Natural Science: Ethics | Quomodo Scimus : IBTOK

areas of knowledge - NAtural science

Ethics of Natural Science

Science can tell you what is possible. It cannot tell you whether to do it.

THE PROVOCATION

The consent form described the procedure as an "HIV vaccine." It did not mention gene editing.

First genetically modified babies

In November 2018, He Jiankui announced that he had used CRISPR-Cas9 to alter the germline of human embryos - changes that are permanent and heritable - with the stated aim of conferring resistance to HIV. He acted without institutional oversight, in violation of international guidelines, and without disclosing to the parents what the procedure actually involved. The scientific community condemned the work near-unanimously. He was sentenced to three years in prison in 2019.

In November 2018, He Jiankui announced that he had used CRISPR-Cas9 to edit the germline of human embryos, resulting in the birth of twin girls whose DNA had been permanently and heritably altered. His stated aim was to confer resistance to HIV. He acted without the knowledge of his institution, in violation of existing international guidelines, and without disclosing to the parents in plain terms what he was doing to their children and their children's children.

The scientific community condemned the work near-unanimously. He was sentenced to three years in prison in 2019. What is harder to answer is what the condemnation was based on - whether the objections were scientific, ethical, or both, and whether settling one kind of objection settles the other.

Big idea 1 - Scientific knowledge does not come with instructions for its use

The physicists who built the atomic bomb understood this better than most. Leo Szilard, one of the key figures in the Manhattan Project, circulated a petition in July 1945, signed by around 70 scientists, arguing that the bomb should not be used on Japanese cities without a prior demonstration of its power. The petition was classified and never reached Truman. Joseph Rotblat had already left the project the previous year - the only physicist to do so on ethical grounds. Germany had abandoned its nuclear programme, removing the original justification, and Rotblat concluded that continuing was indefensible. He later co-founded the Pugwash Conferences on Science and World Affairs, a sustained attempt by scientists to take collective responsibility for the uses of their knowledge.

J. Robert Oppenheimer took a different position, and articulated it with uncomfortable clarity in security hearings after the war. Commenting on the decision to build the bomb, he said: "When you see something that is technically sweet, you go ahead and do it and you argue about what to do about it only after you have had your technical success." Every Group 4 internal assessment asks students to address the ethical implications of their investigation - a formal acknowledgement that the knowing/doing question must be raised, not assumed away.

Jennifer Doudna, one of the scientists who developed CRISPR-Cas9, found herself returning to the same Oppenheimer parallel as the technology spread globally. In A Crack in Creation, she describes a nightmare in which a colleague leads her into a room to meet someone who wants to understand the gene-editing tool she has helped create. The person in the room is Adolf Hitler. She connects the dream directly to Oppenheimer's admission: that scientific communities tend to pursue what is technically possible and address the ethics afterward. Her concern was not only that CRISPR could be misused, but that by the time the question was being asked seriously, the technology was already in hundreds of labs worldwide and available online for a hundred dollars.

The same structure has appeared in artificial intelligence development. Geoffrey Hinton, one of the pioneers of deep learning, resigned from Google in 2023 in order to speak freely about the risks of the technology he spent decades building - a position structurally close to Rotblat's, though the risks are less precisely defined. Stuart Russell, whose textbook has trained most of the field, has argued that AI researchers have been building systems optimised for specific objectives without adequately asking whether those objectives are the right ones. An open letter signed by thousands of researchers in 2023 called for a pause on training AI systems more powerful than GPT-4; like Szilard's petition, it was largely disregarded. Oppenheimer's description of how the bomb was built - pursue what is technically possible, address the consequences afterward - is recognisable as a description of how the most consequential knowledge tends to develop, in field after field.

Big idea 2 - The history of medical research is also a history of whose bodies it used.

Watch the video below. It is the greatest scene ever filmed for television, from probably the greatest documentary series ever made.

Jacob Bronowski - The Ascent of Man

The Ascent of Man is a landmark 1973 BBC documentary series in thirteen episodes, written and presented by Polish-British mathematician and polymath Jacob Bronowski. Inspired by Kenneth Clark's Civilisation, it traced the development of human knowledge from stone tools to nuclear physics, arguing that science and art are expressions of the same creative impulse. Bronowski died in 1974, the year after the series aired. It remains one of the most ambitious works of science communication ever made.

In 1973, Jacob Bronowski filmed the final episode of his television series The Ascent of Man at Auschwitz-Birkenau. Unscripted as always and speaking direct to camera, he waded into the settling pond where the ashes of Nazi victims had been flushed, among them many members of his own family. He stood there as a survivor and a witness and spoke about science, certainty, and what happens when scientific authority is severed from doubt. Near the end of the clip he names his friend Leo Szilard.

Between 1939 and 1945, doctors in the German concentration camp system subjected prisoners to high altitude decompression, hypothermia, deliberate infection with malaria and typhus, surgical removal of muscle and bone without anaesthetic, and exposure to poisons and burns, in the stated service of military medicine and racial science. At the Nuremberg Doctors' Trial, concluded in 1947, 23 physicians were tried for war crimes and crimes against humanity. Seven were sentenced to death. The trial's judgment produced the Nuremberg Code: ten principles for the ethical conduct of research on human subjects, the first of which was voluntary informed consent.

Harriet A. Washington's book Medical Apartheid argues that the pattern of conducting research on human subjects without consent did not end with the camps, and did not begin with them. The history she traces runs from slavery-era medicine in the South through the Jim Crow period to federally funded research programmes in the twentieth century. The populations that served as uninformed or involuntary subjects - predominantly Black Americans - were also the populations least likely to receive the treatments their participation helped develop.

The Tuskegee Syphilis Study had been running since 1932. The United States Public Health Service enrolled 399 Black men in Macon County, Alabama, to observe the natural progression of untreated syphilis. In 1947, penicillin became the standard treatment. The men were not told and were not treated. Tuskegee had been running for fifteen years when the Nuremberg Code was adopted, and continued for a further 25 years after it, until journalist Jean Heller exposed it in the Associated Press in 1972. The continuation proceeded in the presence of an available treatment and a judgment that had already established, in the sentencing of seven physicians, where the absence of consent leads.

Henrietta Lacks never agreed to what was done with her cells. In 1951, during treatment for cervical cancer at Johns Hopkins, a sample was taken during a biopsy without her knowledge and passed to a researcher attempting to grow the first immortal human cell line. The cells, labelled HeLa, survived and replicated where every previous sample had failed. They were used in developing the polio vaccine, chemotherapy protocols, gene mapping, cloning research, and in vitro fertilisation. They have been bought, sold, and shipped by the trillions to laboratories worldwide. Her family learned the cells were still alive in the 1970s, twenty years after her death, when researchers contacted them for blood samples without explaining why. Rebecca Skloot's account of the Lacks family raises the question Washington's book traces across a longer history: when knowledge is produced using someone's body without their consent, who owns what was made?

The cases in this section share a common structure: a decision, often implicit, that certain subjects' suffering matters less than the knowledge to be gained. Harriet Washington's argument is that this decision tracked race in American medical research for two centuries. Peter Singer's argument in Practical Ethics is that the same logic extends further than most medical ethicists are prepared to follow. Jeremy Bentham, writing in 1789, asked what property of a being could possibly be the relevant criterion for moral consideration and answered his own question: "The question is not, Can they reason? nor Can they talk? but, Can they suffer?" If the capacity to suffer is what creates a claim on our moral attention - which is the principle that makes Tuskegee and the Nuremberg Doctors' Trial wrong - then Singer argues the line between human and non-human research subjects cannot be drawn on principled grounds. Many millions of animals are used in laboratory experiments each year. The ethical argument that condemned those trials, extended consistently, reaches further than medicine has been willing to go.

Big idea 3 - The appearance of scientific controversy can be manufactured

Merchants of Doubt

Merchants of Doubt is a 2014 documentary directed by Robert Kenner, adapting Naomi Oreskes and Erik Conway's book of the same name. It follows the careers of a small group of scientists who moved between denial campaigns across four decades - tobacco, acid rain, the ozone hole, climate change - working for hire to manufacture the appearance of scientific controversy. Several of the scientists appear on camera. The film shows the strategy not as a conspiracy but as a professional service, openly offered and openly bought.

An internal memorandum from a major American tobacco company described the strategy plainly: manufacture the appearance of uncertainty about the health effects of smoking. The industry's own scientists had already concluded that smoking caused cancer. If public doubt could be sustained, regulation could be delayed. The strategy worked for decades.

Naomi Oreskes and Erik Conway's Merchants of Doubt traces what happened when the same approach was applied to other fields where scientific consensus threatened powerful economic interests: acid rain, the ozone hole, second-hand smoke, and climate change. A small group of scientists, several of them physicists with Cold War credentials and strong conservative commitments to free markets, moved between these campaigns across four decades, producing alternative reports, disputing consensus findings in public, and appearing before regulatory bodies as credible scientific voices. They were recruited precisely because they were not fringe figures.

The Intergovernmental Panel on Climate Change rates its findings by confidence level. "Very likely" means a probability of 90 per cent or higher. In scientific usage, this is a strong claim. Received by a non-specialist audience or a legislature, the phrase reads as uncertainty. Oreskes and Conway document how this gap was systematically exploited: scientists communicating findings accurately in technical language were outmanoeuvred by people who understood how that language would land with those not trained to read it. The campaign raises two questions the science itself cannot resolve: what the people who manufactured the doubt were doing to the practice of science, and what scientists are obliged to do when they know their accurate findings will be reliably misread.

The manufactured doubt required a distribution mechanism, and the journalistic norm of balanced reporting provided one. If 97 per cent of climate scientists agreed that human activity was warming the planet, editorial balance logic suggested finding a dissenting scientist and giving them equal airtime. The result was the systematic misrepresentation of where scientific opinion actually stood, presented as fairness. The BBC formally acknowledged in 2018 that its coverage had given undue weight to fringe views on climate, having spent years treating the question as more contested than the evidence warranted.

The doubt, once manufactured and distributed, found willing amplifiers in political systems. Donald Trump withdrew the United States from the Paris Climate Agreement in his first term, rejoined it briefly under Biden, and withdrew again on the first day of his second term in 2025. The language of scientific uncertainty, originally produced by industry consultants in the tobacco campaigns, moved into executive orders and budget decisions, removing climate scientists from advisory bodies and cutting federal research programmes. What Oreskes and Conway traced as a lobbying strategy had, by the mid-2020s, become integrated into a strand of government that treated scientific consensus on climate as a political position to be contested rather than a body of evidence requiring a response.

The ethics of Natural Science compared

What does science owe the people it studies - and the futures it creates?

In Mathematics, the ethical questions that arise on the Maths Ethics page come largely after the mathematics has been done, when tools developed without explicit ethical intent are applied to populations in ways that cause harm. The statistician who built the algorithm did not design it to discriminate; the discrimination emerged from application. In Natural Science, the ethical question can arise at the point of production itself. The experiment on the men at Tuskegee, the procedure on Henrietta Lacks, the gene editing in He Jiankui's laboratory: each of these is the moment of decision, not a downstream consequence of knowledge produced elsewhere.

In History, the ethical questions centre on how knowledge about the past is produced and whose perspective it represents: who controls the archive, whose testimony counts, which silences are maintained. Natural Science ethics works on a different temporal axis. The question is often whether to create a capacity that does not yet exist and cannot be recalled once it does. He Jiankui's edited germline, disseminated to future generations, is a permanent alteration with no equivalent in historical interpretation.

Human Sciences and Natural Science both use human beings as research subjects, and both have produced landmark cases of consent failure. The Milgram obedience experiments and the Stanford Prison Experiment in psychology, and Tuskegee and the Guatemala syphilis study in public health medicine, raise structurally similar ethical questions about what researchers may do to participants. The difference is in what the human subject represents. Human Sciences study behaviour, meaning, and social life. Natural Science treats the human body as a biological system to be understood, modified, or used as material. The form of reduction matters to the ethics.

In The Arts, ethical questions more often concern what may not be shown or said: censorship, representation, and the uses of creative authority. The question of whether knowledge should be pursued at all arises less directly, because art does not produce the kind of knowledge that creates irreversible capacities. Natural Science raises the question of whether inquiry has limits that cannot be set from inside the practice of inquiry itself. That is the question He Jiankui answered by proceeding, and that Szilard answered by refusing to.

Think further: questions and resources

Leo Szilard held the original patent on the nuclear chain reaction and spent the years after Hiroshima trying to prevent scientists from remaining silent about the weapons they had built. His argument was that producing knowledge creates an obligation to speak about its consequences, even when the application is decided by governments and not by scientists. Does producing knowledge create that obligation - and does it extend to applications the scientist did not intend?
Harriet Washington argues in Medical Apartheid that the exploitation of Black bodies in American medical research was systematic and deliberate across two centuries, not a sequence of isolated failures. If that is true, what does it mean for the knowledge the research produced? Does knowledge generated through injustice carry the injustice with it, or can it be separated from the conditions of its production?
Jacob Bronowski stood at Auschwitz and argued that the willingness to live with uncertainty is the foundation of both science and decency, and that the certainty of those who ran the camp was what made the atrocity possible. But the same openness to doubt that defines scientific method can be exploited to delay action indefinitely. Is doubt always a scientific virtue - or does it become something else when it is weaponised?
Naomi Oreskes and Erik Conway document in Merchants of Doubt how the same small network of scientists who defended tobacco later defended fossil fuels, using manufactured uncertainty to delay regulation in both cases. The strategy worked for decades. What does this suggest about the relationship between scientific consensus and public belief - and where does responsibility for that gap lie?
Jennifer Doudna draws a line between somatic cell editing (changes that do not pass to future generations) and germline editing (changes that do). He Jiankui crossed that line. Doudna's distinction is both ethical and technical: germline changes are heritable and cannot be recalled. Is that line a principled ethical boundary, or does it mark where current scientific confidence happens to run out?
He Jiankui defended his experiment by claiming he was giving the children a gift: heritable protection from HIV. Every defendant at the Nuremberg Doctors' Trial offered a comparable justification - a collective benefit large enough to override the individual's consent. The Nuremberg Code was written precisely to exclude that reasoning. What would need to be true for benefit to the many to justify harm to an unconsenting individual - and who has the authority to make that determination?

Films
For more see my 10 films for the TOK journey page.

🎬 WATCH — Merchants of Doubt (2014)

Robert Kenner

Robert Kenner's documentary follows the network of scientists and public relations operatives named by Oreskes and Conway, showing how manufactured uncertainty was deployed first against tobacco research, then against the science of acid rain, ozone depletion, and climate change. The film names individuals and shows the machinery in operation rather than describing it in the abstract. The clip used in Big Idea 3 opens the film; watching the full documentary extends the evidence considerably and includes interview footage with the operatives themselves, some of whom defend their work. My students can watch the film here.

🎬 WATCH — Don't Look Up (2021)

Adam McKay

Two astronomers discover a comet on a direct collision course with Earth and spend the rest of the film unable to make anyone act on that fact. What they encounter is not simply public irrationality but something more organised: a media environment that turns scientific consensus into entertainment, a political class that reads the data through the lens of electoral calculation, and a tech billionaire who sees the comet as an extraction opportunity. The film covers the same territory Oreskes and Conway document in Merchants of Doubt, but in real time rather than retrospect - the question is not why the tobacco industry succeeded in the 1950s but how the same logic operates when the timeline is immediate and the evidence is unambiguous. My students can watch the film here.

🎬 WATCH — Human Nature (2019)

Adam Bolt

Adam Bolt's documentary traces CRISPR-Cas9 from its origins in bacterial immune systems to its emergence as a precision gene-editing tool, through interviews with Jennifer Doudna, Emmanuelle Charpentier, and the researchers involved in its development. The film was nearly finished when He Jiankui announced the birth of the first gene-edited babies; it pivots to include that announcement as its closing act. It covers the same ground as Big Idea 1 - what scientists owe to the consequences of their discoveries - but from inside the scientific community as it processes what one of its members has just done. My students can watch the film here.

🎬 WATCH — The Immortal Life of Henrietta Lacks (2017)

George C. Wolfe

George C. Wolfe's adaptation of Rebecca Skloot's book follows Skloot (Rose Byrne) and Henrietta's daughter Deborah (Oprah Winfrey) as they piece together what happened to the cells taken without consent from Henrietta in 1951 and what the family knew across the decades of research those cells enabled. The book is in the reader for this page; the film makes the same argument through Deborah's search for her mother - whose body produced one of the most important biological research tools of the twentieth century, and who received nothing. My students can watch the film here.