Thursday, February 9, 2012

How do institutions change? Not easily

Reading Paul's thoughts about both Sheri Sangji and Jason Altom were interesting and saddening. Go over there and read the whole thing! Here's the conclusion that I found affecting:
Changing the culture of an institution—especially one as intractable as chemical academia—is extraordinarily difficult. But so long as we forgo meaningful changes in favor of cosmetic ones that we don’t even bother to sustain anyway, we will continue to experience frustration and tragedy. One wonders what magnitude of disruption is necessary for our community to commit itself to improvement. Apparently, it is much greater than the death of a twenty-something student.
How do large institutions change permanently?  I can think of a few ways -- readers can probably think of more:
  • The gun to the head: Funding is added or taken away, new laws are passed, lawsuits are lost (or won). Change is forced on the institution. 
  • From the top: Leadership decides (for whatever reason) to make changes. Examples: handover at the White House between presidencies, the transition at Apple between Steve Jobs and Tim Cook, Microsoft's late-90's pivot to the Internet. 
  • From the bottom: As an organization's people changes, the organization changes with it. Imagine the difference between the US Army between 2001-2003 (senior people may have served in Operation Desert Storm, most have some foreign peacekeeping deployments, few combat veterans) and the current US Army (the median infantry soldier has at least one deployment, most officers and senior NCOs have multiple tours and are combat veterans.) 
  • Voluntary change: People within the institution have a change of heart, decide to alter their direction. Examples: I can't really think of any (doesn't mean there aren't). 
Right now, those of us who follow academic chemical safety are hoping that "voluntary change" happens. It remains to be seen whether the charging of Professor Patrick Harran in the death of Ms. Sheri Sangji will be a sufficient gun to the head to change academic chemistry. Somehow I doubt it. According to Paul, Jason Altom's suicide was an insufficient gun to the head of the Harvard chemistry department to institute any permanent change (although I'm guessing that depression in graduate students or postdocs is taken fairly seriously by smart faculty members.) 

If enough people like Paul have faculty positions and attain high enough positions in academic chemistry, will things change? Probably. I think change from the bottom will be slow and painful, but has the best chance of becoming permanent. 

There is, of course, the Langerman/Benderly approach to academic chemical safety: that funding and tenure be threatened in the case of bad safety records. I am on the record as seeing it as (at best) a useful bad idea. It's a Desert Eagle approach -- it will be interesting to see if it suffices. 


  1. The one time I saw drastic, quick change from the top was when McNerny came from GE,took over 3M and instituted Six Sigma. It was amazing how he got management in the room, gave them their little red book (The Six Sigma Way by Pande, et al.), and boom, just like that, they were converted. What everyone knew was that, being the first non-3Mer to run the company since it was started, he had been brought in by the board for just that reason, and so there was no where to turn to counter him. It also helped that it wasn't a distant edict from on high - he was part of the process. That's how you enact change.

  2. Change can come from all these ways but pushing from the bottom is so much more difficult without buy-in from the top.

  3. The former column you wrote on June 6 2009 that was referred to almost seems written by another Chemjobber altogether.

    You should really clarify why you think using tenure and grant funding is less preferable than having criminal charges filed against you. Or why having an obstinate "Bangladeshi postdoc" operating outside minimal safety requirements is in any way acceptable in a modern lab. What's next? "Well, crapping on the floor in the middle of the lab is normal where the new grad student's from, so we shouldn't really say anything to him." And heaven forbid the long arm of the NIH/NSF come down into lab and tell us how to do our jobs! I mean, they set guidelines for human testing and even the treatment of animal subjects. Are grad student lives worth less?

    1. bw, you bring up an interesting point. Let me see if I can reconcile past CJ and current CJ (not that I think that I've changed my mind):

      So here's the bit from Ms. Benderly's Slate article ( that I was responding to:

      "If the point of medical research is to save lives, then the NIH and NSF should insist—on pain of dismissal, like the chemical companies—that researchers follow all accepted safety procedures.

      If Sheri Sangji's death is to mean anything, it must be that no lab chief—and certainly no federal agency—claiming to further human welfare ever again tolerates the risk of harm to lab workers. That means that university administrators from the provost on down must make safety a serious concern and a requirement for career advancement and hiring, and tenure and promotion committees must hold faculty members responsible for seeing that everyone in their labs has the training, skills, and equipment needed to work safely. Funding agencies must make a good safety record and evidence of safety awareness real conditions for getting and keeping grants. Never again should academic research needlessly claim the life of a researcher."

      As I said, it's far easier to fake a good safety record than to prove a good one -- how would you go about implementing such a set of draconian policies? It's much more effective to attempt to engender cultural change by emphasizing best practices and changing the minds of both leadership and the rank and file.

      To respond to your individual points:

      1) I don't think either are preferable, but if it's a choice between NIH/NSF/universities attempting to implement such a draconian set of policies and the risk of criminal charges if someone is OSHA-recordably injured*, I'd much rather risk the criminal charges.

      Here's a fair rubric of where I stand:

      A) University and department EH&S personnel need to draw from a more experienced pool of scientists; they need better funding and they need more support from the university administration.
      B) If it's an OSHA-recordable injury to a student (i.e. a trip to the clinic/hospital), a full rotorooter from the university/department EH&S of the PI/group in question should happen.
      C) If someone dies, all bets are off.

      Sound like a fair assessment?

      2) As for the obstinate postdoc, it's much smarter to have the PI or the department chair deal with the issue than for NIH/NSF to do the same. I don't really think any of this has to do with cultural standards, just that it's much more effective to have personnel problems dealt with at the lowest possible level. Forcing behavioral change from atop and afar can be both ineffective and painful.

      3) Compliance for human/animal subjects is simply different; they're totally at the mercy of the study director. I don't think the power dynamic between scientist/rat and PI/young lab tech is (jokes aside) comparable.

      What do you think, bad wolf? Where have I gone wrong?

      *needs medical attention

    2. Thanks for the well-thought out response, cj, especially to my somewhat scattered thoughts.

      1) I guess i see the NIH/NSF as a potential source of guidelines that set a minimum standard of behavior, and adherence to which would be required for continued funding. The change we all appear to want has to come from multiple directions, and top-down has a role, in that the one thing PIs will do is try to increase their fundability. I don't see them as enforcing or patrolling but i think making funding contingent on a record of safety discussions, promises to work within some guidelines, and a history of compliance. I don't believe in zero-tolerance labs either but conditions can and should be more 'industrial' than today. So i am strongly in favor of your (A).

      2) The hypothetical rogue postdoc should be brought into compliance with safety standards by coworkers or PI, certainly, but the pressure on the others in the lab has to be there, to avoid "Well, she doesn't really speak english well enough to discuss this, let her work however she wants."

      3) I doubt i really know the specific requirements regarding animal subjects to comment but again i'm thinking that these guidelines can be made with the same fiat as the other onerous things demanded of a PI. You want money? Sure, show us that you're conscious of lab safety. Blow off a bunch of inspections and laugh about the stuff you set on fire? Your cowboys can fund themselves.

      It also occurred to me that the DEA mandates that we not cook meth in our labs, but beyond that they leave the enforcement to schools, departments, etc. and of course, the other people in your lab that should strongly disapprove of such activities. This seems to have been largely effective.

  4. I don't think change will come from the bottom (i.e., students) in chemistry because (i) there is so much turnover that it would be hard to sustain momentum for a substantial period of time, (ii) many/most students fear the power held by advisors and other professors, and (iii) it takes several years in grad school before you figure out what is going on.

    Rather, those who truly care are going to have to "play the game" and work their way into positions of power. Once you're a professor, there is a lot you can do as an individual to effect change at the school level, but you're going to have to be willing to spend time working on it.

    1. I agree completely, thus the "If enough people like Paul have faculty positions and attain high enough positions in academic chemistry, will things change? Probably." bit.

    2. This comment has been removed by the author.

    3. Fair enough re: students, but I think the Sangji case has done quite a bit to ratchet up peer safety culture among students, i.e, "Hey, friend, perhaps you should not do that..."

      ...or I hope so, anyway.

  5. Results! Results! Results! This trumps everything. I've seen people who were truly sociopaths, terrorizing their students, yet they got tenure. It was an open secret- research $ trumps all bad behavior. The fact that Harvard dismantled all the (token) programs for student welfare is the most telling, and damning piece of evidence.

  6. CJ:

    I've said this in other places, but people (as institutions) change when the pain of the change is less than the pain of staying staying the same.

    The LA DA's actions in the Sangji case raised the stakes: The pain of NOT changing one's ways is now the actual loss of a career.

    The DA's actions send a single message simultaneously to ALL management levels in the Academy: "You will be held accountable for your actions regarding employee safety." Both "upper management" (Provost/Board) and "front line management" (PIs) at UCLA (should have) received that message loud-and-clear. Certainly other California institutions have gotten that message and the same message should resonate throughout the Academy nationwide. Like it or not, the DA has provided added incentive for "buy in" at all levels.

    Harry Elston
    Editor, JCHAS

  7. You are talking about cultural changes within a community. Although I agree with a democratic evolution over time as knowledge slowly builds, as well as changes made following inquiries into something, like safety, I feel that real change comes about by scientific discovery and paradigm shifts (in our view of Nature). Ernst Mach, a positivist, held back science due to his prestige and stubbornness to accept the existence of molecules and atoms. There are many examples but paradigm shifts are not only quite rare but are usually only established after the old guard has passed on.


looks like Blogger doesn't work with anonymous comments from Chrome browsers at the moment - works in Microsoft Edge, or from Chrome with a Blogger account - sorry! CJ 3/21/20