Editor’s Note: This article was originally published by the Observatory of University Ethics on November 3, 2024. It was translated into English by the Observatory before being edited to align with Minding the Campus’s style guidelines. It is crossposted here with permission.
Academic freedom requires that we be free to choose our research topics, but like any freedom, it must be framed. The first limitation lies in our conscience, as we have known since Rabelais. As a doctor and researcher, I cannot choose a research topic that would harm the integrity of people. We all have in mind what the Nazis had put in place in the concentration camps, supposedly to advance scientific knowledge; since then, strict barriers have been put in place, which some, however, pretend not to see by taking liberties with medical ethics, in particular by dispensing with the opinion of the “Committees for the Protection of Persons” that it is mandatory to collect. The necessary scientific control of our research projects and the publications we draw from them is not a real limitation on academic freedom: there can be no research without evaluation, the simple fact of obtaining a position in a university, a large organization such as the CNRS, Inserm or INRAE, or a hospital for a doctor, is accompanied by an obligation, that of complying with the rules of evaluation.
A second limitation to academic freedom comes from the need to finance the research that we plan. We all need, depending on the discipline, and even for research in philosophy and literature, various products, small equipment, large machines, computing time, etc., but above all, people to work with us, doctoral students, post-docs, engineers, technicians, investigators, statisticians, and others. Public, charitable, and private funds are used after evaluation of the projects by our peers and competent scientists, but the wording of the “calls for projects” often limits our freedoms: we must propose profitable research topics, which can lead to applications, what is called the steering of research downstream. Basic research, not finalized towards the discovery of practical solutions—a drug, for example, for the medical field—is essential; some very fundamental projects will turn out to end up on a dead end, others on the contrary, will be very fruitful, but there is no way of knowing this in advance. To stay in the medical field, basic research is the mother of therapeutic progress, and we could give specific examples.
Is research programmable? It is a question I had asked during the vote on the last research programming law, which was, in fact, a research programming law funding research. Unfortunately, the type of funding favored by this law was actually aimed at programming even more research topics. Should we remind you that this is an aberration? We can only program the me-tool, not the fundamental and creative works. A sentence like “Mr. Fleming, please submit a research project aimed at discovering anti-bacterial substances” does not make sense.
[RELATED: MTC Launches Minding the World]
On the other hand, this other one: “Mr. So-and-so, submit a research project aimed at discovering new antibiotics,” has a meaning. The serendipity governs the world of discovery, not that of research as it is conceived by the guardians. If Fleming had not been free to exploit an unexpected observation, we might not have antibiotics today. I fear that, nowadays, no Inserm commission would give him credit for taking advantage of this observation.
These limitations do not prevent bizarre research from being promoted and funded, such as those aimed at extending human life beyond one hundred years or at developing brain transplants–or whole-body transplants–depending on your point of view. It is necessary to exercise scientific control over the subjects proposed if only to ensure their plausibility. Inserm had to stop funding Benveniste’s work at the end of the 1980s because it was moving away from the framework of science and it was possible to demonstrate this. This decision was taken collegially, by a specialized scientific commission of which I was a member, and I can say that it was painful and was not taken lightly. But if this scientific control is essential for the allocation of research funding—not to mention the a posteriori control that is never done—this control must never be an ideological control as is tending to be put in place. Benveniste compares himself to Galileo, but it was indeed an ideological control that was exercised over the latter’s work, put in place by Pope Urban VIII, it was not a scientific control: the pope had perfectly seen through the telescope the moons of Jupiter that Galileo had shown him and he was convinced that they did indeed exist.
Today, a different type of ideological control is exercised over research, and, as Nathalie Heinich pointed out, confusion has arisen between the essential scientific control of research and an unacceptable ideological control, leading to censorship and sometimes to self-censorship by researchers themselves. Not only does activism, whatever its justifications, operate outside of research, but it harms all research. There are research topics that fall within the scope of single, right-thinking thought and others that do not. Of course, this control is much more prevalent in the human sciences than in the so-called “hard” sciences, but medicine, for example, is not even free of it. Joseph Ciccolini described on this site the cancer research topics that he identified at a major conference in the discipline and which demonstrate the right thinking of the researchers: “Racial/ethnic disparities in locoregional recurrence in patients with hormone receptor-negative node-negative breast cancer, “Real-world outcomes of treatment of black women compared to non-Hispanic white women with advanced triple-negative breast cancer.” Like the research subjects, the results either conform to the pre-established conclusions or do not. Of course, black women always appear disadvantaged in these studies, but these studies are systematically mono-parametric and do not take into account factors other than ethnicity: economic and social level, education, etc.
Along the same lines, an article from the Journal of the National Cancer Institute, whose I did the analysis, assessed the depth of tumor DNA sequencing performed on whites and blacks: of course, to the “disfavor” of blacks, for reasons that have nothing to do with the color of their skin. The depth of sequencing has no effect on diagnosis, prognosis, or treatment, but the authors of the article have found their stripes that will make them well-regarded in woke circles. One may wonder whether the authors would have dared to publish their work if they had found opposite results or, more simply, an absence of differences between the results observed in whites and blacks. This question is not theoretical.
An article published in 2020 about Proceedings of the National Academy of Sciences of the USA (PNAS) has been very successful and cited several hundred times. It was even cited by Supreme Court Justice Ketanji Brown Jackson because it shows the benefits of diversity: “It saves lives,“ she wrote. The study focused on the “patient–physician concordance on health care outcomes for ‘underrepresented’ minorities, as it can mitigate external bias, boost communication, and increase trust.” The authors started from the very real observation that in the United States, the mortality rate of black newborns is three times higher than that of white newborns and looked for the causes. Their results suggest that “newborn-physician racial concordance is associated with significantly reduced mortality among black newborns. [They] further suggest that these benefits occur in more difficult deliveries and in hospitals that deliver more black babies.” So, if the study was well conducted and its results unbiased, the conclusion is that white doctors are not taking enough care of black newborns. This is clearly unacceptable and must be urgently remedied.
Yes, but four years later, the same scientific journal published a reanalysis of the same data, which did not reach the same conclusions. “The effect is considerably weakened and often becomes statistically insignificant once one takes into account the impact of very low birth weight on mortality and the fact that black physicians are less likely to see the population at highest risk of low birth weight infants.” In addition, ” These results […] suggest that, in order to reduce the gap between blacks and whites, we must combat the incidence of these low birth weights in black newborns.” In short, we need to act, not at the level of delivery rooms by matching the race of doctors to that of newborns, but before birth, by ensuring satisfactory socioeconomic conditions for pregnant black women. However, matching a doctor with a newborn costs much less than improving the living conditions of black Americans:
We can recall that Martin Luther King, after years of activism for civil rights, after many battles, ended up being recognized by the entire American political class; but his aura weakened when he said that it would also be necessary to take care of socio-economic conditions, ‘feed the hungry and clothe the naked;’” he was abandoned by the media that had supported him until then … and he was assassinated. Civil rights are all well and good, after you have put your wallet in a safe place.
I have only seemingly strayed from my subject, which is academic freedom. It has been shown in an article that only six percent of American and Canadian social psychologists described themselves as conservatives. The latter “fear the negative consequences of revealing their political beliefs to their colleagues.” They are right to do so: “Many social psychologists have said they would discriminate against openly conservative colleagues whether they had to review a job application, a grant application or an article manuscript. The more liberal the respondents were [in the American sense, that is to say “left” in the French sense], the more they said they would discriminate.”
A renowned psychologist, Paul Bloom, a professor at Yale University and then at the University of Toronto, analyzes the question of self-censorship. in a ticket from the Chronicle of Higher Education. A study in 2024 found that most professors—including those on the left, although it was more common on the right—self-censor their publications on controversial topics in psychology. They worry about the negative consequences for their social and professional lives. While they have little to fear from most of their colleagues who regard these fears as unfounded and have great contempt for colleagues who request retractions of articles on “moral” grounds, there is a minority of professors who believe that an appropriate response to certain points of view may include “ostracization, public labeling with pejorative terms, refusal to publish work regardless of its merits, refusal to recruit or promote a researcher even if formal requirements are met, dismissal, shaming on social media, and removal from leadership positions.”
Finally, some major journals explicitly state that political implications partly determine what can be published. Here are the journal guidelines Nature Communications., developed in 2020 following an investigation conducted by an editorial manager regarding an article (later retracted by the journal) on the supervision of researchers, which concluded that male supervisors could, better than female supervisors, contribute to raising the status of women in science:
We have strengthened our commitment to supporting diversity, equity and inclusion in research. […] We have also revised our editorial practices and policies, and in recent weeks have developed additional internal guidelines and updated information for authors. We recognise that it is essential to ensure that these studies are considered from multiple perspectives, including those of the groups affected by the findings. We believe this will help us ensure that the review process considers the dimension of potential harm and that claims are tempered by consideration of study limitations when findings have potential policy implications.
The most recent editorial of Nature Human Behavior describes the new procedures of this review for the reviewers and the editors thus entitled: “New ethics guidelines address potential harms to human groups who do not participate in the research but who could be harmed by its publication ” and continues: “Science has for too long been complicit in perpetuating structural inequalities and discrimination in society. With this guidance, we are taking a step forward to counteract this. ” This is in line with Nature Communications’s editorial policy: a paper concluding that female researchers perform better with male mentors will not be published because these findings, even if true, will cause “potential harm” or make the journal “complicit in the perpetuation of structural inequalities” and discrimination in society.
So, journals instruct what is acceptable to publish and what is not! Regardless of the scientific content and the quality of the analyses and investigations, they must lead to conclusions that are known in advance. Under these conditions, researchers working on “sensitive” subjects—and we have seen above that these subjects can be found just as much in medicine as in the human sciences, even if this type of research is undoubtedly more frequent there—have the choice between: (i) not submitting for publication a work that leads to “incorrect” conclusions; (ii) cheating a little to make the conclusions presentable, for example by omitting to insert part of the analyses; (iii) denigrating the work that one has carried out, by saying that one has not taken into account factors that would have made it compatible with the doxa; (iv) changing the subject of research.
[RELATED: Gender? It Doesn’t Exist]
In any case, this is called self-censorship.
A wonderful example of this was provided by the New York Times very recently. An American study on the use of puberty blockers was not published in a scientific journal by its authors because the results did not suit them. Let’s put the problem back into context. In the name of the “freedom” of adolescents who choose to change sex, some doctors advocate the prescription of “puberty blockers.” This allows, they say, to delay the onset of puberty, and they claim without proof that it reduces their psychological disorders. One of the doctors involved, Dr. Johanna Olson-Kennedy, wanted to provide arguments to defend this point of view, but the results she obtained are contrary to her hypotheses: no, there is no improvement in the gender distress adolescents who follow this treatment.
But since these results are not consistent with the initial hypotheses and “that they could fuel political attacks that have led to bans on gender-based treatments for young people in more than 20 states,” she refuses to publish them. The entire research project was funded to the tune of nearly $10 million by the National Institutes of Health, but what does it matter? It is not impossible that she will ultimately be forced to publish them, the NIH not hesitating, if a subsidized and carried out work is not published, whatever its results, to demand reimbursement of the grants received.
Image by Ralf Geithe — Adobe Stock — Asset ID#: 312826020
This article was originally published at www.mindingthecampus.org
