Limits of Dual Use

In “The Limits of Dual Use” (Issues, Summer 2018), Tara Mahfoud, Christine Aicardi, Saheli Datta, and Nikolas Rose argue that a “more nuanced” framework than dual use is needed to govern emerging scientific trends. Their concern is valid; their argument, however, obscures the nuances in the debate so far.

An important distinction arises between what people meant when they talked about dual use before 2001, and after. Prior to 2001, the term was used regarding technologies that crossed the civil/military divide, commonly in the context of export control agreements. After 2001, the benign/malign sense of dual use arose, when researchers in Australia inadvertently created an orthopox virus that possessed 100% lethality.

These two senses do not overlap in obvious ways. National Academies reports in 2004 and 2017 rejected export controls as a way to control dual-use research in the life sciences, such as the now-famous Australian study that gave rise to a lethal virus that causes mousepox. Whereas the malign uses of a 100% lethal poxvirus—such as smallpox, which killed half a billion people in the twentieth century—are clear, the military uses are less so. Biological weapons, especially highly lethal ones, are very difficult to control once released, and make for an unpredictable weapon. To an apocalyptic cult, such as the Aum Shinrikyo cult that committed the sarin attacks in 1995 on the Tokyo subway (and two attempted anthrax attacks besides), that might be acceptable.

In the same way, civilian uses do not clearly overlap with benign uses. Arguably, some kinds of weapons are permissible in national defense: only the most ardent pacifist would reject weapons altogether. Moreover, military uses of, say, psychoactive drugs to treat post-traumatic stress disorder—whether caused by combat, sexual assault, or something else—are surely beneficial if they are therapeutic, whether or not they are used in a military theater or context.

The old sense of dual use is less moral, and more strategic. The newer sense is the reverse. Taking care to distinguish between these uses allows us to better navigate the problem.

Strong institutions are an important component of regulating dual-use research, in any sense of the term. In neuroscience, in particular, institutions for regulating its potential applications are weak. Facebook struggled to decide whether or not to shut down InfoWars, a site so egregiously harmful that it perpetuates conspiracy theories such as the one claiming that the young victims of the Sandy Hook school shooting were actors, and that the entire shooting was a hoax. Without a strong set of institutions regulating propaganda, it doesn’t matter if new forms of neuropolitics or neuromarketing or neuropropaganda are used. As the philosopher Neil Levy has noted, manipulation is manipulation whether it is “neuro” or not.

In the military and intelligence spheres, a similar argument can be made. Neuroscience is less likely to lead to a resurgence of torture if the United States would enact strong legislation to reaffirm its commitment to the provisions of the Geneva Convention, famously limited by President George W. Bush through Executive Order 13440. The abuses committed under this order were made possible by weakening the nation’s institutions. Affirming the values that lead us to reject torture requires us to repair those institutions.

Finally, military research in neuroscience should be made as open as possible. In this case, a lot of the fear of future misuse arises in the context of a lack of information about what the military has developed, or hopes to develop. Sunlight can help fix this problem.

Jonathan Moreno

I agree with Tara Mahfoud and her coauthors that there are multiple ways in which benignly intended life science research could be misused, and that there are problems with the way in which the concentration on dual use of concern has distorted debate so that other less immediate possibilities of dual-use applications have tended to be neglected. So I do not see any problem in trying to be more precise about misuse in the political, security, and intelligence domains as the authors suggest. But in the military domain, particularly in regard to chemical and biological weapons, I suspect that dual use will remain as the best terminology for talking about the problem, as it has been regularly used for so long that to try to change the term would only cause unnecessary confusion.

I also disagree with the authors’ statement that the Biological Weapons Convention and the Chemical Weapons Convention five-year review conferences “fail to capture the rapidly evolving advances in the neurosciences.” In regard to the Chemical Weapons Convention, and its Article II.9(d) discussed by the authors, the scientific advisory board of the Organisation for the Prohibition of Chemical Weapons has had a settled view on the problem of so-called incapacitating chemical weapons for many years. As its report for the fourth review conference set for late 2018 noted, the board “maintains the view that the technical discussion on the potential use of toxic chemicals for law enforcement purposes has been exhaustive: the term ‘non-lethal’ is inappropriate when referring to chemicals intended for use as incapacitants … and that the Secretariat should commence preparations for verification activities that could be required during an investigation of alleged use….”

Even in regard to the Biological Weapons Convention, it should be noted that the intersessional process that occurs between the review conferences will in 2018 be concentrating on the dual-use issues caused by the recent advances in genome editing.

However, I do think the authors are correct in their view that we should take a broad perspective on the problem of the potential misuse of advances in the life sciences. In our own work on arms control, we have therefore adopted a holistic strategy that involves three phases. The first involves an examination of the nature of the weapons and weapons-related technology under review. Then there is an exploration of the full range of potentially applicable control mechanisms. And finally comes the development of a comprehensive strategy to improve existing mechanisms, or introduce additional mechanisms, for the effective regulation of the weapon or weapons-related technology of concern.

Malcolm Dando