Is there ever a reason for unethical human experiments?

Let us posit a situation that faces the government of a developed nation. In this situation there are ongoing threats to national security from significant external forces leading to a pressing need to develop highly effective weapons systems for both offensive and defensive purposes. In order to establish effectiveness it is necessary to test such weapons, and this has to be done in a number of ways, depending on the type of weapon. Essentially what these test boil down to are an analysis of the impact of a given weapon on multiple aspects including infrastructure and buildings, the environment and of course on human targets, the primary targets for most weapons. In some cases it is possible to use computer models, or simulations, particularly for infrastructure and environmental impacts, and potentially to use simulacrum to establish the effect on human targets, typically using something like ballistic gel, or animal carcasses for, say a kinetic weapon like a bullet. The question posed by this scenario however is whether there are some types of weapons, or some situations where there is a need to carry out tests using live human subjects?

Suppose that the new weapon was biological in nature. A bacterial or viral agent designed to target enemy forces and incapacitate or kill them. Something of this nature would be very difficult to test on a simulacrum. An animal carcass of ballistic material can not contract a disease. The impact of a biological agent, its efficacy, the time of incubation, the disease progress and spread can not be accurately assessed. What about the use of radiological weapons such as depleted uranium or nuclear weapons? We have an understanding of how radiation affects human beings from research carried out following nuclear power plant accidents, most famously the Chernobyl disaster but this is not necessarily the same as the effects of a deliberate nuclear attack, so how valid is this data? There is at least the possibility that there would be a need to assess such weapons on human test subjects. This raises another serious question and it is one that cuts to the heart of the title of this article.

There are two ways in which human test subjects can be engaged, consensually or non-consensually. In standard human experiments in the field of medical research the idea of using non-consenting test subjects would be completely anathema, but is it ever acceptable in the interest of national security? Consider that one of the requirements of a weapons development program is to keep it secret from enemies both existing and potential. Consider also that there is a potential discrepancy between results between subjects who are aware of the test and those who are not. Given these two thoughts it could be argued that in order to maintain security and in order to give accurate test results there could potentially be a point at which there might be a requirement to engage in experimentation with human test subjects who were not aware that they were being tested. This is, of course completely unethical but there are a number of philosophical proponents of the concept of the greater good, and following this thought pattern, the damage caused to a small number of individuals could be weighed against the benefit of military success in terms of defending a far larger national population.

Of course, there are also arguments to suggest that no matter what the requirements, there can never be a reason for unethical behaviour at this level, and that it is tantamount to torture, but there is evidence that not only has this type of research happened, but that it is still happening in several highly developed countries around the World. Does this make it acceptable? Personally I would say no, but as a pragmatist, I have to acknowledge that I can understand the case for, even though I am personally against.