably only lead to further reputation losses for Wadhams.

One of the famous quotes from the stolen emails of climategate is that Phil Jones writes that he wants to keep two articles out of the IPCC report even if he has to redefine what peer reviewed means. What the mitigation sceptics normally do not say, because that does not fit their narrative, is that these two articles were referenced and discussed in the IPCC report.

Given that there are no further climategate emails on this topic I assume that Phil himself thought better of it later. In this case there is no loss of reputation; scientists are also humans, he corrected his initial error and no damage was done. If not his colleagues made clear what they thought of this and Phil decided that this was not worth a loss of reputation. If the article in question was really bad, the loss of reputation would have been limited; if the article was half-way decent the loss of reputation would have been large.

Summarizing, while visually maybe nothing seems to happen and it is difficult to fire faculty members, spreading evidence of harassment does result in reputation loss, reductions in collaboration and productivity and the ability to do the research you would like to do.

Young and female scientists

Young scientists and female scientists are also more important than they may realise. When you are young, doing science is centred around writing articles and somewhat later also on getting funding for your own (small) project. The next step in a scientific career is leading larger projects. At least in Germany and the EU, it is impossible to get funding for such projects without involving female scientists. If you lose support of (nearly) all female colleagues, your career stops right there.

The final step in a scientific career is getting your people into permanent positions. This stabilises your power and somehow professors tend to think that their topic is so important that it should grow at the expense of others. For this, and for the successful completion of scientific projects, you need to be a good talent scout and find young people you can build up.

When I studied physics in Groningen, rumour had it that there were so many professors that they were fighting to be allowed to give lectures and thus get access to the students. That way you can interest the students for your topic and assess who are the good ones. Also companies often pay for professorships to be able to search for talent and lure the good students to work for them.

Binding good students to your group with interesting bachelor and master projects and with student help positions gives professors the possibility to interest the good ones to pursue a PhD in their group. The best of those will hopefully become important scientists and grow the field.

Gossip and punishing bad behaviour

When I was a PhD student, two American scientists misbehaved, I made sure that everyone knew, especially all Americans I knew. When they did it to you, they likely did it to others. Multiple similar bad stories can do real damage to a scientists reputation and their work.

Even if your case is not strong or big enough to go to the police, talk to other female scientists. There are bound to be more victims, which makes the case stronger and makes it more important that the harassment stops. Talk to other male scientists, you may be surprised how many will support you. I know of one multiple harassment case. Too small to be a legal case, but once the leaders of that community heard about it, they talked to the guy, made it clear that any future complaint would have severe consequences and I did not hear about new complaints.

Fighting back is necessary; make sure people know what happened. If bad behaviour goes unpunished it will proliferate. Economic games have shown that collaborative behaviour will only prevail if cooperative people are willing to punish bad behaviour. It also showed that people from all over the world are willing to punish bad behaviour at a cost and derive pleasure from that. Gossip and punishing bad behaviour are the two main pillars of our human civilization.


The media likes to portray scientists as solitary geniuses. In reality scientists need to be good collaborators to become heroes in their field. Many young scientists think that senior scientists are one big pack and may feel that it would not help to talk to senior scientists about the poor behaviour of another. But this is not true either: there are conflicts, they are just not very visible.

It would be good if (female) scientists would speak about (sexual) harassment at work and this likely does more damage than they think and see. I know actually doing this is difficult and there are unfortunately likely repercussions if you do so openly. In addition much should change in the organization of science and in society to reduce harassment. I just want to say: you have more power than you may think.

Related reading

How to end sexual harassment in astronomy on changes in the organization that are needed.

Nature: Scientific groups revisit sexual-harassment policies. Interesting: The American Geophysical Union did not receive any harassment complaints since it started tracking them 5 years ago. It could be underreporting, I hope it means that the situation is not as bad in the geosciences.

Nature: Science and sexism: In the eye of the Twitterstorm. Social media is shaking up how scientists talk about gender issues.

Astropixy: How not to "deal with it". A example of harassment by a non-famous astronomer, makes clear that unfortunately getting rid of a faculty member is hard.

Nature Blog. The faculty series: Learning to collaborate. "Collaborations are the key to success in modern scientific research, says Michelle Ma."

The ultimatum game, a key experiment showing intrinsic fairness and altruism among strangers

*PI: Principle Investigator. The leader of a scientific project and often a group.


  1. Excellent post, Victor, on a wide range of topics. I'm surprised at how few posts, columns or articles about the "nuts and bolts" of day to day research seem to have been written up by working scientists in academic or other research organization settings.

    I won't claim this shortage of discussion is behind the cartoonish view many denialists have of scientists and their work, since the deniers have shown they are capable of ignoring any amount of information they don't like.

  2. The cartoonish view of scientists in the public makes it easier for the mitigation sceptics to link any wrong doing (often just in their eyes, often spin that the public cannot assess) to a problem with the scientific results.

    If it depended on scientists being holy men, we would not have seen any scientific progress. Starting with the conflict between Newton and Leibniz, where Newton abused his position as chair of the Royal Society to write a report that he was right and Leibniz was wrong. And likely there are earlier cases.

  3. Thanks for an excellent article, Victor. Your article and the cartoon sums up what can happen very well. Speaking up is important, but in my experience you need to pick the person/people you speak up to carefully.

    Organisations have norms, sometimes unhealthy norms of behaviour (in which case it often/usually stems from the attitudes and behaviours at the top, which makes it very difficult to change).

    Sometimes for people at the bottom of the pecking order, the only solution is to leave and move to a healthier environment. Other times, getting to the right person who is in a position to influence or change things is the way to go - that person doesn't need to be at the top of the pecking order, just in a position to influence those who are. Most often, when alerted to what is wrong, people are quite horrified and eager to change or learn what to do, but not always.

    I thought that by now men and women working together would be the norm, but I've read of the experience of some young women in science that shows that's not the case everywhere.

  4. Sou, yes, that was the part I did not write much about. The "further reading" list does provide a lot of ideas.

    I had written "speaking up" (public) once by accident and deleted it again. "Speaking with" (private) is likely better in most cases. Building coalitions. Speaking up hurts both parties a lot and will make the situation unworkable if you do not manage to get the harasser fired, which is hard in case of faculty; that is for severe cases with good evidence where you can go to the police.

    That it is hard to fire faculty is likely good for scientific progress, it makes scientists less afraid to publish controversial results. In case of harassment, it is a problem, especially combined with the lower jobs having nearly no protection. There is an unhealthy difference in power due to the project funding system.

    I was surprised that in one case even just a student body speaking with the problematic person solved the problem. Many scientists are somewhat autistic and may not notice that their behaviour is inappropriate. More often colleagues will have to make clear that continued harassment will have severe consequences.

    Leaving does not have to be large step. I used to measure clouds, now I work with climate data. There is hardly an overlap between these two fields. It is not optimal for your (short-term) productivity, but I feel it is good for creative work to change fields once in a while.


Comments are welcome, but comments without arguments may be deleted. Please try to remain on topic. (See also moderation page.)

I read every comment before publishing it. Spam comments are useless.

This comment box can be stretched for more space.