How a dispute at Harvard led to a grad student’s forced mental exam...

January 22, 2017 2:48 AM

At 1 a.m. on 4 June 2016, Gustavo German, a doctoral student in biomedicine at Harvard University, heard a knock at his door. It was three police officers. A doctoral student at Harvard is forced to take an in-patient psychiatric evaluation. Concern for the student or a reprisal for blowing the whistle on his advisor? "The judge issued an order that has created an extraordinary situation: Rubin must allow German to work in his laboratory, but stay at least 30.5 meters away from him, and have no direct or indirect contact. Rubin must also provide German with all of the lab resources he had before the problems began."
posted by Rufous-headed Towhee heehee (39 comments total)

I'm not a scientist, but a doctoral student trying to order around a post-doc sounds weird to me. A very bad situation either way though.

Honest question: how hard would it be to transfer German to another lab to finish his doctorate?
posted by YAMWAK at 3:56 AM on January 22, 2017

An ugly and not surprising situation- grad school definitely takes its toll on mental heath, and supervisors become supervisors because they are good scientists, not because they are good managers.

To answer Yamwak's questions- transferring labs usually means starting over from scratch, and other lab heads would be unlikely to offer a position to a student who has a reputation for causing problems. So pretty hard in this case.
posted by emd3737 at 4:18 AM on January 22, 2017 [13 favorites]

Honest question: how hard would it be to transfer German to another lab to finish his doctorate?

When my professor moved to a different university my choises were to either go with him, graduate with a masters (After 4 years), or essentially start a new project with someone else if they would take me. I wouldn't want to take on German if he was causing this much ruckus for fear of him being the root cause even if he isn't. I doubt that anyone at Harvard would allow him into their group.
posted by koolkat at 4:19 AM on January 22, 2017 [4 favorites]

I was part a large lab where the whistle was blown on sexual harassment/assault. One student out of more than 15. After it became known only one other student officially backed the whistle blower's claims. The department did not and in fact tried to diminish the student's academic standing and also accused them of being mentally unstable.

She pursued a formal university complaint and her complaint was upheld.

About half of the lab still followed the adviser to his next position at a different university and maintained professional contact.

You really can't know if German is right or not simply based on the rest of lab thinking he is crazy or a department saying they have investigated. Both the lab and they department have incredibly powerful incentives to make a student go away in these scenarios even if he is telling the truth.
posted by srboisvert at 5:41 AM on January 22, 2017 [50 favorites]

I thought exactly what srboisvert thought about the very powerful incentives on either side, which are combined with very driven personalities and people who are brilliant at creating narratives from a series of objective and testable facts.

When there's a set of subjective and untestable theses under debate, then the issue comes up that these people who soar like albatrosses when dealing with work problems at the Angstrom level start walking like albatrosses dealing with work problems at the angsty level.*

Which is to say that the issue is about preemptively monitoring labs so that there's facts on hand before any issues get this big.

Also, if MA's compulsory mental health assessment laws are accurately described there, I feel that they're not as good at their deprivation of liberty safeguards as they should be.

*A million apologies for the purple prose and puns. I really couldn't think of a better way to put it. Also, I'm not implying that scientists are any more flawed than anyone else outside the work environment.
posted by ambrosen at 6:19 AM on January 22, 2017 [5 favorites]

This is a hideous lose-lose situation - either :

a vindictive PI tried to destroy a whistleblower by the equivalent of mental health SWATting

or

a concerned PI tried to stop a student who was in danger of self-harm or violence to others - and who could damage the progress of the lab and careers of dozens of other scientists.

...or a mix of either of these situations. BOTH extremes (PI screwing over student, student wrecking/killing lab/lives/self) have happened infrequently but repeatedly in academia. I honestly don't know who to believe in this case.


One bright spot in this the journalism was done with the fantastic Retraction Watch, a beacon of meaningful self-policing by a reality-based community in this time of post-truth.
posted by lalochezia at 6:40 AM on January 22, 2017 [9 favorites]

Regardless of what happens, German's career is toast.
posted by k8t at 6:46 AM on January 22, 2017 [3 favorites]

the exact nature of the resources, such as mice and research assistants, that German is entitled to receive.

ha.

no but seriously even if every single thing they said about him was true, what a horrible man the other one was.

(I would feel better leaving that ambiguous but Rubin wins horrible.)
posted by queenofbithynia at 7:01 AM on January 22, 2017 [1 favorite]

Lalochezia, the third option seems very feasible. Rubin was notified in May of 2016 about the allegation, but things had already deteriorated to the point of meeting with the ombudsperson the month before.

Unless someone gave Rubin a heads up that a complaint had been filed (a possiblity), the tension wasn't solely generated by the investigation. To be fair to German, I imagine wrestling with being a whistleblower probably was stressful, on top of the normal amounts of stress that comes with being a graduate student. But once Rubin found out about the investigation he escalates, using the concerns of lab members as an excuse. German feeling vulnerable retreats, giving Rubin more ammunition. Rubin may have even convinced himself that he was solely motivated by German's well being.
posted by ghost phoneme at 7:12 AM on January 22, 2017 [1 favorite]

Looking through the article, the first mentions of tension are this: "In January 2016, a postdoctoral researcher reported that German asked her to perform a tedious task, which she believed was disrespectful, leading to a dispute. German was “red in the face and shaking with rage,” she said in her sworn statement." It then talks about how another researcher felt unsafe in the lab and worried that German was sabotaging her equipment. The students go to Rubin in April leading to a meeting with the ombudsperson. Everything else happens after these incidents. Now, unless you're willing to credit the idea that these two women fabricated their sworn statements, or that there were tensions between German and Rubin's prior to January 2016 that were left out of the article entirely, Rubin's position seems to be the more reasonable.
posted by Grimgrin at 7:51 AM on January 22, 2017 [6 favorites]

I think Rubin's position was more reasonable until the forced mental-health check.

I'd love to know how Rubin's conversations with the clinician went. If he was pushing for the check it's one thing; if the clinician took it upon themselves to act for what they believed was the greater good, then Rubin may be the unfortunate victim of someone else's overzealousness.
posted by YAMWAK at 8:06 AM on January 22, 2017 [2 favorites]

Obviously we don't have all the details. But having a doctor call for a mental health check for somebody they have never met, with the possibility of involuntary commitment is completely screwed up. Especially in the middle of the night. That sounds an awful lot like malice or retribution to me. If I was German I would be going nuclear with the legal options at this point. If it affects the lab, maybe a vindictive person shouldn't be running that lab.
posted by bh at 8:54 AM on January 22, 2017 [5 favorites]

Now, unless you're willing to credit the idea that these two women fabricated their sworn statements

Obviously the article isn't providing us with a detailed timeline of who said what when, but the sworn statements are what leave me with the impression that something besides just German was wrong with the lab. Lab members were fearful of German causing them professional and physical harm. They reported their concerns in April, after which there was a meeting with an ombudsperson. That seems like a fairly underwhelming response to me. That doesn't make me question the women's statements. Rather, it makes me suspicious that Rubin was more concerned with his reputation than the health and safety of his lab members. Or he didn't take their concerns seriously until he himself was directly impacted.

It definitely seems like some part of the process needs to change, either at the lab/department level or higher.
posted by ghost phoneme at 10:54 AM on January 22, 2017 [2 favorites]

It strikes me that you can have a grad student who is difficult or a jerk or struggling with mental health issues and a vindictive PI. I have to say, the PI is the one who bears the most responsibility here, because he is the PI. He is the adult with the successful career and the ample funding. (His lab is huge! The most successful researcher I've met has maybe eight full time grad students and a lab manager. Thirty people in a university lab is bananas, even assuming you've got a few staff scientists/lab managers in there.)

It is quite possible that the university itself screwed up and precipitated this by not having a good strategy to handle whatever German was going through to begin with. In any case, seizing someone at 1am for an involuntary psychiatric exam is abusive and terrifying unless they are literally at that moment a danger to themselves or others.

When I think of how difficult it was to get a relative hospitalized when she was a terrible danger to herself and probably others, I could cry - we struggled and struggled to get her help (she got better; it was a meds interaction/stress/aging issue that proved very treatable) and yet there's people who think it's okay to grab someone at 1am. It's like something out of Stalinist Russia.
posted by Frowner at 11:58 AM on January 22, 2017 [14 favorites]

Bad, bad things do happen in academia. At one stage I worked for a PI who was not just a classic "kiss up, kick down" person but pretty much a full blown sociopath. If someone outlived their use to her, she would psychologically destroy them until they left - not just people that couldn't handle the pace and the abusive work environment, but in one case a guy who'd been in a serious car accident and had to take some time off as a result. She knew he couldn't lift his arms above his head so she'd ask him to get things off high shelves etc. When people quit she'd salt the earth by calling people she knew working in the same area and warning them about this student looking for work that wasn't very good. In one case she tracked down a guy who left academia and tried to get him fired from his new company. I basically kept my head down and survived and did one more short postdoc elsewhere and gave up on academia.

An unrelated sad case was a guy in my Comp Science class in college. He was a pretty prickly guy in college and once failed a project because he wouldn't hand over the code in case the university claimed ownership. Long after we graduated someone would look him up every now and again because basically "what's the weirdest guy in our class doing now?". He was a PhD student in Portugal and he attacked a professor with an axe. He'd already been diagnosed with paranoid schizophrenia.
posted by kersplunk at 1:05 PM on January 22, 2017 [7 favorites]

Yeah, this was an interestingly ambiguous article. You can spin a number of plausible scenarios from the facts as given --- clearly the lab seems to have closed ranks against this post doc. Maybe because he was a whistleblower, or maybe because he was an unstable jerk they wanted rid of, and he decided to blow the whistle to get his own back. Maybe the PI was unethical and the post-doc really did have a crisis of conscience months after the paper was submitted; maybe that "crisis of conscience" only happened after his own position got shaky; maybe the paper was above board after all. Tough to say from the facts.
posted by Diablevert at 1:22 PM on January 22, 2017 [1 favorite]

I'm not going to comment on this institution's handling of poorly behaved academics.

Most labs have lab chores - these are divided up between the lab members. So it is entirely possible that a grad student might ask a postdoc to do a menial task. It is also entirely possible that one lab member would ask another to do a favor and that favor could be a menial task. I think that as a grad student I definitely asked postdocs to do things like change a buffer on dialysis so I could go home at a reasonable time (for reasonable time, please understand that we're talking a 12 hour day as opposed to an 18 hour day); and I'd expect to be asked to do the same by them. And I could understand if this guy asked someone to do something like turn on or off a machine and they refused how he'd get really pissed off. Especially if he'd done other similar things for this person.

Working through the night is common in many labs.

I've worked in and near labs where people were known to sabotage others work. I've worked near labs where there were people whose data I didn't trust. I've worked near labs where everyone knew that one of their major papers in a top journal had extensive non-reproducible data.

I am not going to comment on harassment in labs.
posted by sciencegeek at 1:58 PM on January 22, 2017 [8 favorites]

Rubin was notified in May of 2016 about the allegation, but things had already deteriorated to the point of meeting with the ombudsperson the month before. Unless someone gave Rubin a heads up that a complaint had been filed (a possiblity), the tension wasn't solely generated by the investigation.

I'm not sure we can conclude that from the facts given. The alleged misconduct goes back almost a year before that.

"In the fall of 2015, he took a refresher course in the conduct of science, which covered topics such as ethics and misconduct. The course reminded him that, around June 2015, a member of Rubin’s lab had told him about the alleged data fabrication in the Cell Stem Cell paper. After talking about the situation with faculty and students—in hypothetical terms—German concluded that it was “grave misconduct,” he stated in an affidavit, and that he had an obligation to correct the scientific literature."

German's people skills don't sound that good, so his attempt to protect his identity by "talking ... in hypothetical terms" may not have fooled anyone. The judge's statement of facts is very damning for Rubin. (It's no longer on the SocialLaw.com site but I linked to the cached copy).

The ethics committee finally met with Rubin on May 4th, and again on May 10th, with Rubin and Natalia Muela, one of the other researchers German accused. They showed them German's email, with his identiity blanked out. On that same day, May 10, Muela and Tobias Grass (her beefy boyfriend in the lab) sent an email accusing German of "erratic and threatening behavior" and worrying about him sabotaging their research.
posted by msalt at 2:56 PM on January 22, 2017 [12 favorites]

The mental health check stuff is probably driven by the various grad student on grad student and lab tech on grad student murders over the past decade. And the middle of the night thing might have been when the police found him at home. This "resolution" sounds like it's just going to make things worse either way though.
posted by fshgrl at 3:29 PM on January 22, 2017 [2 favorites]

Thanks msalt for posting the cached copy of the judges ruling. It really is damning of Rubin. Prior to that I thought that maybe Ruin was right about German.
posted by biggreenplant at 3:48 PM on January 22, 2017 [3 favorites]

I took a sec to throw a PDF of the cached ruling on my Dropbox if the cached copy ages out.
posted by Samizdata at 4:21 PM on January 22, 2017 [3 favorites]

Comments such as yours are what make viewing things like this through the prism of MetaFilter so worthwhile, msalt.
posted by jamjam at 4:29 PM on January 22, 2017 [4 favorites]

Thanks!
posted by msalt at 4:33 PM on January 22, 2017 [1 favorite]

msalt: "Thanks!"

As well as going to the trouble of finding the cached copy of the results, which added a whole different angle to the article.
posted by Samizdata at 4:40 PM on January 22, 2017 [1 favorite]

Thirty people in a university lab is bananas, even assuming you've got a few staff scientists/lab managers in there

I personally think 30 is way too big and I would be very reluctant to join a lab that large, but it's honestly not that unusual among well-funded R1 big shots in molecular/cell biology. I think most established labs at places like the Broad Institute would be around that size. Someone like George Church might have as many as 50. Mike Snyder has almost 30 postdocs alone.
posted by en forme de poire at 5:30 PM on January 22, 2017 [6 favorites]

I'll also note that if I remember right, the first investigation of Anil Potti "cleared" him of misconduct despite the careful and extremely damning statistical sleuthing by Baggerly and Coombes. The investigation was only reopened once they figured out he'd actually faked his CV by pretending to have won a Rhodes scholarship. At the time, there were clinical trials in progress based on the results of his research.
posted by en forme de poire at 5:36 PM on January 22, 2017 [5 favorites]

Strange, disturbing case. The judge's statement of facts is indeed strongly in German's favor, but her conclusions and reasoning don't seem to reflect a good understanding of what lab research culture is usually like, and from her statements of the facts, I'm not particularly convinced one way or the other regarding Rubin's or German's version of events. What is definitely clear is that Rubin, the department, and the President's office did not handle this particularly well.

sciencegeek may not want to comment on Harvard's handling of academic misconduct, but I will. I have a fairly oblique connection to a high-profile case of scientific fraud perpetrated by a professor at Harvard. The end result was a finding of guilt, loss of (tenured) job, and legal indictment, all of which seems to have been appropriate. But throughout the proceedings, Harvard was extremely secretive about what was going on, protecting its reputation at the cost of jeopardizing the careers of graduate students and other junior researchers, as well as leaving the professional literature in the field a muddy mess, with no one quite sure which of the ex-professor's papers were fraud and which were legitimate. If German was indeed instructed to communicate only with the President's office, this culture of secrecy and pointless intrigue apparently contributed to German's unnecessary involuntary hospitalization.
posted by biogeo at 8:30 PM on January 22, 2017 [5 favorites]

I wonder if the institutional memory of Amy Bishop may have influenced this situation
posted by knoyers at 9:13 PM on January 22, 2017 [1 favorite]

Yeah, the first thing this made me think of was Amy Bishop. Did they ever work out what was going on with that pipe bomb mailed to her then postdoctoral adviser at Harvard?
posted by Sonny Jim at 12:34 AM on January 23, 2017 [1 favorite]

I was also thinking of the UCLA murder that happened a couple of days before this incident – which doesn't, of course, resolve anything: it could mean that people were extra worried about possible angry and/or unusual behavior in a colleague, or it could be an excuse to take retaliatory action in a setting of heightened fear.
posted by taz at 1:19 AM on January 23, 2017 [1 favorite]

> Now, unless you're willing to credit the idea that these two women fabricated their sworn statements, or that there were tensions between German and Rubin's prior to January 2016 that were left out of the article entirely, Rubin's position seems to be the more reasonable.

Yeah, I don't think this is a one-or-the-other thing. German could have been an "intense" jerk who was increasingly losing his shit with his lab members and the womens' statements utterly and completely credible -- but this has nothing to do IMO with Rubin's position on his research ethics as a PI or the quality of his mentorship.
posted by desuetude at 9:21 AM on January 23, 2017 [4 favorites]

Even if we assume that Rubin initiated the mental health check as retaliation, do we have any reason to believe that Rubin knew that the check was going to involve Hannibal Lectering German in front of his parents at 1 in the morning?

I will admit my ignorance of how mental health evaluations work in cases like this, but my assumption would have been something like Harvard serves German with papers saying he isn't allowed back on campus until he completes the eval.

I've read the findings of fact from the rulings, and I think that the facts seem to support a different interpretation, where Rubin may have been damned either way. If we assume that Rubin is even partially correct about German being a problem well before the whistle-blowing, then you've got a situation where an erratic person with reason to hold a grudge is making a serious and career-threatening accusation. If Rubin (and his other grads and post-docs) were completely innocent of wrong-doing, then yeah, that kind of accusation is just another tickbox on the "we are dealing with a crazy person" checklist. And we have a whole green-background subsite full of people making less that perfect decisions in their dealings with crazy people.

But on the other hand, if Rubin really did oversee academic dishonesty, then all the same facts take on a different light. So I guess my overall take on this situation rests on whatever comes out of Harvard's investigation of the dishonesty accusation).
posted by sparklemotion at 9:56 AM on January 23, 2017 [1 favorite]

The judge's findings of facts didn't really paint a pretty picture of Rubin. As desuetude said, it's very possible that there is something a bit off about German and that Rubin committed legally actionable wrongdoing as well.
posted by grouse at 10:56 AM on January 23, 2017 [3 favorites]

By nearly every account on both sides of this dispute, things soured dramatically in March 2016 which is exactly when German filed his formal misconduct complaint. So it's not really accurate to say that the problems predated the ethics complaint, though Rubin and Muela weren't formally notified about it until May.

It's possible that German became paranoid about the people he had complained against and they had no idea why he was acting weird. It's possible that they remembered his "hypothetical" discussions 3 months earlier, knew about whatever happened that underlay his complaint, and became defensively hostile toward him. (Remember that, by his account, someone else in the lab had told him about the falsified data by Muela the previous summer. So it was floating around and obviously Muela and her boyfriend would know about it if she did something untoward.) It's also possible that someone with Harvard unethically leaked word about the complaint to their star researcher.

We do know -- from Rubin's own affidavit -- that when he was told of the misconduct allegation he immediately suspected German was behind it. And part of the complaint is that Rubin was told about the falsified data before publication and approved it anyway.

By the way, this appears to be the paper with the allegedly falsified data:
Genome-wide RNA-Seq of Human Motor Neurons Implicates Selective ER Stress Activation in Spinal Muscular Atrophy

by Shi-Yan Ng*, Boon Seng Soh5, Natalia Rodriguez-Muela, David G. Hendrickson, Feodor Price, John L. Rinn, Lee L. Rubin
*Co-first author
Published Online: August 27, 2015
posted by msalt at 3:09 PM on January 23, 2017 [1 favorite]

Thank you, msalt, for the link to the judge's ruling. For those interested, here is a key excerpt:

I find that Rubin engaged in five separate willful and malicious acts...

...In the present case, all of Rubin's malicious acts, which were carried out by various methods of calling plaintiff's mental health into question, were taken after Rubin learned of German's allegations [of Rubin's academic fraud]..., despite Rubin's contention that he had been concerned about plaintiff's alleged "unkempt," "disheveled" and "tired" appearance since the beginning of 2016.

...I find that Rubin fabricated concerns and encouraged others to fabricate concerns about plaintiff's mental health in response to his belief that German was responsible for the allegations against him, and Rubin thus acted maliciously.

posted by darkstar at 10:19 AM on January 24, 2017 [2 favorites]

> The judge's statement of facts is indeed strongly in German's favor, but her conclusions and reasoning don't seem to reflect a good understanding of what lab research culture is usually like, and from her statements of the facts, I'm not particularly convinced one way or the other regarding Rubin's or German's version of events. What is definitely clear is that Rubin, the department, and the President's office did not handle this particularly well.

Agreed. Frankly, I don't envy anyone trying to make a legal and/or common-sense decision about this kind of case who is not already familiar with the culture in major research labs, because you have to wrap your head around "norms" that frankly sound a little insane to the average lay person. Also seconding that universities tend to not handle this sort of thing well, it tends to be all secrecy and extreme damage control first, addressing the actual situation second. (Or third. Or worse.)
posted by desuetude at 10:37 AM on January 24, 2017 [1 favorite]

I don't envy anyone trying to make a legal and/or common-sense decision about this kind of case who is not already familiar with the culture in major research labs, because you have to wrap your head around "norms" that frankly sound a little insane to the average lay person.

In this case, which norms do you think the judge didn't understand?
posted by grouse at 12:03 PM on January 24, 2017 [1 favorite]

...I find that Rubin fabricated concerns and encouraged others to fabricate concerns about plaintiff's mental health in response to his belief that German was responsible for the allegations against him, and Rubin thus acted maliciously.

That's the thing... the mental health concerns are "malicious" only if you believe that Rubin made them up. The Judge believes that Rubin made them up because Rubin didn't act on them until after the allegations of dishonesty.

On the other hand, I see it as possible that Rubin's actions were a reasonable escalation in light of what seemed like increasingly erratic behavior (especially if the allegations are false).

Like, it's one thing to accept that you have to work with a paranoid weirdo, it's another thing entirely when the paranoid weirdo starts lashing out at you. Anecdata: I used to be good friends with someone who, even at the time, I wouldn't be surprised to hear had gone on a shooting rampage. But he and I got along, and his crazy expressed itself through rants that were vague enough that I wasn't going to pull a "see-something-say-something" on him. On the other hand, if he'd ever even vaguely threatened me (or more specifically threatened someone else), I would have felt compelled to try to get him some help, which would of course involve talking about his past behavior that I hadn't previously reported/recorded. At which point, I could totally see him accusing me of maliciously defaming him (which, I understand, is how most people who are involuntarily committed feel about the people who initiated that process).
posted by sparklemotion at 1:23 PM on January 24, 2017

In this case though, by Rubin's own words he hadn't seen German for weeks when he called the alarm in. And everyone who had talked to or checked in with him found German to be unproblematic. But Rubin had found out (or strongly and correctly suspected) that German was behind the ethics complaint. So the judge felt that the "concern" Rubin allegedly felt was not consistent with his actions and I have to agree.

Most notably, Rubin could have just talked to German but didn't.
posted by msalt at 12:50 AM on January 25, 2017 [5 favorites]