Talking about science in crisis is something that is becoming increasingly common, and the proof of this is what happened on January 2nd, when the prestigious magazine Science withdrew one of the articles by Frances Arnold, winner of the Nobel Prize in Chemistry in 2018. This is a sign of the deep scientific crisis we are experiencing.
But what are the implications? It is not common, of course, although every day more and more scientific articles are withdrawn due to failures of all kinds: from the reproducibility of approaches, to the falsification of results. The conclusions are very clear: science has a problem and it must be solved.
Science in crisis
Frances H. Arnold is one of the most brilliant women of this century: she has behind her the Nobel Prize in Chemistry in 2018, being the first woman to win the Millennium Technology Award or be one of the most efficient professors at Caltech. With hundreds of publications signed and run by herself and her team, her reputation is (almost) unimpeachable. Not long ago, Arnold was experiencing the worst possible blow to the world of science: retraction.
The magazine Science recently announced the withdrawal of the article “Site-selective enzymatic C‒H amidation for synthesis of diverse lactams“, at the request of the authors themselves, including Arnold herself, although not the first, who would be Inha Cho. According to the journal,
“Efforts to reproduce the work have shown that enzymes do not catalyze reactions with the activities and selectivities claimed. A careful examination of the first author’s (InHa Cho) laboratory notebook revealed that contemporary entries and raw data from key experiments were missing”.
The lack of raw information prevents a comparison of whether the information provided in the article has an actual source or has been modified or “invented”. On the other hand, not being able to reproduce the experiment is not always the worst sign in the world of science, but it is when the original information is missing, since it incites suspicion about the article’s lack of veracity. This should not serve to question Arnold’s career, since in any case the main responsibility falls on the first author, but it is a problem, as the researcher herself admits, of lack of responsibility on her part, for not ensuring the quality of the data under scrutiny.
For my first work-related tweet of 2020, I am totally bummed to announce that we have retracted last year's paper on enzymatic synthesis of beta-lactams. The work has not been reproducible. https://t.co/Dk1TFw0FY9
— Frances Arnold (@francesarnold) January 2, 2020
In the end, the authors, including Arnold, have decided to withdraw the article because of a serious lack of information or because of a fabrication of the data. In other words, this is the worst thing that can happen to a scientist: he can be accused of falsifying research and have his article retracted, calling into question his judgment and scientific integrity, which is more important to the trust of this discipline than skill or ingenuity.
How does science work?
To understand why trust in scientists is so important, one must understand how “science” works. The current scientific system is a modus operandi that discriminates by nature between what can be considered science and what is not science and even what can be categorized as bad science. Today, this has a clear manifestation in the form of articles or papers. These are documents that describe in a concrete way a scientific advance, with its result, the methodology followed, the most current review of the “state of the art” and other information that allows other scientists to follow each finding to its very origin.
One of the most important points which are published through a method known as peer review is that other experts in the field review the research before it is published. These independent experts, or peers, give you the go-ahead to ensure the quality and scientific integrity of the finding. But, of course, these reviewing scientists may be wrong or, worse, they may have helped commit fraud by the authors.
This often becomes apparent when, once the study is published, other scientists try to reproduce a discovery in order to move on to a new one and find that what was proposed in the original research simply doesn’t work. And this is exactly what happened with Dr. Arnold’s retracted paper.
When this fails, and new “science” is published based on false or erroneous publications, the system ends up failing completely. How can we find new technological solutions, new medicines, new sources of energy if their development is based on scientific errors or attempts at deception? For this reason, simplified, modern science puts more weight on the integrity of scientists than on their own abilities.
The crisis of science
Currently, the scientific system is experiencing a crisis whose problems are supported by various points. One of them, precisely, is that of scientific fraud. Every year, more than 1,400 articles are withdrawn, a number that is growing little by little. This is partly because many are not reproducible, that is, it is not possible for other scientists to obtain the same results proposed by researchers. However, this does not always mean that the studies are false. Other articles are simply withdrawn as a precautionary measure because sufficient information about the research is lacking.
But what is really worrying is that most of the articles withdrawn are withdrawn because of fraud demonstrated by the evidence presented after the review. And this usually happens after they have gone through the peer review, which highlights the two main problems of this crisis. The first problem is the frenzy of publications and the need to publish in the journals with the highest impact rate. This is known as “publish or perish“.
Those who do not publish quickly enough and in sufficiently prestigious publications are left “out of science”
Those who do not publish quickly enough and in sufficiently prestigious publications are left “out of science” for lack of funds, support and prestige. This false premise is taken as true by researchers, who in fact already live in an extremely competitive world with little recognition. In many cases, we may dare to assume that because of this enormous pressure, researchers may be tempted to commit fraud in order to maintain their status.
Secondly, there is another major problem in the world of publications, and that is the publishing crisis. Precisely, as we said, the impact factor is a tremendously important issue for researchers. However, peers, or experts, are neither paid nor practically recognised beyond the fact that they can be “peers”. Nor is it possible for one of these experts to reproduce the experiment and carry out the relevant tests to ensure that the research is entirely reliable.
The worst thing about this complex scientific crisis, where there are professionals drowning in the needs of the system and experts who are not motivated by the system to fulfil their role as professionals who scrutinise the studies, is that it is the system itself that suffers. Science then, begins to see failures as serious as a Nobel Prize winner getting involved in a plot to retract papers because of falsified results, something terribly serious for the researcher and for science as a whole.