Skip to content

Guest Post: Is it ethical to use data from Nazi medical experiments?

  • by

Written by Dr Lynn Gillam

Academic Director/ Clinical Ethicist,

Children’s Bioethics Centre at the Royal Children’s Hospital,

and Associate Professor in Health Ethics at the Centre for Health and Society at University of Melbourne

Results of Nazi hypothermia experiments were cited in papers from the 1950s-1980s. Rich Engelbrecht/Flickr, CC BY-NC-SA

 

During World War II, Nazi doctors had unfettered access to human beings they could use in medical experiments in any way they chose. In one way, these experiments were just another form of mass torture and murder so our moral judgement of them is clear.

But they also pose an uncomfortable moral challenge: what if some of the medical experiments yielded scientifically sound data that could be put to good use? Would it be justifiable to use that knowledge?

Using data

It’s tempting to deflect the question by saying the data are useless – that the bad behaviour must have produced bad science, so we don’t even have to think about it. But there is no inevitable link between the two because science is not a moral endeavour. If scientific data is too poor to use, it’s because of poor study design and analysis, not because of the bad moral character of the scientist. And in fact, some of the data from Nazi experiments is scientifically sound enough to be useful.

The hypothermia experiments in which people were immersed in ice water until they became unconscious (and many died), for instance, established the rate of cooling of humans in cold water and provided information about when re-warming might be successful. Data from the Nazi experiments was cited in scientific papers from the 1950s to the 1980s, but with no indication of its nature.

The original source appears as a paper by Leo Alexander, published in Combined Intelligence Objectives Subcommittee Files. This is an unusual type of publication to be mentioned in a scientific journal, and it’s unclear that it comes from the trial of Nazi doctors at Nurmemberg.

In the late 1980s, US researcher Robert Pozos argued the Nazi hypothermia data was critical to improving methods of reviving people rescued from freezing water after boat accidents, but the New England Journal of Medicine rejected his proposal to publish the data openly.

Use of data generated by the Nazis from the deadly phosgene gas experiments has also been considered, and rejected by the US Environmental Protection Agency, even though it could have helped save lives of those accidentally exposed.

A tricky conundrum

So should the results of Nazi experiments ever be taken up and used? A simple utilitarian response would look to the obvious consequences. If good can come to people now and in the future from using the data, then its use is surely justified. After all, no further harm can be done to those who died.

But a more sophisticated utilitarian would think about the indirect and subtle consequences. Perhaps family members of those who were experimented on would be distressed to know the data was being used. And their distress might outweigh the good that could be done. Or perhaps using the data would send the message that the experiments weren’t so bad after all, and even encourage morally blinkered doctors to do in their own unethical experiments.

Of course, these bad consequences could be avoided simply be making sure the data is used in secret, never entering the published academic literature. But recommending deception to solve a moral problem is clearly problematic in itself.

The trouble is that focusing on the consequences – whether good or bad – of using Nazi data, misses an important point: there’s a principle at stake here. Even if some good could come of using the data, it would just not be right to use it. It would somehow deny or downplay the evil of what was done in the experiments that generated them.

This is a common sentiment, but if it is to hold ethical weight we need to be able to spell it out and give it a solid foundation. A little reflection shows that, as a society, we don’t have an absolute objection to deriving some good out of something bad or wrong. Murder victims sometimes become organ donors, for instance, but there is no concern that is inappropriate.

Paying our debt

So how to decide when it’s all right to derive some good from a wrongdoing? I think the answer lies in considering what society owes ethically to the victims of a wrongdoing. The ongoing investigations into institutional child sexual abuse in a number of Western countries have brought this question sharply into focus.

The wrongs done to victims of abuse are over but that’s not the end of the matter. Victims are ethically owed many things: recognition that what was done to them was indeed wrong, a credible indication that the society takes this seriously, an effort to identify, apprehend and punish the perpetrators, and compensation for their ongoing suffering and disadvantage. But beyond this, we have an obligation not to forget, and not to whitewash.

Victims of Nazi medical experiments are owed these same things. If society’s obligations to them have broadly been met through the Nuremberg trials and the ongoing global abhorrence of the awful things done to people in World War II, then it might be ethically possible to use the data if it could lead to some good.

But this must only be done with absolute openness about the source of the data, and clear condemnation of the way it was obtained. Citation of the Nazi hypothermia data in the medical and scientific literature from the 1950s to the 1980s gives no hint at all about of what is being referred to, and so falls ethically short.

Share on

22 Comment on this post

  1. This is a complex issue which is all too easily resolved by mindless ‘simple utilitarian response’. I remember discussing this at the CSEP, University of Manchester . In the audience there were the members of an ethics committee for one of the city’s teaching hospitals (some of them had done the PG ethics and law course at Manchester). Without a slightest moment of reflection they all gave a resounding ’Yes’ to your question. I attempted to raise some caveats they might like to consider, but they were all rejected as being irrational because there was only one correct answer to the question.

    I think your assertion that ‘science is not a moral endeavour’ is not quite correct. If we look at science conducted within oppressive and ideological environments, we not only discover the obvious bad science but, if we scratch a little deeper, we find minor “errors” because researchers – knowingly and unknowingly – distort their findings to fit the paradigm (I do not want to overplay the Kuhn, but the paradigm is in part formed by the social and political environment). Obviously this is a problem for science in all societies, which brings into question any strong claim of it being value free. If, as in the case of these Nazi experiments, the experiments cannot be repeated, it becomes extremely difficult to assess their veracity. It is not impossible, but these epistemic issues cannot be avoided by simple utilitarianism without repeating the same error. (Thought experiments that postulate that the “knowledge” is objective and value free only obfuscate the problem.)

    You say ’If society’s obligations to them have broadly been met through the Nuremberg trials and the ongoing global abhorrence of the awful things done to people in World War II, then it might be ethically possible to use the data if it could lead to some good.’ This is a big ‘If’ given that thousands of Nazi scientist, many of whom were directly implicated in the deaths of Jews and others, were spirited away by the Allies (USA’s Operation Paperclip and the USSR’s Operation Osoaviakhim redeployed most of them); there was no real assertive effort to de-Nazify German and there was little or no compensation paid to the victims of slave labour.

    That said, I agree there may be occasions were some Nazi science research can be used so long as it is, as you say, used openly with citations and condemnation. The experiments conducted at Dachau are perhaps in a league of their own and should not have been used. (As far as I am aware there has been no attempt to seriously involve the survivors of these experiments in this discussion. There is still time but not long.) It is, as you say, a tricky conundrum which may require us to reject information extracted under torture.

  2. The use of a non-utilitarian framework for adjudicating this issue is compelling, but I feel it may be too permissive – allowing the use of the data as long as it’s accompanied by a condemnatory statement.

    I think we can draw an analogy to the legal principle of Exclusion, or the related Fruit of the Poisonous Tree doctrine – where evidence improperly/unconstitutionally obtained is inadmissible in court. In the US, at least, this is actually justified on somewhat utilitarian terms, for the sake of deterring future abuses of legal rights, rather than a sense of debt to the victims. But it is certainly a rule utilitarian justification, insofar as it is not generally sensitive to public benefit that might be gotten from inclusion of the tainted evidence.

    The analogous rule for the Nazi research would be to exclude it from citation and study, even if one is open about and condemnatory towards the source. The idea would be to provide further deterrence against abuses in medical research. What’s interesting to me, though, is that this may exclude citation of a lot of unethical research that falls well short of Nazi inhumanity. This would probably end up excluding from citation and study the classic Stanford Prison Experiment, for instance, given numerous lapses in subject protection. But the field will survive – there are enough more ethically-sound corroborations of the basic findings so that future research and insights will not be significantly diminished.

    1. I don’t think that’s necessarily analogous. That kind of evidence is inadmissible because it is being specifically used against the party whose rights were violated in the first place to obtain it. Data acquired by unethical Nazi experiments is not being used to further violate the rights of the research subjects.

      The data already exists. Let it exist in the open as a reminder of the atrocities that were committed. As a reminder that “good science” can be a tool of oppression so that we don’t repeat our mistakes.

      1. The analogy’s not perfect, to be sure. Still, I believe that evidence against A would be inadmissible even if it was obtained in an illegal search of person B – the justification is not rectification for rights violation, but deterrence against rights-violations more generally.

        But as Eric Black notes below, maybe there’s not much need for this sort of deterrence – current oversight regimes, including IRB review, are robust enough to prevent atrocities – as well as various ethical violations that fall short of atrocities. So maybe deterrence is insufficient justification for prohibition on its own – but combined with the argument of moral offense to victims/families in the article and reflected in some comments, I think there is probably sufficient reason to maintain a prohibition.

  3. Mr. Schaefer proposes an interesting analogy. Although using genuine scientific knowledge (assuming we have any from such experiments) while condemning and prosecuting the experiments- to a degree- inoculates the results from any benefits to those conducting said experiments. Whereas in the legal example, cops and prosecutors benefit directly from corrupt tactics. Unsatisfactory (compromise) solution, but I’d argue using knowledge (however obtained) while having strong ethical, political and economic sanctions in place to ensure those who utilize unethical research methods are universally condemned and excluded from the community of scientists (ie, strong deterrent). If you go far beyond basic human rights you could exclude a lot of pure science from bad regimes – I know this argument is fundamentally utitarian with a baseline of natural rights but it’s the best I have.

  4. Surely it would be possible to argue the opposite — that if we don’t use the data, the torture and death of the victims become purposeless. If I look at it personally, I know what I’d choose if my options were to be tortured to death with no result beyond the satisfaction of my torturers, or with the result of eventually saving others. This would be directly analogous to the organ donation.

    We could condemned the torturers and simultaneously honored the vistims when we cited the results.

    1. I agree and I know that there are survivors of the camps who would like to think that some good came from of their horror. On the other hand, rejecting information extracted under torture might prevent others from justifying research that uses torture or indeed harmful non-consensual methods in the future.

      I think Lynn’s question needs addressing while there are still survivors of Nazi science.

  5. Ethical or pragmatic ?
    A sense of high irony in bitter ,utterly bitter , nauseous perspective.
    Do you really think never was made use of findings?
    Morally objections?
    The ethics of brutality ?

    I am flabbergasted by calling it an ethical question.
    Ought to remember once Wiesenthal said :”…military defeated,not the ideas….
    “[e.g. Euthanasia ;which characteristics a child should have;the idea of race and so on].

  6. I lack the expertise to answer the question directly. But it strikes me that while the Nazi experiments were abhorrent, science has often progressed by harnessing the evil of war and oppression. I suppose the question is whether the person observing and recording the consequences of such evil was complicit in the acts themselves, or whether they were simply a reliable witness of someone else’s acts of evil. If the latter, I think the learnings are sufficiently far removed from the evil to be both practically and morally justifiable. The problem with the Nazi experiments is that it is difficult to disentangle the findings from the evil motives and practices of those recording them. The fact that the experiments cannot be replicated means that, if we are to cite the findings, we are forced to trust the scientific integrity of people who plainly lacked integrity. However, if there are findings which can be separately corroborated by innocent observers at the time, then I think we can perhaps credit the innocent observer with the finding (in the same way that we would have no qualms about citing field studies from Hiroshima or Nagasaki). I’m not sure whether many of the Nazi experiments meet this standard, though?

  7. Before considering ethics, I’d find it helpful to determine whether the data is scientifically valid to begin with (as others have mentioned above). What pedigree do we need to achieve? To me it seems contextual – in the hypothermia case presented here there may be no good alternative to achieving warming/cooling rates in humans. In this context, even a flawed scientific study (for the reasons mentioned by others here) may provide benefit as a “redundant” study. Here we have no method of validating the data, but it’s enough to provide an initial guess towards, say, bringing a person out of hypothermia. I’d argue that any study, as long as its shortcomings and assumptions are understood, can provide benefit.

    To use the data, there seems to be agreement that full disclosure of its origins must be made, or it would inherently lack in scientific integrity. But even as we condemn the authors, we are still prizing the study for its data. Weightings of both ethics and scientific advantages, even though subjective, must be made. As victims and their immediate families pass on to new generations, is weight reduced from the ethics side? Is data from previous wars any more ethical to use? Again, it becomes contextual and we’ll need to estimate the societal benefits of the data – I’d like to think that no generalizations can be made.

  8. In the 1980s in a medical school in Manitoba, my mother was presented with data from Nazi experiments. She, a daughter of a holocaust survivor whose sister was experimented on by Mengele to the point she couldn’t have children, was disgusted. The experiments done by the Nazis were not science and should not be viewed as such. Murder victims who donate organs have the chance to chose before they die if they want to donate their organs, or their family can choose for them after their death. That is very far away from what was done in the holocaust. There was entirely no choice.
    Nazi experiments were not just completely unethical by our standards of human subject experimentation now, the nazi experiments were war crimes that progressed and assisted in their genocidal goals and should not be considered to be anything close to what we consider science.
    Presenting this question as it is only a matter of how to interpret ethical rules and paradigms is so far removed from the atrocities of these experiments. Sitting in an office on a computer debating whether inhumane “science” should be used to create the most minuscule amount of good is not looking at the bigger picture, on the level of basic humanity, of what was done to people and their families and their children in this war. It is not a posthumous honor to use this data but a slap in the face to those who suffered and continue to suffer generations later because of it.
    I hope the scientific journals continue to reject citations of nazi data and any other data collected on populations who had entirely no choice in the matter. These experiments were torture.

    1. Consider this, if a loved one of yours could directly benefit in a life or death situation from the data obtained during the experimentation, would your feelings change, or would you be prepared to watch a loved one suffer knowing a cure was there, but was gathered unethically?

    2. Obviously many experiments were atrocious.
      And we should honor those people by speaking of the things that happened, and making sure they do not happen again.
      That is how you honor them.

      You yourself are not looking at the bigger picture, and only looking at it on a personal level.
      If you don’t want any data or technology by which anyone ever suffered, then I’m afraid you’re not allowed to use much technology.

      Bad things happened a long while ago – remember that.
      Then try to make it in something good.
      For only by progressing will we have better living standards for everyone, and thus making sure such things don’t need to happen again.

  9. (For the sake of argument I’ll assume Data leads to technology or other form of advancement.)

    Why do people always associate nazi’s with pure evil? In the end Hitler, and nazis just wanted was best for them.
    Obviously I don’t sympathise, or condone actions warcrimes and other a that were done.atrocity.
    But simply limiting these actions to “the nazi’s” is a limited viewpoint.

    Thinking all technology that was gained from WWII is ethical is just naive. What about Hiroshima and Nagasaki? No data was gathered from those things I’m sure … .
    Surely it’s unethical to also use the technology that was gained from the data as well?

    Or how do you feel about data and technology that was gathered in wars before that? As we all know wars have Always been a great age for technology, but also of ethical low grounds. Do you believe wars in the 1800’s didn’t have any unethical gathered data & technology?
    Do you dismiss those as well?

    Or how about data and technology that was gather in a way that we currently view as unethical, but was ethical at the time?
    Experiments on slaves for example? (which is also how nazi’s saw some populations)

    Your question should be – is it ethical to use any data from any experiment that was attained by any experiment we currently think of as unethical.

    I’m afraid we wouldn’t have much technology if we didn’t. Isn’t it technology that helps us improve our lifestyle, and with that our ability to do ethical things?

    As for the question – wouldn’t we ‘encourage’ other people to do unethical things by accepting the data?
    Do you believe that the people who gathered the data saw it as unethical? Of course not.
    What is ethical and what is not is subjective. Knowing your data will be used won’t change your idea on which experiments are ethical or not.

    1. Why do people always associate nazi’s with pure evil?

      Because of the economic basis of the Nazi regime where the undesired people were calculated to work themselves to death and, when that didn’t go fast enough, the extermination camps.

      Thinking all technology that was gained from WWII is ethical is just naive. What about Hiroshima and Nagasaki?

      Nobody’s naive here. If you want to call the Nagasaki bombing unnecessary and a war crime, I won’t dispute it. I’m also well aware that none of the allies had any problem employing morally reprehensible figures for their engineering qualities, like the SS-er Wernher von Braun.

      Your question should be – is it ethical to use any data from any experiment that was attained by any experiment we currently think of as unethical.

      What sets science apart from activities like psychology and sociology is the possibility for an independent scientist to replicate an experiment. Additionally, modern science has ethical guidelines, such as how to deal with test subjects, privacy, statistics, funding, etc, which would make it impossible to replicate the experiments done by the Nazi’s. Using data you can’t replicate is bad science.

      In practice we deal with this flexibly. The hypothermia experiments have been used in citations, see the article, although perhaps that was before universities and magazines demanded adherence to ethical guidelines. It’s fair to say that whatever Mengele c.s. were doing in Auschwitz was nothing more than torture in a labcoat, we can safely use that as an example of what “pure evil” is.

  10. Owen Schaefer’s comparison to the doctrine of admissibility of evidence is a very interesting one. But there are a couple of reasons why it might break down in this case.

    These two are related but distinct, I think:
    1) Shared concern with the past crimes: the admissibility of evidence doctrine applies to a group of people who operate within a specific system (current US law, for example), who, because of the nature of that system, may be motivated to obtain evidence improperly; and who share a common interest in not doing so. The Nazis, by contrast, did not share any of our concerns about this kind of experimentation. They were outside of any ethical or legal system we establish, and retroactively incorporating them into our ethical system doesn’t seem to do anything except in some bizarre way to tie us ethically to them.
    2) Shared commitment not to repeat the crimes: There doesn’t seem to be any immediate danger of anyone in the US recreating Nazi experiments; and if the Nazis returned, they would not think themselves bound by any rules which we make up.

  11. Mankind has been Using Nazi knowledge since the end of WWII– and it has been using it (mainly) for good… without the V2 programme there would not have been moon landings, without Hitler wanting to spread his speeches, we would not have had the magnetic tapes, “flying wings” … there are dozens of technologies we have harnessed from them, and all of those came with a high cost of human life. even if V2-s were not developed through experimenting on humans, thousands of forced laborers died making them, same thing goes for jet engines , and BASF factories— should we have discarded those and their obvious advantages because they came from the NAzis— as much as I am concerned the medical data falls into the same category..
    TRue- the experiments were horrendous, and many people suffered horribly — but, if there is data that can help save lives, then I believe we should use it to do so
    actually I would go even further and say that we owe it to the victims to use the data if it will bring overall good… Although it will never bring the victims back, or make their suffering, and the Nazi experiments less evil, at least it will not mean that all of their suffering was in vain. If we discard the results of those experiments, we are just allowing those victims f the Nazis to die for nothing… By using the results to do good– their deaths will be honored

    Do not mistake me for someone agreeing with the Nazi experimets- all I am saying is that we can not change what happened, but if there is something good to come out of it- why not take it

  12. Educate yourself on the history of bioethics and you will see that the Nazi experiments are not the only reprehensible experiments that have taken place in the name of Science. Prior to the war, doctors, governments and scientists were conducting research on people without any consent whatsoever, without even telling the people they were being experimented on.

    For example, look at the experiments in Tuskegee.
    Doctors wanted to know how syphilis spread in a population. They told the populace they were getting treatment and gave them sugar pills (placebo) instead. Many people died from lack of treatment while trusting the doctors to help them, and the doctors monitored and recorded the symptoms and deaths that resulted as their “experimental data”.

    There were also experiments done on mentally retarded children using radioactive isotopes (children were made to ingest them) to discover the effects of radioactivity on the human body. I do not remember the particulars off the top of my head but I’m sure they can be searched online.

    The data is not invalidated from having come from a “morally bereft” source. If it was conducted with proper scientific controls it is absolutely valid.

    Consider this: a human is just another animal. We conduct experiments on animals all the time. Many people (including me) consider this to be disgusting and completely immoral, yet the data is not considered to be invalid because it was obtained immorally (even though experimentation on animals is no substitute for experimentation on humans, if scientists could ignore morality I’m sure they’d agree that results obtained from drug experimentation on humans instead of an animal substitute is far more scientifically valid – for example, cats die when fed paracetamol, had we tested paracetamol on cats we would have rejected it as a toxic non-useful compound, test it on a human and you can see it’s perfectly viable).

    I will not say that we shouldn’t use the data we have collected from previous experiments on animals, even while I protest that continued animal experiments should NOT take place. If we did throw away the data, then those animals have died FOR NO REASON. It makes their sacrifice totally in vain.

    Similarly if there is useful data in the Nazi experiments, I think it would be good to keep and use that data, because people died or were harmed in obtaining it – so we should not let their sacrifice go to waste if there is something useful in the data (that could potentially help someone or something in future).

    However I would still like ALL non consensual experiments on ANY higher vertebrate stopped. Anything that can experience suffering should not be experimented on. Full stop.

  13. Educate yourself on the history of bioethics and you will see that the Nazi experiments are not the only reprehensible experiments that have taken place in the name of Science. Prior to the war, doctors, governments and scientists were conducting research on people without any consent whatsoever, without even telling the people they were being experimented on.

    For example, look at the experiments in Tuskegee.
    Doctors wanted to know how syphilis spread in a population. They told the populace they were getting treatment and gave them sugar pills (placebo) instead. Many people died from lack of treatment while trusting the doctors to help them, and the doctors monitored and recorded the symptoms and deaths that resulted as their “experimental data”.

    There were also experiments done on mentally retarded children using radioactive isotopes (children were made to ingest them) to discover the effects of radioactivity on the human body. I do not remember the particulars off the top of my head but I’m sure they can be searched online.

    The data is not invalidated from having come from a “morally bereft” source. If it was conducted with proper scientific controls it is absolutely valid.

    Consider this: a human is just another animal. We conduct experiments on animals all the time. Many people (including me) consider this to be disgusting and completely immoral, yet the data is not considered to be invalid because it was obtained immorally (even though experimentation on animals is no substitute for experimentation on humans, if scientists could ignore morality I’m sure they’d agree that results obtained from drug experimentation on humans instead of an animal substitute is far more scientifically valid – for example, cats die when fed paracetamol, had we tested paracetamol on cats we would have rejected it as a toxic non-useful compound, test it on a human and you can see it’s perfectly viable).

    I will not say that we shouldn’t use the data we have collected from previous experiments on animals, even while I protest that continued animal experiments should NOT take place. If we did throw away the data, then those animals have died FOR NO REASON. It makes their sacrifice totally in vain.

    Similarly if there is useful data in the Nazi experiments, I think it would be good to keep and use that data, because people died or were harmed in obtaining it – so we should not let their sacrifice go to waste if there is something useful in the data (that could potentially help someone or something in future).

    However I would still like ALL non consensual potentially harmful experiments on ANY higher vertebrate stopped. Anything that can experience suffering should not be experimented on. Full stop.

  14. This article suggests that obligations to victims are the crucial test of “when it’s all right to derive some good from a wrongdoing.”

    This principle appears sound, but the article only focuses on the direct victims of experimentation. The interests of victims generated by the decision to use the data openly must also be considered.

    Moral results cannot themselves be immoral.

    If the clear and open usage of Nazi data will incentivize unethical experimentation, then we must also consider our obligations to the potential and probable victims of this unethical experimentation.

    These calculations will likely be difficult and perhaps even negligible in their ultimate moral significance. We may agree that this open declaration will have little to no effect in generating more victims of unethical experimentation.

    Nonetheless, the ethical theory must incorporate the entirety of our obligations to victims: not just direct victims, but the indirect ones caused by our decision to clarify the sources of our data.

Comments are closed.