There is a classic formulation of the problem of controlling emergent technology- when an innovation is contained enough to control, its implications are impossible to predict; and by the time implications become clear, the technology is impossible to control. Couple this with the increasingly anticipatory nature of national security- and you start to appreciate the ambiguous and contestable situation in which experts in this policy space operate. In this article, Yong-Bee Lim considers a key issue facing experts on biosecurity- what is the risk of bioterrorism, and how can we make this call? He discussed how these debates have played out, with particular emphasis on emergent technologies.
Mind the Gap(s) – The Importance of Field Research in Biorisk Assessments
National security communities have become increasingly troubled by the spectre of bioterrorism, with discussion revolving around a number of key cases. In 1984, a cult in Oregon deliberately contaminated local salad bars and restaurants with Salmonella enterica Typhimurium to influence a local election. In 1993, a Japanese millenarian cult called Aum Shinrikyo attempted several biological attacks in Tokyo using what they believed was Bacillus anthracis, the causative agent of anthrax. In 1995, a white supremacist named Larry Wayne Harris committed fraud to acquire samples of Yersinia pestis, the causative agent of plague, from the American Type Culture Collection. In 2001, an anthrax attack distributed through the U.S. postal system resulted in 5 deaths, 17 non-fatal injuries, and financial cost estimates that range between $300 million to $800 million.
Some experts have continued to see these incidents as the harbingers of a new wave of bioterrorism in the world; a claim based on three key working assumptions:
Globalization and advances in the life sciences make it easier for terrorist organizations and lone-wolf actors to research and develop biological weapons.
Terrorist organizations appear to seek a larger number of casualties per incident, potentially making weapons of mass destruction ( WMD), including some types of biological weapon, more appealing.
The costs associated with defending against and recovering from biological attacks are orders of magnitude greater compared to what it costs terrorists to research, develop, and initiate such attacks.
However, other experts see such bioterrorism incidents as anomalies: attacks that were, and would continue to be, atypical for how terrorists operate. These experts use the same cases above, but tend to operate with a different set of working assumptions:
Terrorist groups require public support for their activities, which means avoiding morally reprehensible weapons like biological weapons.
Terrorist organizations are much more likely to have access to firearms and explosives, and relevant expertise. Terrorists organizations will continue to opt for the tried-and-true methods of guns and bombs, particularly if they are small organizations without many resources [i]
How can the views of experts vary so wildly on assessing the future risk of bioterrorism? Marie Isabelle Chevrier’s chapter in Wenger and Wollenmann’s Bioterrorism: Confronting a Complex Threat (2007) provides insights into explaining this phenomenon. In her chapter titled “Why Do Conclusions from the Experts Vary?”, Chevrier explains that expert conclusions vary because of methodological inconsistencies in bioterrorism risk assessment. Several inconsistencies she brings up in chapter include:
Strongly distinguishing between biological, chemical and nuclear weapons research, production methods, equipment, and delivery mechanisms vs. treating these different weapons types as essentially the same.
Specifically characterising threat scenarios (e.g setting a defined timeframe) versus more unspecified forms of threat which may not be bound by time or place.
While the importance of operationalization is a clear takeaway, one of the important messages that I took from this chapter is the importance of good data. Given the small number of bioterrorism-related incidents since the 1980s, the lack of data makes it difficult for researchers to make empirically based assertions of the trends and probabilities of different bioterrorism scenarios.
A Resurgence of Concern: Technology Emergence, Convergence, and Bioterrorism
Despite this issue of data, there has been a resurgence in expert concern in recent years. After looking through current literature, three factors stand out to explain this heightened concern:[ii]
As technological capabilities grow, the potential for diverse applications (including harmful applications) grows.
As technologies converge with new communities, the outcomes of emerging scientific disciplines and unique societal applications become less predictable.
As technologies become easier and cheaper to use, it becomes more difficult to control both who uses a given technology and what people use the given technology for.
In the non-state actor context, these factors potentially contribute to the empowerment and enhanced capabilities of small groups and individuals to do greater amounts of harm (economic, psychological, and physical) than historically possible. The theory is that these actors, primarily terrorist groups and lone wolves, may find biological weapons easier to grow, develop, and distribute as:
Capabilities like gene editing become less prone to off-target effects and is done with a deeper understanding of genomic sequence- broadening the scope what is practically possible;
New disciplines, like bioinformatics, both provide additional information on biological systems and provide alternate ways to do biological experimentation;
Developers provide cheaper and easier-to-use tools like CRISPR kits and access to cheaper gene sequencing and synthesis providers.
A Problem of Data: Bioterrorism in the Modern Day
There are two issues with this view in the bioterrorism context. If I were channelling my inner Chevrier, the first issue is that available data on modern bioterrorism incidents do not lend much weight to concern about the threat of increasingly technically sophisticated and deadly attacks. Gryphon Scientific’s 2016 report titled “Risk and Benefit Analysis of Gain of Function Research” provides a comprehensive terrorist and extremist events tied to biological weapons from 1972 – 2014. Out of the 42 incidents listed, several recurring themes include:
Low-to-No Success Rate: Of the 20 cases of use and attempted use of biological agents, 15 of them either failed or resulted in less-than-anticipated casualties. This translates to a failure rate of 75% for use and attempted use of biological agents.
Anomalies, Not Trends: Larger-scale morbidity and mortality rates like those experienced with the Amerithrax Incident in 2001 and the Rajneeshee food poisoning incident of 1984 are anomalies in bioterrorism incidents. Most cases result in little to no deaths, or apprehension by authorities before an attack even commences.
A Problem of Abductive Reasoning: The Case of DIYBio
As I continue to channel my inner Chevrier, the second issue I notice is one of abductive reasoning: if it looks like a duck, and it quacks like a duck, then it probably is a duck. In the context of bioterrorism risk, abductive reasoning can lead individuals to conclude that any emerging phenomenon with certain characteristics may constitute a bioterrorism threat. In our case, these characteristics include increased technical capabilities, novel socio-technical convergences, and greater accessibility of scientific tools may constitute a bioterrorism threat.
One such group that emerged in a time of increased technical capabilities, new interactions between technology and society, and cheaper and easier-to-use tools is the Do-It-Yourself Biology (DIYBio) community: a community of professional and amateur scientists that engage in science education and research outside of traditional academic, industrial, and government laboratories.
This group has received a significant negative attention from security practitioners and media sources. In the DIYBio context, proponents of heightened bioterrorism threats focus purely on the increasing availability of life sciences techniques, equipment, and capabilities to a broader range of users. Using this abductive reasoning, people reach the conclusion that DIYbio participants could constitute a bioterrorism threat. Examples of this threat include practitioners using genetic engineering or synthetic biology to manipulate an organism and cause harm to crops, animals, and humans. Negative media reports that paint a picture of sneaky people doing dangerous biotechnology experiments in their garages, houses, and kitchens reinforce this security concern [iii]
Unlike most cases of researching bioterrorist groups (which would either involve doing extensive case studies or directly trying to infiltrate some terrorist or extremist organization), it is much easier to gather data on the bioterrorism risks associated with DIYBio: going out and interacting with the community! Some takeaways I have developed after several years of interactions and research with the community include:
Capabilities: While community labs have some donated and home-made equipment to use in the laboratory, they also experience significant delays and issues like any other lab. These can include analysis errors, equipment breakage, environmental contamination issues, and difficulties keeping a fully-stocked laboratory due to financial constraints.
Convergences: Converging individual and community needs with technical capabilities can lead to unusual or unexpected projects. However, factors such as the community lab director’s goals, community-wide expertise, and the community the lab is a part of can effectively direct and limit what projects are pursued.
Accessibility: Increased accessibility of technology and knowledge do empower people to do independent research, learn basic lab skills, and acquire equipment. However, community labs do not just exist because of cheaper equipment and dispersed knowledge on the internet. There needs to be a critical mass of scientific expertise in the community, an actual desire and/or need for a community lab space in an area, and enough resources (which includes significant financial contributions) to start and support a space.
Conclusion: Mind the Gap(s)
I do not doubt that safety and security practitioners will continue to identify and project problems as technological and scientific advances continue. Some of these problems will manifest in one form or another. Other problems may never arise.
As the safety and security communities forecast emerging bioterrorism risks, we will hopefully notice the gaps in our data. These gaps in data directly interfere with our ability to understand an existing or emerging phenomenon. This lack of understanding, in turn, directly interferes with our ability to recommend good policies to mitigate future risks.
Some of these gaps, which include future bioterrorism incidents, are difficult to investigate in advance; we can only research them after the fact. However, there will be gaps that can be investigated immediately. As hinted at in the case of DIYBio, minding the gap and filling it with data can help in understanding an emerging phenomenon, building a data-driven understanding of the benefits and risks, and providing accurate information to develop good recommendations and policies.
[i] Dr. Gregory Koblentz (2011) wrote a great synthesis of CBRN terrorism prediction titled “Predicting Peril or the Peril of Prediction? Assessing the Risk of CBRN Terrorism” in the journal Terrorism and Political Violence.
[ii] Literature that highlights these main points include John P. Caves, Jr. and W. Seth Carus’ The Future of Weapons of Mass Destruction: Their Nature and Role in 2030, Jonathan B. Tucker and Raymond A. Zilinskas’ “The Promise and Perils of Synthetic Biology”, and Natasha E. Bajema and Diane DiEuliis’ Peril and Promise: Emerging Technologies and WMD
[iii] Some examples of negative media portrayals of DIYBio include Howard Wolinsky’s (2009) “Kitchen Biology”, Jeanne Whalen’s (2009) “In Attics and Closets, ‘Biohackers’ Discover Their Inner Frankenstein’”, and Emily Baumgaertner’s (2018) “As D.I.Y. Gene Editing Gains Popularity, ‘Someone is Going to Get Hurt’”
Cover image rights here