Wednesday, April 30, 2014

Keep it in the family

Here’s my latest Crucible column in Chemistry World.

_____________________________________________________________________

When Jenny Pickworth Glusker of the Fox Chase Cancer Center in Philadelphia delivered a talk on the past, present and future of crystallography at the opening ceremony of the International Year of Crystallography (IYCr) in January, she not only described but personified the traditions of the field. For Glusker worked in the laboratory of Dorothy Hodgkin, who was a PhD student of J. Desmond Bernal, who was a protégé of William Bragg – who of course started it all, after Max von Laue’s seminal discovery of X-ray diffraction in 1912.

This sort of scientific genealogy is properly a source of pride to those concerned, but it is more than that. From her mentor a scientist acquires not just a technical training but a culture – a sense of what matters, in terms of the scientific questions one asks, an approach to answering them, and the attitude one adopts in researching them. It is hard, for example, to imagine anyone emerging from under the wing of Bernal with much regard for rigid disciplinary boundaries. A part of that culture is surely also the sense of moral and ethical responsibilities that a good mentor will supply.

This is one reason why Alexander Petersen of the IMT Lucca Institute for Advanced Studies and colleagues suggest in a preprint that the trend in science towards large teams is not uncomplicated. That trend has been much remarked on, one implication being that the old mechanisms of rewarding scientific achievement – individual prizes, not least the Nobels – are becoming obsolete. Petersen and colleagues confirm, with a number of metrics, the observation that the number of coauthors on papers is rising across the field in the sciences, and that singleton Nobel prizes are now rather rare.

This poses challenges for attribution of credit (at the same time that such credit is becoming more vital to young researchers), not to mention for the task of simply organizing large teams so that they work efficiently. What has been less remarked is that the potential problems arise not just for team members and leaders, but for the whole scientific community and indeed beyond. As teams get larger, they become less transparent. It is harder to monitor who is doing what, and becomes increasingly necessary to take each contribution on trust.

Petersen and colleagues suggest that this trend could make it easier for misconduct to happen unnoticed, and less likely that there will be channels of mentorship to discourage it in the first place. There are ample examples. Some highly respected scientists have had their reputations tarnished, whether fairly or not, by their apparent failure adequately to scrutinize results falsified by junior colleagues: biologist David Baltimore in the case of Thereza Imanishi-Kari in the late 1980s, and physicist Bertram Batlogg in the case of nanotech fraudster Jan-Hendrik Schön in the early 2000s. Both senior figures were busy people in big labs. But such situations have surely got even harder to manage since those days. Poor management procedures were cited as a reason why the young forensic chemist Annie Dookhan was able to falsify perhaps thousands of drug-test results in the Hinton laboratory in Massachusetts, leading to Dookhan’s prison sentence later last year.

These cases may be extreme, but Petersen and colleagues suggest that it is difficult to maintain chains of responsibility and good conduct in large teams. When things go wrong, for example requiring retraction of a publication, it might be all but impossible to trace the blame. Big teams increase the potential for conflicts of interest, say with researchers peer-reviewing a collaborator’s manuscript, at the same time as making them harder to spot. “In this respect,” the authors say, “we have been witnessing the emergence of a conflict between the scientist and the scientific commons.”

Some of these concerns relate to a sense of values. “Many young scientists have likely been ‘lured’ into postdoctoral traps within large projects”, Petersen and colleagues write. “Are the next crop of scientists trained to be leaders or to just fit into a large production line? And once they enter the tenure track, do the lessons they observed reflect positive scientific values? Or do they reflect a system engaged in productivity at the expense of quality… and pathologically competitive attitudes that run counter to socially beneficial progress?”

Such attitudes may be forced onto young researchers by the prevailing culture. At the IYCr ceremony, a panel of young crystallographers debated the challenges they and their peers face, and in a signed declaration from that event [coming to this site soon…] they say that “Young researchers face problems with long working hours, high pressure and expectations to obtain results and to publish papers quickly and in top journals, job insecurity, and large teaching commitments. These pressures are intensifying, and… they hinder the freedom to explore original and innovative directions or to think about long-term research goals.” They can also motivate misconduct: Dookhan admitted, for example, that she faked results because of “her desire to be seen as a particularly hard working and productive.” Large teams and increasing competition might be an inevitable trend in science, but their consequences for mentorship and ethics need to be faced.

1 comment:

Chemdiary said...

The part about "large teams" and "individual prizes" urged me to ask why we don't have great minds anymore (like Einstein, Pauling etc.) ? I am sure there are still geniuses around, but we don't get to see great discoveries or groundbreaking theories. I wonder if you have any thoughts or a post on this.