The American Medical Association's (AMA's) Council on Ethical and Judicial Affairs' (CEJA's) new “Guidelines to Prevent the Malevolent Use of Biomedical Research” are both timely and appropriate. These guidelines are a product of the increasing realization of the “dual use” potential of life science discoveries. Although biomedical research usually aims at the development of new medicines, vaccines, diagnostics, and so on, the very same discoveries that could benefit humankind in these ways also often have implications for the development of biological weapons.¹

The CEJA Guidelines draw attention to this fact and hold that physician–researchers have responsibilities regarding the uses to which their discoveries are put. Medical researchers should assess the likely social benefits and harms of their work and avoid projects where the latter outweigh the former. Knowledge and the advancement of science should not be the only aims of scientists; “commitment to public welfare and safety” are also essential.² Toward the aim of preventing the malevolent use of research discoveries, the CEJA states that “regulatory oversight” of research (with an eye to potential harmful uses of potential discoveries), increased ethics education of physician researchers, and sometimes perhaps even censorship of research findings, are warranted.³

This CEJA should be praised for broadening the focus of research ethics guidelines, which have traditionally been primarily concerned with the protection of human participants in human experimentation,⁴

and for (implicitly) expanding dialogue on ethical implications of the new genetics. Whereas initial discussion of ethical, legal, and social implications of advances in genetics focused on the worry that dangerous material might result from recombinant research and escape into the environment, more recent bioethics discussion has focused on issues of genetic testing, discrimination, eugenics, cloning, stem cell research, and (to a lesser extent) DNA fingerprinting and the patenting of genetic sequences. Though comparatively neglected by ethics discourse, in the meanwhile, biological weapons development may turn out to be the most serious consequence of the genetic revolution.⁵ According to the CIA, for example,

The kind of danger that the CEJA and CIA are concerned about is illustrated by the recent publication of two controversial studies (alluded to in the CEJA Guidelines). In one case, Australian researchers aiming to find a way to reduce mouse fertility as a means of pest control accidentally discovered that insertion of the IL-4 gene into the mousepox virus genome resulted in a superstrain of mousepox that killed mice that were naturally resistant to, and also mice that had been vaccinated against, the disease. The scientists proceeded to publish their findings—along with a description of their materials and methods—in 2001.⁷

In a second case, American researchers sponsored by the U.S. Department of Defense synthesized a “live” polio virus from scratch by stringing together commercially available strands of DNA (purchased over the Internet) in accordance with the map of the polio virus (RNA) genome (published on the Internet). The addition of protein resulted in a live virus that paralyzed mice. In 2002 they too published their findings, materials, and methods.⁸

Both of these studies have implications for smallpox—a disease that commonly tops lists of feared biological weapons agents. The polio study, for example, reveals that it might be possible to produce the smallpox virus through similar procedures. Although the smallpox virus is much larger than polio, the technical feasibility of artificially synthesizing smallpox is perhaps dubious. More fearsome is the possibility that the straightforward technique used on mousepox could allow the genetic engineering of vaccine-resistant smallpox. As there is no treatment for smallpox, vaccine is our only defense against this disease, which is believed to have killed more humans than any other infectious disease in history. Because routine vaccination ended more than 20 years ago, the world population now largely lacks immunity to smallpox, and modeling has shown that a smallpox attack could cause the devastation of nuclear attack(s).

Critics claim that neither of these discoveries should have been published. Their complaint is that publication of such studies both alerts would-be bioterrorists of possibilities and provides them with explicit instructions for producing potential weapons of mass destruction. The scientists and editors involved, however, claim that publication was warranted given the importance of alerting the scientific community to the kind of things we may need to prepare to defend ourselves against. These are clear examples of “dual use” discoveries: the same knowledge that may be harmfully used in weapons production has benefits with regard to biodefense.

Whether or not the mousepox and polio studies should have been conducted and/or published, the CEJA is right to draw the medical community's conscience to the potential dangers of both research and publication. It would be wrong to think that scientists should only be concerned with the generation of knowledge—which is often assumed to be inherently valuable for its own sake or, at worst, neutral. Because the misuse of knowledge would not be possible without its generation and dissemination in the first place, those who generate and disseminate potentially dangerous knowledge are ethically implicated in any misuse that occurs, especially when the harmful application of the knowledge in question is foreseeable.

This raises a difficult question left unanswered by the CEJA Guidelines. What should the process of censorship be? Given the importance of openness for scientific progress and the importance of transparency for public trust in science, the CEJA sounds generally (though not entirely) resistant to even self-censorship. For similar reasons, the U.S. National Research Council has explicitly argued against governmental censorship of “sensitive” information resulting from life science research, advocating “voluntary self-governance” of the scientific community as the preferred alternative.⁹

Though scientists should be encouraged to refrain from research with implications for weapons development and to voluntarily limit dissemination of research findings in cases where likely harms outweigh benefits, it is doubtful that the scientific community should be relied upon to regulate itself in such matters. This is simply because scientists will not always, contrary to what is said by the CEJA, be “in the best position to assess the potential for and the ramifications of misapplication of their research.”¹⁰

In the case of the mousepox study, for example, the danger of publication is largely a function of the likelihood that there has been proliferation of the smallpox virus from the Soviet biological weapons program (or other secret sources), insofar as aspiring bioterrorists would need to have access to the smallpox virus in order to apply the mousepox technique to it. The true danger of publishing the mousepox study, therefore, depends on information that the scientific community (like the general public) is systematically denied access to. Details about the likelihood of smallpox proliferation are classified.¹¹ Assuming they lacked security clearance, the editors and authors of the mousepox study would not have been able to accurately assess the danger of publication. Given that the government will sometimes be in the best position to judge the danger of publication, censorship by government may sometimes be justified.