Recently a Nobel prize-winning scientist had to retract her team’s latest paper. I won’t go into the specifics of it, except that it turned out that their work wasn’t reproducible, the lab notes were incomplete, and generally the results weren’t what they said the results were.
This is no big deal.
First, mistakes happen all the time, even to Nobel prize winners. Heck, they are somewhat busy people (as the winner herself, Frances H. Arnold, said in a tweet), and somewhat busy people don’t always pay attention to the details, trusting their team to make the right decisions. And in the case of scientific research, your “team” is made up of some combination of either colleagues who are just as busy as you, or junior researchers (postdocs, grad students, and even undergrads) who, by definition, aren’t experienced, and hence are prone to making mistakes.
So it shouldn’t be surprising at all that papers contain mistakes. Indeed, I’m willing to bet that every single paper every published contains at least one mistake, or at the very least, one assumption or detail that could be reasonably argued against.
But a paper doesn’t have to be entirely right to be useful; it depends on the severity of the mistake. Sometimes a mistake isn’t big enough to dislodge the point of the paper. Sometimes a flawed paper still introduces a useful idea or concept that goes on to flourish.
Most of the time, however, a typical paper is so minor that a flaw or mistake doesn’t really change the trajectory of the field. If only a dozen people were going to read the paper in great detail and go on to cite it, the mistake won’t have a great impact.
This is the real reason that paper retractions are so rare. It takes a lot of work from outside referees or the wider community to spot such a significant flaw in a paper that a retraction becomes necessary. Most mistakes simply fly under the radar. Or, what usually happens is that the community realizes some flaw in a paper and “works around it” by simply ignoring that section or just moving on – it’s not worth the effort to pursue a full, formal retraction.
In the end, scientists make a lot of mistakes, and slowly over time – whether through retractions or less formal methods – science still advances.
- Substantial undocumented infection facilitates the rapid dissemination of novel coronavirus (SARS-CoV-2) – Science Magazine
- England coronavirus testing has not risen fast enough – science chief – The Guardian
- Coronavirus Tests Science’s Need for Speed Limits – The New York Times
- Trump Falsely Distorts New York Times COVID-19 Science Story – FactCheck.org
- This is the brightest supernova ever seen – Science Magazine
- Coronavirus Today: Science will save us – Los Angeles Times
- Italians stuck at home are measuring light pollution for ‘science on the balcony’ – TechCrunch
- ‘Oumuamua might be a shard of a broken planet – Science News
- College of Arts and Science converts thriving academic programs to departments – Vanderbilt University News