Science has struggled with contrition for a while, but the problem reflects a wider trend in society.

The reproducibility crisis in science is old news now. We’ve gotten worryingly inured to the regular stream of paper retractions, the newer and oddly euphemistic “withdrawals” (a term designed to destigmatise, but one that sounds sexual or transactional instead), and earnest editorial expressions of concern. The stream continues. There’s a certain sardonic comedy to journal tables of contents being ballasted now with an end section entitled “latest corrections”.

It’s worth emphasising that a lot of irreproducible or erroneous data aren’t deliberately produced with an intent to deceive – such cases are the sensational but unrepresentative thin end of the wedge. Rather, reproducibility failures are more likely to stem from a lack of oversight, substandard supervision, inadequate training, pressure to publish, and a concomitant failure to instil and enforce standards of data quality, collection, interpretation, and curation. It’s a fair bet that most serial offenders aren’t really aware they’re offending, or if they are, aren’t quite aware of the magnitude of their transgressions. Until the reproducibility detectives come knocking, that is.

When that happens, those first authors should always be given the chance to explain themselves. They are, after all, often one of or the most junior person on the team publishing the paper(s). Granted, they get a larger share of the credit and as such they have to be prepared to take a larger share of the responsibility. But how much of the overall responsibility for the paper(s) should they carry?

Surely the greatest weight of responsibility, given their status as senior and often corresponding author, falls with the group leader instead. That’s what being a group leader means, and it’s why “group leader” is a better and more accurate term than the alternative and rather supercilious “principal investigator”. A “principal investigator” sounds like someone with special status, while “group leader” more properly defines the role of that person in the team. They might not be the cleverest, or the hardest working, or the most gifted, but they are the one with the responsibility of leading the group. And that means it’s their duty to take responsibility when something goes amiss.

How often do they really take responsibility though? Group leaders are usually all too happy to soak up the plaudits on the group’s behalf when things are going well, but get coy when problems arise. The blame for retracted papers is almost always shifted to the first author (see here and here for two examples from Retraction Watch), who will be portrayed as a rogue element, cynically and knowingly profiting from the gullibility of those around them. This overlooks the fact that the author(s) were clearly operating in an environment that enabled problematic data to be produced, and may have been doing so despite the unease of other group members (regardless of a group leader’s opinion, group members usually have a very finely tuned radar for things going too well to be true).

It’s easy to understand the reticence to admit failures. Group leaders without tenure cannot afford to jeopardise their careers, while group leaders with tenure are going to rightly conclude that the reputational risk outweighs the moral imperative. There is more to lose from showing contrition than there is from deflecting blame and riding out the criticism. Whoever heard of a tenured professor voluntarily and pre-emptively offering their resignation as a result of research misconduct that occurred within their group?

And as is so often the case, this behaviour reflects a more general trend in society. The willingness to reap the benefits of light-touch regulation in the good times, the unwillingness to take the hit when problems arise, a new cost/benefit equation that rewards chutzpah over humility, and an awareness that shamelessness and brazen denial will burn off any rain of criticism so long as those defences are deployed with sufficient vigour.

Nowhere is this more visible than in the behaviour of current political “leaders” in the neo-strongman mould (no gender bias here: they are all men – Trump, Johnson, Putin, Bolsonaro, Erdogan, Modi, Xi, Duterte, and more), who never take personal responsibility for events that occur on their watch.

Trump was not the first to use this tactic, but he is the one who has most legitimised it. He disclaims responsibility for his administration’s response to the coronavirus response, he blames ongoing civil unrest on others rather than his own policies of division and factionalisation, and he denies climate change and rolls back environmental protections even as the West Coast burns.

Taking this as his lead, UK Primer Minister Boris Johnson is now not taking responsibility for the implications of a treaty he negotiated with the European Union less than one year ago. Instead, he and his cronies are insisting that they need to break international law rather than see through the consequences of their actions. Over in Russia, Putin has disavowed responsibility for the use of military-grade chemical weapons against political opponents. And on and on and on. The people at the top are never responsible.

This behaviour filters down. These leaders’ underlings have unsurprisingly taken their cue from the boss and decided that it’s better to brazen mistakes and misbehaviour out. In the UK, chief adviser Dominic Cummings is still in his post despite breaking lockdown and offering up a farcical excuse as justification; Secretary of State for Education Gavin Williamson is still in his post despite a exam results debacle that forced a screeching governmental U-turn; Health minister Matt Hancock is busy scapegoating and dismembering Public Health England for the UK’s shambolic pandemic response before the emergency is even over.

Contrition, it seems, is going out of public life. Previously when things went wrong, the person in charge would admit mistakes had happened, apologise, and if need be, take personal responsibility for those failures by resigning. Nowadays this is a vanishing philosophy. Never admit you’re wrong, never take responsibility for mistakes made on your watch, blame others, and make them leave is the new order of the day. Hydroxychloroquine makes a good recent example in both scientific and political terms, with French professor Dider Raoult refusing to admit the flaws in his work, and political supporters of the drug like Trump and Bolsonaro refusing to row back on their endorsement of it.

There is therefore a rather depressing societal moral being drawn, that it’s better to not admit mistakes and ride out the controversy than it is to take responsibility for them. If people misbehave and refuse to admit their wrongdoing, this creates doubt as to whether any wrongdoing actual occurred (a symptom of moral corrosion), and despair if the culpable parties simply refuse to go and are not pushed.

This may in itself be a consequence of the blowback that often accompanies attempts at contrition nowadays in certain (usually online) channels. Taking responsibility can sometimes be like blood in the water, an invitation to one’s opponents to try and claim as many heads as they can (and these days, when debate is valued above discussion and polarisation and scandal are the heartbeat of the news, opponents are everywhere).

Science knows more than most professions about the importance of making mistakes. When dealing with complex and unprecedented situations (research itself is a journey into the unknown), errors are inevitable. But good scientists also know that the way to learn from them is not to run away from them or hide from them, but to face up to them. And that means taking responsibility. Especially so when you’re at the top.

If errors have been made due to a failure of oversight, supervision, training, or pressure, then things are not being run properly. If that’s the case, then you should change the structure. If you can’t or don’t know how, then you should admit as much, and go. People need to start owning up.

Acknowledgements: Many thanks to Ivan Oransky (Retraction Watch) for assistance with this posting.