The relationship between national security and scientific research has received considerable attention in recent years. As part of this, particularly in North America and Europe, questions are being raised regarding whether the knowledge and techniques generated through fundamental and applied life science research might facilitate the production of bioweapons and therefore whether controls should be placed on what gets done, how, and whether information is widely circulated. Such concerns, in some part, informed the 2005 meetings in the inter-review conference process of the Biological and Toxin Weapons Convention examining “the content, promulgation and adoption of codes of conduct for scientists.” The final report of the 2005 meeting States Parties stated that:
codes of conduct should require and enable actors to have a clear understanding of the content, purpose and reasonably foreseeable consequences of their activities, and of the need to abide by the obligations in the Convention.
In response to this emerging discussion, this paper elaborates a pragmatic empirical research agenda for engaging with practicing scientists regarding what is to be done. It is meant to provide further background the presentation by Rappert at the 10th PIIC Beijing Seminar on International Security (25-28 September 2006, Xiamen).
In over 50 seminars Rappert and Malcolm Dando (University of Bradford) have engaged some 1300 scientists in discussions regarding what they thought of the security concerns being debated about the potential malign use of their work. This paper focuses on certain methodological dimensions of those seminars. Various dilemmas, decisions, and difficulties of discussing the dual use status of life science research are recounted with a view to reflecting on the unavoidable choices made in efforts to promote a questioning of the practices of research. This research suggests that current discussions within the life sciences are such that careful attention and sustained commitment to education and awareness raising should be considered internationally.
The Life Sciences, Biosecurity, and Dual-Use Research:
Further Details on a Proposed Method for Engaging with Scientists
Introduction In recent years, the continuing high public profile of ethical, social, and political issues associated with scientific research has renewed attention to long standing questions about its place in society. Particularly after the events of 9-11 and the anthrax attacks that followed in the US, the relationship between national security and research has been one of those topics that has received significant attention in many countries (Alberts, 2002). Perhaps in a manner unprecedented, as part of this the life sciences have come under scrutiny regarding their security implications (Marburger, 2003). Not only have concerns been voiced about the possibility of diverting dangerous pathogens and toxins from laboratories, in manner analogous to that in nuclear sciences or cryptography, questions are being raised whether the knowledge and techniques generated through research might facilitate the production of bioweapons and therefore whether controls should be placed on what gets done, how, and whether it is widely circulated.
As with the emergence of other areas of public controversy about scientific practice such as the safety of laboratories, the participation of human subjects, the retention of human organs, and the use of animals in experimentation, the current security focus potentially poses considerable challenges to established practices and preoccupations. This in turn means that attention to matters of security and research not only raises concerns about the public understanding of science, but scientists’ understanding of their own activities.
This paper specifies a highly pragmatic strategy employed for engaging with life science researchers regarding the ‘dual use’ implications of their work. It is meant to provide further background the presentation by Rappert at the 10th PIIC Beijing Seminar on International Security (25-28 September 2006, Xiamen). During 2004-5, the author and Malcolm Dando (University of Bradford, UK) undertook numerous research seminar workshops with British researchers. In 2005-6 the basic format seminar format was extended to the Netherlands, the US, Finland, and South Africa. These seminar workshops were conducted as interactive sessions that combined the goal of awareness raising with data collection. In their structure and rationale, they most closely approximated ‘focus groups’, yet there were significant deviations from typical procedures employed for the latter. Our work entailed undertaking seminars as part of existing university departmental seminar series and employing a problem-orientated and dialogic methodology that transformed over time to achieve a greater mutual understanding of the issues associated with bioweapons and life science research. In discussing the preparation for, the planning of, and the conducting of these workshops, this paper aims to propose a strategy of engagement and learning relevant for other areas of emerging controversy. Various dilemmas, decisions, and difficulties are recounted with a view to reflecting on the unavoidable choices made in efforts to promote a questioning of the practices of research.
Dual Use Life Science Research as an (re-)Emerging Social Problem With the heightened attention to bioweapon threats in the aftermath of 9-11, security-related debate has taken place regarding topics such as the protection of human subjects in experimentation (Trotter, 2003), the public health response to bioattacks (Kipnis, 2003), the procedures for regulating experimental drugs (Shamoo, 2003) and the physical containment of dangerous pathogens (Epstein, 2001). While the possible contribution of the advancement of the life sciences for enabling novel forms of biological weapons has been a matter of discussion in the past, today there is unprecedented attention to whether the knowledge and techniques generated in fields such as immunology, molecular biology, virology, toxicology, and molecular genetics could ease the development of weaponry. Possibilities envisioned include manipulating viruses and bacteria to make them resistant to anti-virals and antibiotics, modifying the virulence and pathogenicity of known bioagents, rendering the detection of bioattacks more difficult, enhancing the capacities for the dissemination of agents, and reducing the effectiveness of the body’s defense system.
Just what the identification of such possibilities should mean for responsive measures though has been less than straightforward. Shortly after 9-11, the Natural Research Council of the US National Academies established a committee chaired by Professor Gerald Fink of the Whitehead Institute for Biomedical Research to examine possible changes to research practices. Fink succinctly summarized many of the commonly identified binds associated with controlling life science research in his preface to the committees’ report Biotechnology Research in an Age of Terrorism: Confronting the Dual Use Dilemma (NRC, 2003, p. vii):
…[A]lmost all biotechnology in the service of human health can be subverted for misuse by hostile individuals or nations. The major vehicles of bioterrorism, at least in the near term, are likely to be based on materials and techniques that are available throughout the world and are easily acquired. Most importantly, a critical element of our defense against bioterrorism is the accelerated development of biotechnology to advance our ability to detect and cure disease. Since the development of biotechnology is facilitated by the sharing of ideas and materials, open communication offers the best security against bioterrorism. The tension between the spread of technologies that protect us and the spread of technologies that threaten us is the crux of the dilemma.
As argued herein, attempts to curtail research out of fears about its eventual use may well prove counterproductive because they threaten the open communication of science.
Three experiments have come to epitomize the tensions identified by Professor Fink. One, the insertion of the interleukin-4 gene into the mousepox virus by Australian researchers in early 2001 to find an infectious contraceptive to combat mice plagues. With the high morality rates achieved for immunized and non-immunized mice with the over expressed IL-4, this experiment (unexpectedly for the research team) suggested a technique for enhancing the lethality of other pox viruses (e.g., smallpox). Second, the 2002 announcement of the successful artificial chemical synthesis of poliovirus that brought to the fore a way to create other viruses from scratch. Third, the comparison of variola major and vaccinia viruses published in 2002 that indicated how the vaccinia virus used to immunize against smallpox might be made more lethal. While recognizing the potential malign applications of such experiments, many have defended their undertaking and publication because of their value in warning about impending capabilities or because of their importance in elucidating fundamental biological mechanisms. Indeed, the initial results were extended in follow-on experiments such as the adding of the IL-4 gene to rabbitpox and cowpox and its refined insertion into mousepox (BMA, 2004).
The initial dilemmas associated with assessing the appropriateness of conducting and communicating possible ‘contentious research’ (Epstein, 2001) are further complicated by uncertainty and disagreement over the severity of bioweapons threats. Much of the emphasis today is with the use of agents by terrorists. The limited number of bioattacks in the past and the difficulties experienced by even well funded groups using naturally occurring pathogens (for instance, the Japanese Aum Shinrikyo cult) suggests a low likelihood of mass casualty attacks. Following from this, the possibility that such groups could or would employ advanced life science research then are even more remote. Yet, the situation is more complex than this. The potential for state sponsored terrorism significantly increases the possibility of successful weaponization. Even if one regards this as taxing for small sized state programs, as illustrated in the case of the anthrax letters, bioattacks need not inflict mass casualties to be highly disruptive. Many though have cautioned that developments in biotechnology in the near future will enable a much wider range of destructive options which will be within the reach of many groups (Fraser and Dando, 2001; Petro et al., 2003; Poste, 2003). With this uncertainty about threats, questions are being asked about the principal drive for present deliberations, whether that be the recently appreciated terrorist threats, the novel possibilities generated by rapid scientific developments, or the renewed public profile of biological weapons (Rappert, 2003a).
Various measures relating to the oversight of research have been initiated. In early 2003, a group of 32 largely American based scientific journals met to agree guidelines for reviewing, modifying, and perhaps rejecting research articles where ‘the potential harm of publication outweighs the potential societal benefits’ (Journal Editors and Authors Group, 2003). The 2003 Biotechnology Research in an Age of Terrorism report recommended the establishment of an oversight system to review and assess so-called ‘Experiments of Concern’. Initially this category includes activities such as increasing the transmissibility of pathogens, enhancing the virulence of agents, and rendering vaccines ineffective. The report also called for ‘national and international professional societies and related organizations and institutions [to] create programs to educate scientists about the nature of the dual use dilemmas in biotechnology and their responsibilities to mitigate its risks’ (NRC, 2003, p. 3). Many of NRC recommendations are to be implemented by a newly formed National Science Advisory Board for Biosecurity (NSABB). This board has been charged with developing criteria for identifying and evaluating the risks and benefits with research and also to develop ‘mandatory programs for education and training in biosecurity issues for all scientists and laboratory workers at federally-funded institutions’ (NSABB, 2004). In contrast, to date the responses initiated in relation to ‘dangerous research’ outside the US have been more limited. In the UK, for instance, much of the policy discussion has centered on community self governance measures such as professional codes of conduct and university undergraduate and postgraduate teaching provisions (UK House of Commons Foreign Affairs Committee, 2002; Royal Society, 2004; Report of Royal Society and Wellcome Trust Meeting, 2004), though the exact aims and content of these initiatives remains poorly specified (Rappert, 2003b; Rappert, 2004).
Dual Use Research and the Life Science Communities As outlined in the previous section, the long held presumption in relation to the life sciences that national security is best served by the beneficial and protective innovations deriving from research free from security constraints or oversight has been called into question. While the storage and security of pathogens as well as the vetting of personnel working with such agents have been regulated for some time, today the possible future consequences of the data, conclusions, and techniques of fundamental research are under scrutiny (Marburger, 2003). Arguably the extent and type of scrutiny is historically unmatched. Those in the life sciences have been urged to ‘lose their innocence’ and devise responsive measures (Morse, 2003; Poste, 2001) before they are imposed (Albright, 2003; UK House of Commons Science & Technology Committee, 2003). Lively debate has taken place about what, if any, security review or oversight procedures are prudent (Block, 2002; Knezo, 2003).
In these conditions of contestation and uncertainty, the proper governance of research is a matter of some dispute. With respect to dual use issues in the life science community, two points are worth stressing. First, as suggested above, one topic upon which many agree is the need for the education of scientists (see as well ICRC, 2004; Oborne, 2004; Royal Society, 2002; WMA, 2002). Yet, that overall agreement is belied by the lack of specification about the content and specific aims such provisions. Should that, for instance, consist of providing information on the history of biological warfare, stimulating generic concerns about the responsibilities of scientists today, alerting researchers to security considerations for individual decision making, or confronting scientists with the malign potential of their work?
Addressing what sort of education is sensible is hampered by a second important point: the past dearth of empirical data on the extent of researchers’ knowledge about dual use issues or evaluations of possible oversight measures.1 Following from the points made above, it is fair to say that issues surrounding the security implication of research findings have not been a topic of widespread professional discussion in the past (see Barnaby, 1997). Yet, the extent to which researchers have considered these issues has obvious bearing on what sort of educational provisions would be prudent. While policy-orientated conferences about dual use issues abound post 9-11, the extent of the participation of practicing researchers is necessarily limited.
Strategies for Engagement As suggested above then, the growing attention to the dual use issues associated with advanced life science research poses significant dilemmas for the conduct of research, ones which could have significant implications for future practice. Yet, at the same time, it is less than clear how much or what practicing researchers have thought about the issues at stake. In this situation, analysts wishing to investigate these issues face important choices about the types of interactions they foster through their own research.
Consider some reflections on conventional interviewing techniques. As part of a pilot collaborative project about genetics and bioweapons,2 in 2003 the author conducted 16 semi-structured one-to-one interviews with university sector British life science researchers regarding dual use issues. On the basis of a technical review conducted by Dando, it was decided to focus on those investigating the functioning of muscarinic acetylcholine receptors in the brain;3 this provided a bounded sub-population where all those doing significant research in the UK could be approached. Prior to the interviews, interviewees were sent a one-page sheet outlining both past military interest in acetylcholine transmission and a summary of key recent scientific trends. The interviews sought to determine how scientists defined the possible biological weapons applications of their research, where problems with research derived from, and what they thought of ongoing debates about security regulations or oversight measures. To summarize, interviewees indicated little awareness of bioweapon prohibition agreements or ongoing security-orientated deliberations and in addition only three stated ever having considered the weapons applications of their work (two of whom as a result of being approached by US military establishments) (Rappert, 2003c).
For the purpose of this paper, two reflections are worth noting about the interviews. First, they repeatedly bordered on the awkward and confrontational. As scientists were being asked about possible negative consequences of their work that they had largely hitherto ignored (and thus had not formed well-thought out rationales regarding) as well as the prospect of restrictions on their activities, the management of confrontation was a continuing preoccupation. So, many participants offered blanket reasons against any additional security controls on research by suggesting such measures would comprise the open character of science or that restrictions would be futile given the extent of knowledge already in circulation. In light of commercialization and competitiveness pressures in research (e.g., Thackray, 1998), for instance, it was readily possible to question the veracity of some claims, but in the situation of one-to-one interviews this sort of challenging threatened to degrade the interview into an adversarial to and fro inquisition. In this case that meant an opposition between a junior sociologist and (almost always) a more senior biologist about the implication of his work. Second, despite the ongoing tensions, by getting scientists engaged with issues which few of them had given prior consideration, the interviews arguably provided something of an educative experience. While most interviewees doubted the merits of limitations on publications or research agendas as well as the need for pre-project oversight reviews, despite the often initial doubt about the relevance of dual use concerns, through the interaction of the interviews none in the end refuted at least the potential for the malign application of their research. Yet, not least for practical reasons relating to cost and time, one-to-one interviews are limited in their ability to form a strategy for education.
Given the argument up this point in the paper, the understanding of the dual use issues with life science research could usefully benefit from a strategy of combining research and education. As part of any approach, it would be necessary to question the merits of current and proposed policies while questioning how that questioning was conducted. Given this, a formalized survey method would risk asking questions that are not understood by, have different or little meanings for, or are dismissed by scientists. One-to-one interviews allow for more interaction, but also threatened to decay into oppositional exchanges.
Focus group research is one technique that has gained considerable popularity in recent decades, particularly in marketing but more recently in the social sciences. ‘Focus group’ methods differ considerably in terms of their make-up, but generally consist of a group of 5-9 people that collectively discuss a predetermined set of issues regarding a given topic through the guide of a so-called moderator (or facilitator) (see Stewart and Shamdasani, 1992). Two advantages are frequently claimed for such groups. One, they are ‘ideal for exploring people’s experiences, opinions, wishes and concerns. The methodology is particularly useful for allowing participants to generate their own questions, frames and concepts and to pursue their own priorities in their own terms, in their own vocabulary’ (Kitzinger and Barbour, 1999, p. 5). As such, focus groups allow for an examination of the whys behind individuals’ thinking. Two, they entail ‘the explicit use of the group interaction to produce data and insights that would be less accessible without the interaction found in groups’ (Morgan, 1998, p. 12). As Krueger (1998, p. 20) states, ‘focus group interviews produce data derived from a group process in a focused manner. As a result, participants influence each other, opinions change, and new insights emerge. Focus group participants learn from each other, and things learned can shape attitudes and opinions. The discussion is evolutionary, building on previous comments and points of view.’ As suggested by these quotes, the open-ended character of focus groups provides a flexible and responsive way to undertake research.
In relation to exploring the dual issues aspects of life science research, such characteristics are desirable for a variety of reasons. As the security implications of biological research is a rather novel topic for bioscientists (and security analysts), understanding how they conceive of and frame the basic issues at stake is vital. Potentially at least, the interaction between scientific peers could be a way of minimizing both the asymmetrical relation between outside researchers and scientists vis-à-vis technical expertise as well as the potential oppositional relation between interviewer and interviewee(s). As interviewers do not have to press particular individuals with potentially threatening questions, group interviews allow for a space for personal reflection and withdrawal. Indeed, some have suggested focus groups might be particularly useful for examining ‘sensitive’ issues (Farquhar and Das, 1999; Kitzinger, 1994). The interactive dimensions can, in turn, serve educational purposes and foster change in people’s thinking (Baker and Hinton, 1999).
The flexibility and openness afforded by this general methodology though have also been objects of criticism. The typical use of ‘purposive’ sampling and the often low number of focus groups conducted means those employing this methodology rarely strive for ‘statistical representativeness’ (O’Brien, 1993). In marketing at least, focus groups often serve the purpose of informing other forms of research. When focus groups are held with pre-existing groups, then their interaction can be said to be ‘contaminated’ by past relationships. Even if members are unfamiliar which each other, the group dynamic is said to result in conformity and individual censorship (Albrecht et al., 1993). Of course, with any type of (overt) social research, the potential for individuals to offer socially preferred responses and rationalized justifications has long been recognized (e.g., Scott and Lyman, 1968). As well, in terms of analyzing focus group discussions, because of the interactions between participants and the scope for argumentation, it is difficult to ascribe ‘views’ to individuals (see below). In short, the scientific basis of focus groups has been called into question. On a more practical level, the resources, expertise and planning required for successful focus groups are said to nullify many of the advantages of conveying group-type interviews.
In response, proponents of focus groups have acknowledged the often lack of statistical generalizability and the resource demands, but defended them by arguing that even when done on their own, if conducted properly, they can produce verifiable results (Krueger, 1998). Defining a research protocol, conducting disciplined moderation, and establishing feedback between researchers and group participants, other researchers and outside experts are all presented as vital for producing systematic results.
Despite the burgeoning literature about focus groups, arguably it remains deficient in addressing crucial issues surrounding the openness of and rationale behind moderator questioning and subsequent probing. As both focused and disciplined as well as open and reciprocal, those moderating such groups must balance or otherwise resolve questions about how to question. In this respect, Morgan (1998, 58) advises that ‘…focus groups allow you both to direct the conversation towards topics that you want to investigate and to follow new ideas as they arise’ but how the competing aims desired should be reconciled is rarely a topic of detailed consideration even when noted as a crucial. So Kitzinger (1994, p. 106) advocates that ‘trying to maximize interaction between participants could lead to a more interventionist style: urging debate to continue beyond the stage it might otherwise have ended, challenging people’s taken for granted reality and encouraging them to discuss the inconsistencies between participants and within their own thinking’ but without any further elaboration of the rationales for how choices are made about what to do. The tensions with questioning not only have implications for assessments of ‘rigor’ but also claims that focus groups allows participants to ‘generate their own questions, frames and concepts and to pursue their own priorities in their own terms, in their own vocabulary’ (Kitzinger and Barbour, 1999, p. 5). Krueger (1998, Appendix) provides various examples of questions posed as part of group sessions, but all can be read as asking relatively specific matters and they are static over time. Instead of treating the openness and expressiveness as given properties of this method, future sections of this paper treat them as matters in need of continuing attention.