Many researchers stress that powerful technologies, in particular biotechnology, involve risks. I agree with Martin John Rees, who notes that we are heading for an age in which some individuals, by mistake or in order to create fear, can cause social breakdown with such extreme speed that resulting government action is completely inadequate. Therefore, our priority should be to find ways to normatively contain these existential or catastrophic risks. As a first step, an effective universal biotechnology governance regime should be agreed upon and implemented.
In past decades, we have seen rapid advances in many areas of science which have improved our lives significantly. Scientific and technological progress helps us to cure deadly diseases and fight hunger. Nobody denies that they provide great benefits, not only in medicine and agriculture. However, it also can hardly be denied that this progress is mirrored by new and emerging risks. It seems to be common sense that “[t]he physical limits on how much damage any individual or small group could do are becoming less and less constrained”. Especially areas of biotechnical progress are linked to risks that are even existential and catastrophic. At the beginning of 2018, it was Bill Gates who stressed the increasing risk of a bioterrorism attack and the need for an effective global preparedness and response system against pandemics as an issue of global health security.
For the people that suffer or die because of such a pathogen, it seems irrelevant whether the pandemic was caused by a natural mutation or by error, by terrorist attack or by act of war. From the ethical and legal point of view, however, there is an important difference. If a risk is human-made, humans are responsible if this risk materialises; therefore, looking at human-induced biosafety and biosecurity risks, it should be a priority for all states and for the international community and world society as a whole to govern these risks in a legitimate way. But despite the interdisciplinary debates during the last years and policy initiatives that focus on how to govern existential and global catastrophic technological risks, it is unclear what principles and rules should be seen as a foundation for a legitimate and universal future biotechnology governance regime.
I want to close this lacuna, and argue that legally binding human rights are decisive pillars of such a multi-layer regime; moreover, I want to solve the problem of how to govern existential and global catastrophic technological risks, and lay down in more detail principles and rules that should be seen as a foundation for a legitimate and universal future biotechnology governance regime. These principles and rules should be applied – mutatis mutandis – to other areas of existential and global catastrophic technological risks, such as artificial intelligence.
Biotechnology Experiments Today
Before we think about principles of a legitimate biosecurity governance regime, we should look closer at the field of biotechnology experiments that are carried out today. These include so-called gain-of-function studies of concern (GOFsoc) and dual use research of concern (DURC). Recent examples of high-risk research are the widely discussed experiments on influenza viruses in 2012, in which more dangerous forms of the virus were created than those that naturally existed. The same is true with regard to experiments published in 2014, which revealed the possibility that cancer cells can be passed on through common cold viruses, and research that created infectious horse pox viruses through gene synthesis in 2017.
These biotechnology risks are also linked – not exclusively – to experiments that rely on CRISPR-Cas9, the new tool of genome editing. With this method, it is possible to edit DNA more easily, cheaply, and precisely than before. Genome editing research holds great promise for future biotechnical and biomedical applications but there are also major concerns since the discovery gives scientists the power to rewrite the codes of life and so-called off-target effects cannot be excluded. The same scientific breakthroughs in genome editing which can be used to cure a disease, can also be used to create a weaponised version of a certain pathogen.
A special challenge of the CRISPR-Cas9 genome editing technology is posed by gene drives (mutagenic chain reactions). Gene drive systems promote the spread of genetic elements through populations, by ensuring that they are inherited more frequently than Mendelian inheritance would predict. Gene drives can be used to change the genes of certain mosquito species, shutting off their ability to carry malaria and suppressing the size of their population. Gene drives can be used to change the genes of certain mosquito species, shutting of their ability to carry malaria and suppressing the size of their population. These genetically modified mosquitoes (GMMs) have emerged – for some – as a promising new tool to combat vector-borne diseases like malaria and dengue. In agriculture, ‘undesirable’ plants could be made more susceptible to herbicides. What risks these intrusions carry for ecosystems remains unclear. Gene drives might carry existential risks because there are so-called low threshold drive systems and only small numbers of individuals need to be released to start the drive process. At its theoretical extreme, only a single individual may be sufficient (including accidental release). As a result, there are, as far as is known, currently no field experiments involving gene drives and even natural scientists warn about setting free organisms equipped with gene drives because of the high associated risks. The risk of abuse, for instance by employing this technology as a weapon, cannot be excluded, thereby making it a case of dual use research of concern.
A final example of current dual use (or even single use) research of concern is a recent research programme with the aim to create genetically modified viruses to infect staple food crops in fields using an insect-based dispersion system, as the stated agricultural benefits of this research seem to be highly implausible but its potential for generating a new class of biological weapons might be considerable. This DARPA Insect Allies programme is not research of concern because it is a military-funded programme, but because an easy simplification of the viral work using already existing technology could generate devastating bio-weapons.
Shortcomings of the Current International Governance Regime
These scientific tools and biotechnologies are becoming increasingly powerful, cheap, and available to more and more individuals. A key question for a governance regime is therefore whether we are sufficiently prepared to detect risky uses and developments and prepared to counteract destructive uses and developments of high-risk biotechnology. The answer is – so far – negative. The common problem of biotechnological development is that there is no sufficient international governance regime. As a general overview, one can state the following: thus far, no comprehensive universal international treaty on biosecurity, or – more generally – on dual use research of concern, exists. There are only a few treaties, such as the Convention on the Prohibition of Military or Any Other Hostile Use of Environmental Modification Techniques (ENMOD-Convention), the Biological Weapons Convention (BWC), the Convention on Biological Diversity, the Cartagena Protocol on Biosafety, and the Nagoya – Kuala Lumpur Supplementary (Liability) Protocol (that entered into force in 2018) that are applicable to some areas of biotechnology. However, even if international treaties exist, they have major drawbacks. Not all of the relevant states, where research that may be linked to biosecurity risks is conducted, are parties of existing treaties.
For instance, the Cartagena Protocol entails binding rules for living modified organisms (LMOs) resulting from modern biotechnology that may have adverse effects on biological diversity. The Cartagena Protocol is indisputably applicable if foreign DNA is integrated into the target organism’s genome, as this is the case – for instance – with gene drive systems. According to the Cartagena Protocol, when a party knows of an occurrence under its jurisdiction which may result in an unintentional transboundary movement of a living modified organism and is likely to have significant adverse effects on the conservation and on sustainable use of biological diversity, the party shall take appropriate measures to notify affected or potentially affected states as well as the Biosafety Clearing House. Each party shall adopt appropriate domestic measures aimed at preventing transboundary movements of LMOs. Apart from this, the so-called advanced informed agreement (AIA) governs the import of LMOs. However, these rules bind only states parties to the Treaty: having 171 parties, 170 states (including Germany) and the European Union, relevant state actors such the United States have not signed or ratified the treaty. The same is true for the Kuala Lumpur Liability Protocol, which is applicable – for instance – to gene drives. It binds states parties to questions of liability and redress if the transboundary movement of living modified organisms has caused damage.
Most importantly, disarmament treaties, especially the ENMOD-Convention or the BWC, do not provide sufficient protection against the aforementioned risks of misuse of biotechnology because research conducted for peaceful purposes is neither limited nor prohibited. The ENMOD-Convention expressly lays down that it “does not hinder the use of environmental modification techniques for peaceful purposes”. The BWC does not prohibit the development of biological agents of “types and in quantities that have a justification for prophylactic, protective or other peaceful purposes”. But these prohibitions are ‘toothless’ since the BWC does not possess a verification and implementation regime.
This means as a first interim result: although there are concerns that biotechnological progress in the twenty-first century may cause great, even existential and global catastrophic, risks, international treaty law is not sufficient to assess, significantly minimise, or eliminate these risks. Even if international treaties exist, they have major drawbacks. Not all of the relevant states, where research that may be linked to biosecurity risks is conducted, are parties to existing treaties; moreover, there are few relevant international soft law norms.
Shortcomings of Private Rule Making and Codes of Conducts
Even more loopholes and drawbacks exist with regard to private rule making, including codes of conduct, and to mere recommendations by stakeholders and expert bodies. This does not mean that the usefulness of ‘bottom-up’ rule making by private actors and stakeholders, including research organisations, funding bodies, and transnational corporations should be neglected or negated, and it should not be assumed that codes of conduct – as bottom-up approaches that rely on the consensus of stakeholders – cannot be effective. There have been examples of private rule making and codes of conduct that had and still have a major impact in framing and limiting specific areas of research, as for instance the Declaration of Helsinki on ‘Ethical Principles for Medical Research Involving Human Subjects’ by the World Medical Association (WMA).
These scientific tools and biotechnologies are becoming increasingly powerful, cheap, and available to more and more individuals.
Codes of conduct are part of a multi-layer governance that consists of rules of international law, supranational and national law, private norm setting, and hybrid forms that combine elements of international or national law and private norm setting. Promising ‘bottom-up’ initiatives are underway in order to limit problems of dual use research, for instance by framing guidelines for dual use research in general or minimising risks of misuse with regard to a certain field of scientific development, as in the area of Artificial Intelligence (AI) by way of the 2017 Asilomar AI Principles. But the normative force (in the sense of – de facto – bindingness and effectiveness) of these private rules significantly depends on complex criteria and preconditions.
Additionally, a moratorium on a specific kind of research that is based on the consensus of the relevant scientific community and backed up by the relevant scientific journals – which will not publish experiments that violate the moratorium – might be even more effective than a moratorium or prohibition based on an international treaty that is implemented – top-down – by states parties. A moratorium based on the consensus of the scientific community might close lacunae if states which are parties to a specific treaty, as for instance the Convention on Biological Diversity, could not agree upon a worldwide moratorium for certain types of high-risk biotechnology, such as gene drives.
Therefore, multi-layer governance should combine (‘top-down’) state-based rule making and implementation with (‘bottom-up’) norm creation and implementation by private actors and stakeholders to close lacunae and enhance the effectiveness of a biosecurity regime.
Hence, what is missing in the area of biotechnology – different from other fields of scientific research and technological development – and what should be agreed upon by 2030 is a global ‘Code of Conduct for Responsible Biotechnology and Biosecurity’ that should be based on an overlapping consensus of the global scientific and biotechnology community and should be coherent with principles of the ‘Declaration (or: Treaty) for Responsible Biotechnology and Biosecurity’ that are spelled out below.
Global and Catastrophic Biosecurity Risks
In order to find rules that can form the basis of a legitimate governance regime of existential and global catastrophic biosecurity risks, a closer look must be taken at the notion of legitimacy and the special factual features of these scenarios. I view legitimacy primarily as a normative, not a descriptive, concept: legitimacy refers to standards of justification of governance and – possibly – obligations. Hence, if I speak of ‘legitimate governance’, it means that the guiding norms and standards have to be justifiable in a supra-legal way (possessing ‘rational acceptability’).
If one thinks about the principles of a legitimate biotechnology governance regime, one has to note that nearly all the experiments mentioned in the first section can be viewed as so-called ‘low probability/high risk scenarios’ or ‘unknown probability/high risk scenarios’. This is the case, since – for instance – the probability that a modified, deadly virus may escape from a laboratory can be quantified and is very low. But it is uncertain whether there will be reliable models to quantify the probability that certain types of gene drives pose existential or global catastrophic risks. This is especially true with regard to the problem of ‘dual use’ of scientific results, the above-mentioned risk that terrorists or criminals misuse research findings or technical developments to cause severe harm.
If we agree that legitimate governance requires, as a necessary condition, a rational reaction to the scenarios mentioned before, one could argue that a decisive element of a governance regime should be that the relevant actors, especially the states, assess and reduce existential and global catastrophic risks in a way that mirrors the actual probability and severity of a certain risk.
The Human Rights Framework
My thesis is that a legitimate governance regime of existential and global catastrophic risks should be based on human rights, more precisely on legally binding human rights. If we turn to the human rights framework as part of international law, it can be seen that international legal human rights make it possible to spell out the decisive values that have to be taken into account for assessing the scenarios mentioned above.
First of all, the freedom of research is not only a justified (moral or ethical) value, it is also a legally binding human right: What is missing in the area of biotechnology and what should be agreed upon by 2030 is a global ‘Code of Conduct for Responsible Biotechnology and Biosecurity’.there is the shared view that the freedom of research is entailed in the rights of freedom of thought and of freedom of expression in international human rights treaties (as for instance the International Covenant on Civil and Political Rights and the European Convention on Human Rights). But to protect the freedom of research as a human right does not mean that this freedom is absolute. According to legal international human rights, the protection of the life and health of human beings is – inter alia – a legitimate aim that can justify proportional limitations of the right of freedom of science. The human rights framework therefore stresses that, if states limit the freedom of science for legitimate purposes, it is necessary to find proportional limitations, even if the probability of the realisation of an existential and global catastrophic risk is close to zero or cannot be quantified.
Second, and more importantly, the so-called first generation human rights oblige states not only to respect, but also to protect the fundamental rights of individuals. If we concentrate on those human rights treaties, ratified by the most industrialised states, including Germany and the United States, they state that states parties are obliged by international human rights treaties to take appropriate (legal) measures to protect the life of individuals.
In order to avoid the irrational bias of neglecting small risks, one can argue that the higher the severity of the possible damage, the lower the demands on how probable a risk must be. From this, one can conclude as well that the duty to protect the life and health of individuals includes a duty of the state organs to assess existential and catastrophic risks as long as these measures are proportionate.
This does not mean that all problems are solved by applying human rights; the central question remains whether there are measures to assess and reduce existential and global catastrophic risks that are neither ineffective nor disproportionate. This is an especially challenging question if we talk about research that is aimed at protecting the health and preventing the death of human beings. If a virus is modified to find a vaccine against influenza or if genome editing is used to change mosquitoes with gene drives in order to wipe out malaria, the research itself aims to serve the life and health of human beings.
Nevertheless, according to the proposed interpretation, human rights oblige states to assess and reduce existential and global catastrophic risks in a proportional way so that the probability and severity of a certain risk can and should be taken fully into account. As a result, a rational and legitimate perspective to govern existential and global catastrophic risks is given, which includes the shared values of the states.
Elements of a Legitimate Multilayer Governance Regime
Legally binding human rights are decisive elements of a legitimate multilayer governance regime, but they give only general guidance on how to govern existential and global catastrophic risks. A legitimate governance regime of existential and global catastrophic risks should be based on human rights, more precisely on legally binding human rights. In order to frame the specific problem and to spell out rules and principles for the necessary proactive and preventive regulation – including procedures of risk assessment and risk reduction –, a step forward would be a new international soft law declaration or an international treaty – the ‘Declaration for Responsible Biotechnology and Biosecurity’ or the ‘Treaty for Responsible Biotechnology and Biosecurity’ – that specifies the human rights-based values and the implementation of those values in the area of biotechnology and – especially – in cases of existential and global catastrophic risks. The declaration or treaty should lay down in more detail principles and duties to assess and reduce biosecurity risks that are compatible with the human rights framework, and with other justified ethical approaches including justified rules of decision-making.
- I propose that, inter alia, the following five demands are a sound basis for such a declaration or treaty and coherent with a human rights approach. They enhance substantive legitimacy – to spell out in more detail how to enhance procedural legitimacy is beyond the aim of this article.
- As a demand of global justice, one could justify that since existential risks and global catastrophic risks endanger humankind, the benefits of biotechnological progress connected with those risks must also be shared with humankind as well.
- Further, it does not seem to be a disproportionate limitation of biotechnological progress that if there is a worst case and there is no probability for benefits, states have to prohibit the studies or products that can cause the worst case (according to the so-called Maximin Rule, which entails making the choice that produces the highest payoff for the least advantaged position).
- It does not seem to be a disproportionate limitation of biotechnological progress to lay down a rule that there is a burden of proof for those who fund science (or for the scientists) to show that there are more benefits than risks if there is plausibility for an existential or global catastrophic risk.
- It is unclear what the guiding principle should be if after a risk assessment, there is still an unknown probability of an existential and global catastrophic risk and – at the same time – an unknown benefit of the proposed biotechnology. One could stress that, as far as democracy is a value enshrined in international law, a state should solve the problem by relying on the democratic process to determine whether to allow or prohibit this kind of technological progress. On the other hand, because of the global dimension of the risks, it seems to be more plausible to argue that we need a global consensus to solve the problem and to determine whether to allow or prohibit these experiments or techniques.
- There may be deontological limits to a utilitarian approach that can be derived from human rights in the following scenario and which should be spelled out: if there is a probability of an existential or global catastrophic risk as a consequence of certain experiments or biotechnological developments, but a great number of people could be saved if the research is conducted or the technology is applied, one could conclude that according to a rational risk-benefit-assessment – on the one hand – there is no reason (and no justification) to prohibit or limit this kind of experiments or biotechnology. On the other hand, I would stress that there is a deontological limit that can be deduced from legal international human rights and the core of human rights law, namely the duty to protect the human dignity of individuals. Accordingly, no human being may be treated as a means to save other individuals and therefore the research or technique could be prohibited or limited if there is no informed consent by those who might be affected.
Rules for Responsible Biotechnology and Biosecurity
The most important step forward seems to be that as many like-minded states as possible, at least those of the European Union including United Kingdom, negotiate the above-mentioned international Declaration (or: Treaty) for Responsible Biotechnology and Biosecurity which shall be applicable, especially, with regard to dual use research of concern in the area of biotechnology.
The international Declaration (or: Treaty) for Responsible Biotechnology and Biosecurity should lay down – inter alia – the following rules:
- An important aim is to raise the level of awareness for questions of biosecurity in the scientific community. States shall ensure that individuals involved in biosecurity-relevant research acquire the competence to identify the dual-use potential of their research. To achieve this aim, questions of biosecurity should be integrated into undergraduate and graduate curricula and into the training of scientists.
- States shall ensure that in funding, planning, and executing biosecurity-relevant research, the following obligations to minimise risks are applicable to all relevant actors (especially researchers, funding organisations, and biotechnology corporations). This also applies if research cooperation programmes are carried out. Additionally, those acting in the capacity of reviewers or editors as part of the scientific process shall adhere to the principles listed below:
a) As a first step, the purposes, benefits, and risks of the planned research shall be identified.
b) Researchers, research funding organisations, and/or corporations shall assess whether a research programme falls within the scope of dual use research of concern (DURC) or within the scope of gain of function research of concern (GOFsoc); the results of these assessments shall be documented.
c) Research programmes shall be examined in order to assess whether the research purposes and benefits could also be achieved by research programmes that involve less risks.
d) Research programmes shall be examined in order to establish whether the benefits are sufficient to justify the risks involved. Here, it is of particular importance to examine whether the research programmes carry an unreasonably high risk for protected rights and values such as the life and the health of people and/or the environment. Should the assessment reveal that the risk is not justifiable, the research programme shall not be funded and shall not be pursued.
e) The risk can be presumed to outweigh the potential benefits in the case of a research programme in which the virulence of a micro-organism with respect to its pathogenic effects for humans or animals is enhanced to such an extent, or such enhancement is to be anticipated, that its dissemination outside the laboratory is likely to produce a pandemic amongst humans (gain of function research of concern). Such research programmes shall not be funded and not be carried out unless a direct, concrete, and overwhelming benefit in terms of diminishing hazards for human life and health is probable.
- Dual use research of concern and gain of function research of concern shall be listed in a global open access registry (or several supranational/national open access registries) in order to promote transparency of research and open communication.
a) The results of biosecurity-relevant research shall, in principle, be published; however, researchers, funding organisations, and corporations shall assess whether risks of misuse justify that research results are not (entirely) published. Here, it is of particular importance to examine whether the publication carries an unreasonably high risk for protected rights and values such as the life and the health of people and/or the environment.
- An interdisciplinary Dual Use Research of Concern Commission/Gain of Function Research of Concern Commission shall be established at the federal (national) level by every state party. The Commission shall include, inter alia, scientists and biosecurity experts. The Commission shall issue consultative votes. Researchers, funding organisations, and biotechnology corporations shall be obliged to consult this Commission before dual use research of concern/gain of function research of concern is conducted or funded.
- Public or private funding bodies, and corporations in like-minded states that operate in the field of biotechnology, should ensure that dual use research of concern/gain of function research of concern projects only receive funding if:
a) the scientist entrusted with project management has agreed to comply with this Declaration or Treaty; and
b) the research is listed in the (global/supranational/national) open access registry.
- States should consult whether a more specific definition of dual use research of concern or gain of function research of concern should be part of the Declaration or Treaty. The definition could include research intended to enhance the harmful consequences of listed agents.
Risk Assessment and Reduction Regime Based on Public International Law
Even in a world of disorder, we should not underestimate the advantages of a risk assessment and risk reduction regime based on public international law. The advantages are:
- international law is a global legal regime and is legally binding; hence, it can be enforced by various means (including national and international courts);
- non-anthropocentric goods and values (as for instance knowledge or the environment) are protected by existing rules of international law, especially human rights law (freedom of science, right to life) and environmental law;
- as far as there are lacunae in international law, new international treaties (and/or soft law declarations) can be negotiated to govern developments in biotechnology in the near and distant future;
- the non-hierarchical structure of public international law and its fragmentation give it the flexibility to find rational and legitimate ways of evaluating existential and global catastrophic risks and to prioritise those risks in the twenty-first century.
Binding international human rights are and should be decisive pillars of the future multi-layer biotechnology governance regime. International human rights protect relevant values and can be interpreted to oblige states to assess and reduce existential and global catastrophic biotechnology risks in a proportional way that mirrors the probability and severity of a certain risk.
The specific rules that make it possible to differentiate between reasonable and unreasonable impositions of risk should be laid down in a new international Declaration (or: Treaty) for Responsible Biotechnology and Biosecurity by 2030.
This is how we should develop a biotechnology governance regime in coherence with relevant and morally justified values of a humane world order that is aiming for future scientific and technical advances in a responsible manner.
Biotechnology and Security
How We Must Act
Binding international human rights should be the key pillars of the future biotechnology governance regime. International human rights oblige states to assess and reduce existential and global catastrophic biotechnology risks. The probability and severity of a certain risk should be the basis for standardisation. So, this is how we must act until 2030:
- A priority should be to find legitimate ways to govern the cluster of risks with existential or catastrophic consequences.
- All relevant states should agree on an international Declaration (or: Treaty) for Responsible Biotechnology and Biosecurity.
- Dual use research of concern and gain of function research of concern shall be listed in a global open access registry (or several supranational/national open access registries) in order to promote transparency of research and open communication.
- Like-minded states should agree on innovative mechanisms, including big data tools, for improving the implementation and compliance with existing international treaties that are decisive for limiting dual use research of concern.
- In addition, a global ‘Code of Conduct for Responsible Biotechnology and Biosecurity’ should be drafted that binds private actors – including corporations – and is coherent with the international Declaration (or: Treaty) for Responsible Biotechnology and Biosecurity.
- We need a European and/or German institute that is devoted to the interdisciplinary research of existential and global catastrophic risks.