When bad things happen in an academic laboratory, what should happen to people who bear responsibility for those bad things -- even if they didn't mean for them to happen?

This is the broad question I've been thinking about in connection with the prosecution of chemistry professor Patrick Harran and UCLA in connection with the laboratory accident that killed Sheri Sangji. Potentially, Harran could face jail time, and there has been a good bit of discussion (as in these posts at Chemjobber) about whether that's what he deserves.

I'll be honest: I find myself uncomfortable weighing Harran's actions (and inaction) as worthy of jail time or not, let alone assigning the appropriate number of months or years behind bars to punish him for Sheri Sangji's death. And, other than satisfying our appetite for retribution, I am utterly unsure whether such a penalty in this case would help. I don't know that it would do much to change the conditions and institutions that ought to be changed in the wake of this accident. (On the matter of changing institutions, read the excellent posts at ChemBark and Chemjobber.)

Sheri Sangji's death should alert us that things need to change. Conditions in academic labs need to change. Attitudes and behaviors of PIs, students, and technicians need to change. University departments (which are both builders of knowledge and trainers of new scientists) need to change. What kind of resolution of the prosecution of Prof. Harran could bring about the needed changes?

The best way forward should keep lab accidents like the one that killed Sheri Sangji from happening again. Of course, if we're talking about avoiding such lab accidents, we're assuming this one was preventable through some combination of proper safety equipment and attire, training, supervision, and the like.

Jailing the PI would certainly get the attention of other PIs and would underline the message that they are responsible for safety in their labs, as well as for addressing deficiencies identified in safety inspections (and maybe even for identifying and addressing the deficiencies themselves). Maybe jailing the PI in this case would also make Sheri Sangji's family feel that justice had been served.

But, jailing the PI here might also move him, and the larger problem of making research activities reliably non-lethal, out of the sight of the people who really need to be focused on learning the lesson here.

Maybe jail would make him appear like more of the monster; his lab must have been much worse than ours. Or maybe his absence from the academic research milieu might simply mean the other PIs would return their focus to the pressing problems of securing funding, generating data, and cranking out manuscripts. Perhaps their institutions would be stricter about future safety inspections, but the PIs would do what they needed to do to return to the business as usual. Given the extent to which universities rely on external grants secured by such scientific business-as-usual, it's hard to imagine universities doing much to shake PIs out of this routine.

If we're interested in justice that actually addresses the dangers of business as usual, I think there is another option we should explore.

I don't think Prof. Harran should be allowed to continue with the lines of research he was pursuing when the accident in his lab claimed Sheri Sangji's life. The way he conducted that research --- the way he supervised activities and personnel --- killed someone employed to advance the research. That's a big enough strike to bench him and let other PIs play that knowledge-building zone.

Instead, Harran should devote the remainder of his career to creating a scientific culture --- at UCLA and beyond --- in which the safety of the people performing the experiments (and making the reagents, and fixing the equipment, and cleaning the glassware) is never sacrificed to the goal of getting more and faster results. His mission should be to communicate just how easy it was for a "good PI" to allow lapses in safe procedures, to assume students and staff will figure out how to be safe when using materials or techniques that are new to them, to find tasks more important than supervising lab work, to discourage questions about how to be safe.

This shouldn't be a new service requirement on Harran in addition to his research and his teaching. This should be the core of his job.

He should not only grapple with the soul-searching a decent person does when he's allowed conditions that have killed and underling, but also do that soul-searching in a space where the rest of the scientific community can participate and include themselves in the examination. Harran's presence in this role --- his active involvement with his department in this role --- means that Sheri Sangji and the circumstances that killed her will not be forgotten.

Since research grants would be unlikely to pay for this new set of professorial professional responsibilities --- and since UCLA likely bears some share of responsibility for creating the conditions that killed Sheri Sangji --- UCLA should fully fund these new responsibilities of Harran's position moving forward. As well, UCLA should provide what support is necessary to allow Harran's colleagues (and students and other personnel in their labs) to adapt their own practices in ways that incorporate his lessons. And, it might have a meaningful impact if professional organizations like the American Chemical Society provided funds for Harran to travel and speak to others running academic labs about how to make them safer.

In short, my hunch is that the best way to achieve progress on safe conditions and practices (not to mention relationships in lab groups that help everyone promote safety) is not to separate Harran from his professional community but to return him to that community with a new mission. His new charge would be to help build a better business-as-usual.

It might not be the science career he envisioned, but I reckon it's a job that needs doing. Harran now has ample first-hand knowledge of why it matters.