4

Charting Human Values

Why should we consider developing gene-drive modified organisms and releasing them into the environment? How should we select sites where such organisms could be released? How should we assess the outcomes? Do we need additional oversight mechanisms to govern gene drive research and development? These and many other questions underlie discussions within the scientific community and broader society about gene drives. Because gene drives are designed to alter the environments we share, in ways that might turn out to be very hard to anticipate and impossible to reverse completely, these questions are very complex and require careful exploration. The answers depend on values—deeply held, complicated, sometimes evolving beliefs about what kinds of things, in human lives and the world at large, should be fostered, protected, or avoided, and therefore about what people should and should not do (Elliott, 1992; Macrina, 2014). Values are critical components of human identity and society. They permeate our perceptions, understanding, hopes, fears, decisions, and actions. They are reflected in our views about what morality requires of us and in our views about what is in our interests, both individually and as a society. Values sometimes find expression in the sets of ethical principles formulated to guide science and medicine (Elliott, 1992; Macrina, 2014), such as the requirement that medical research on human subjects provide a positive balance of benefits over harms, the harms of participation are not borne disproportionately by disadvantaged or vulnerable people while the benefits go to those in positions of power and privilege, and that research not be conducted without the voluntary, informed agreement of the subjects (National Commission, 1978; WMA, 2013). Such values are understood to be important enough that they need to be treated not just as conventions but as obligations that can be enforced through a system of governance. Values are also the starting point of any attempt to decide what to do with emerging technologies. The committees and commissions charged with those decisions, identifying principles and making recommendations where possible, are engaged in the task of trying to articulate and sort through the implications of values (President’s Commission, 1982; Presidential Commission, 2010).

This chapter focuses on the values involved in gene drive research. The chapter begins with a brief overview of the scholarly debate that has unfolded over the last few decades about genetic engineering. Using the case studies presented in chapter three, the committee explored in depth three broad categories of concern:

• The potential benefits and harms of gene drive research for people,
• The potential impact of gene-drive modified organisms on the environment (understood both in terms of outcomes for people and, for some individuals and cultures, as a concern about the environment in its own right), and
• Who will be affected by gene drives and make decisions about them.

The exploration of these questions provides a conceptual framework for decisions about whether and how to move forward with the science and what kinds of constraints are appropriate in making decisions about field release. The chapter thus provides a conceptual underpinning for the specific recommendations found in later chapters.

CENTRAL VALUE CONSIDERATIONS IN DEBATES ABOUT GENETIC ENGINEERING

Genetic engineering sparked ethical debate as soon as it was imagined. Initially, in the 1960s, public debate focused on the prospect of using genetic engineering on humans; the possibility that genetic engineering might be a new and acceptable way of producing better human beings was exciting to some people and raised questions about eugenics for others. In the early 1970s, as scientists developed the ability to produce recombinant DNA, some of the researchers at the forefront of the work began to ask questions about the safety and environmental impact of the new molecules. At that time the questions focused chiefly on toxicity (Macrina, 2014). But as scientists learned how to produce a variety of genetically engineered organisms—primarily agricultural plants and animals at first, and later with the emergence of “synthetic biology,” microbes that could be used in industry—critics raised additional questions about environmental, public health, and social effects (Presidential Commission, 2010). Just as gene drive technology builds on earlier kinds of genetic engineering, ethical debates about gene drives are likely to build on these earlier considerations.

The most prominent moral questions about genetic engineering have always been about its prospective benefits and harms to human beings. The guidelines developed at the 1975 Asilomar Conference on Recombinant DNA focused on ensuring safety in the handling of potential biohazards (Berg et al., 1975). The seminal report Splicing Life: The Social and Ethical Issues of Genetic Engineering with Human Beings, issued in 1982 by the President’s Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research, identified “balancing present and future benefits and risks” as the overarching ethical and social question that would have to be answered to decide whether and how to use genetic engineering technology (President’s Commission, 1982). In a 2010 report on the ethical issues of synthetic biology, the Presidential Commission for the Study of Bioethical Issues identified “public beneficence” as the first of five “ethical principles” that should be used to assess synthetic biology and other emerging technologies (Presidential Commission, 2010). For decades, US regulation of crops produced using genetic technologies has focused on questions of safety to consumers (under regulations enforced by the US Food and Drug Administration), possible harms to other crops or plants in the environment (regulated by the US Department of Agriculture) and the safety for humans and the environment of any pesticides that the plant may be engineered to produce (under regulations enforced by the US Environmental Protection Agency).

A second set of questions turns attention away from defining the potential human benefits and harms to discussions about who will benefit or be harmed and who will make decisions about genetic engineering. In its 1982 discussion of human genetic engineering, for example, the President’s Commission addressed parents’ rights and responsibilities to make decisions about how genetic engineering might be used on their children, a general societal commitment to equality of opportunity, and to “a more basic question about the distribution of power: Who should decide which lines of genetic engineering research ought to be pursued and which applications of the technology ought to be promoted?” (President’s Commission, 1982). The Commission argued that, in most cases, the public could rely on “the judgment of experts in the field” (President’s Commission, 1982). However, in the Presidential Commission’s 2010 report on synthetic biology, the thinking had changed: The Presidential Commission argued for the “intellectual freedom and responsibility” of experts in the field, but also insisted on “justice and fairness” in “the distribution of benefits and burdens across society,” and it called for a principle of “democratic deliberation.” The 2010 report argued that because biotechnology would affect the public, the public should participate “both in the development and implementation of specific policies as well as in a broader, ongoing national conversation about science, technology, society, and values” (Presidential Commission, 2010).

Third, and finally, the arc from the President’s Commission in 1982 to the Presidential Commission in 2010 reveals a set of questions that are less easily articulated but are sometimes very deeply felt and have often been important in the public’s reception of genetic technologies. The central theme in these questions is the possibility that some ways of using genetic technologies conflict with underlying moral norms that are implicit in how human beings understand the world, including their own nature and relationship to the rest of the world. In 1982 the President’s Commission considered, and dismissed, a variety of objections to the very idea of “splicing life,” such as that it would usurp powers properly left to God (p. 53) or would constitute an “arrogant interference with nature” (p. 55). In 2010, the Presidential Commission agreed that engineering a genome is not intrinsically wrong: “After careful deliberation, the Commission was not persuaded by concerns that synthetic biology fails to respect the proper relationship between humans and nature” (p. 139). It allowed, however, that the use of that power should adhere to a principle of “responsible stewardship,” and it elaborated this principle as a responsibility to be good “stewards of nature, the earth’s bounty, human health and well-being, and the world’s safety” (p. 123). This way of talking about stewardship leaves some room for asking questions about the human relationship to nature: Although genetic engineering can be consistent with social standards for the human relationship to nature, using it to destroy significant natural phenomena might not be. Moreover, it might not be responsible even if the destruction of those natural phenomena were consistent with human health and well-being.

All three of these broad kinds of value considerations are raised by research into gene drives. There are significant potential benefits and harms for humans. There are also questions about who would benefit, who would be harmed, and who would be empowered to make decisions about gene drive technologies. Additionally, there are significant potential environmental benefits and harms, and how to understand the values relevant to the potential environmental outcomes can be challenging. Although other genetic technologies have raised questions about environmental outcomes, the power of a gene drive to alter an entire population or species, perhaps even to bring about the local or global eradication of a species, is a meaningful expansion of the human capacity to alter the shared environment (Esvelt et al., 2014; Oye et al., 2014; Caplan et al., 2015; Webber et al., 2015). It raises questions about both public health and about the human relationship to nature.

POTENTIAL HUMAN BENEFITS OF GENE DRIVES

The primary rationale for pursuing research on gene drives is the hope that it might produce human benefits. The potential human benefits envisioned in the case studies presented in Chapter 3 will be significant to many people. The potential public health benefits are particularly promising, but agricultural benefits may also be possible. Given the early stage of the research, as-yet-unrealized benefits may become evident as the science develops. For many researchers, the possibility of uncovering new kinds of benefits and of gaining new scientific insight itself can be important motivating factors.

Potential Public Health Benefits

Creating gene drives in mosquitoes to combat infectious diseases like dengue and malaria (Case Studies 1 and 2) holds potential public health benefits, particularly the control of arthropod vectors, such as insects and ticks. Case Study 1 illustrates the potential use of gene drives to prevent mosquitoes from transmitting dengue, a virus that occurs predominately in urban environments throughout the tropics. Dengue can also occur in rural and temperate zones, typically due to introduction by travelers from dengue-endemic areas. Dengue remains a major source of human morbidity worldwide, with more than 50 million cases occurring annually and 2.5 billion people at high risk of getting the disease (WHO, 2009). Another estimate places the burden at 390 million infections per year with 96 million clinical manifestations (Bhatt, 2013). More than

70 percent of people who are at higher risk of dengue infection (around 1.8 billion people) live in Southeast Asia and the Western Pacific region (WHO, 2009).

There are currently no curative treatments for dengue. However, in April 2016, the first ever dengue vaccine, Dengvaxia (CYD-TDV) by Sanofi Pasteur, was approved by the World Health Organization for use in endemic countries. Strategies using Wolbachia infected Aedes aegypti mosquitoes to reduce their populations or cause refractoriness to dengue infection are being evaluated (Dobson et al., 2002; Joubert et al., 2016); however, to date, the prevention of dengue has relied on ultra-low volume spraying of insecticides and removal of Aedes aegypti breeding sites, which are typically human-made containers. These strategies are laborious and typically reactive rather than proactive (Achee et al., 2015). Resistance to insecticides among targeted species is also challenging the efficacy of currently available chemicals. In addition, because dengue disease alternates between high and low incidences depending on the year and season, and because of the potential serotype interaction and co-circulations, predicting and therefore preventing possible dengue epidemics is extremely complex. Given these challenges, a gene drive could, theoretically, provide enhanced sustainability for disease prevention, because repeated mosquito releases may not be required. A gene drive that suppresses the mosquito population might also provide a broader health benefit to human populations, since Aedes aegypti also serves as a vector for a range of other viruses responsible for human disease, including yellow fever, West Nile, chikungunya, zika, and eastern equine encephalitis. A suppression drive would also lead to a reduction in nuisance mosquito biting.

Case Study 2, on human malaria, describes a gene drive intended to prevent mosquitoes from transmitting the protozoan parasite that causes malaria, a major cause of human illness and death worldwide. Malaria occurs predominately throughout the tropics, but it can also occur in temperate zones, typically when travelers visit areas where malaria is present and bring the disease home with them. In 2013, 198 million cases of malaria were estimated to have occurred, leading to 584,000 deaths (WHO, 2014). Most of these cases occurred in sub-Saharan Africa where the species of parasite responsible for severe disease, Plamodium falciparum, is most prevalent. Ninety percent of global malaria deaths occurred in Africa, with children under the age of five years accounting for 78 percent of deaths (WHO, 2014).

Human malaria infections can be cured using drug therapy, but therapy requires that the parasite be detected and that the infected person have access to health care. These requirements can be extremely challenging in many settings where malaria is endemic. In addition, the parasites have developed resistance to many first-line drugs. Insecticide treated bed nets, larval source management, and indoor residual spraying are strategies for preventing transmission, but they require organized campaigns and resources. Moreover, malaria carrying mosquitoes can develop resistance to the chemicals used in currently available insecticide treated bed nets and indoor residual spraying programs, making control difficult. Malaria vaccines are under development and have shown promise, but will take many more years before they become fully effective, scalable for use and approved for wide application. The possible benefits of a gene drive that prevents mosquitoes from transmitting malaria would, in theory, include an impact on morbidity and mortality caused by disease, a reduction in nuisance mosquito biting experienced by inhabitants, and a sustainable approach to delivering an intervention within remote communities where resources may be limited and efforts for disease control most challenging.

Although these case studies are particularly prominent examples of how gene drives might be used to advance public health, a number of other, similar uses of gene drives have been envisioned. These include proposals to develop a gene drive to modify deer ticks so that they cannot transmit the bacterium Borrelia burgdorferi, which causes Lyme disease (Pennisi, 2015b), and a gene drive to eradicate the parasitic flatworms that cause Schistosomiasis (Esvelt, 2016). Other possible uses of gene drives to prevent infectious disease are likely to emerge. The news in 2016 that Zika may pose a surprising and exceptionally significant public health threat shows that potential uses of gene drives may have a very great sense of urgency. Given the fear prompted by

such threats, it may sometimes be difficult to make a reasoned decision about whether a gene drive provides a good possible solution.

Potential Agricultural Benefits

Agricultural uses of gene drives are a second significant source of human benefit. For example, gene drives might turn out to be useful for controlling some weeds, a possible use explored in Case Study 6. As Palmer amaranth has developed resistance to glyphosate, it has become the most economically detrimental weed of cotton in the American South. The weeds compete with crop plants for water, light, and nutrients, resulting in lower yields. They can also become stuck in harvesting equipment, slowing production. The benefits of a gene drive that restored Palmer amaranth’s susceptibility to glyphosate could include improved crop productivity and economic gains for farmers.

Agricultural uses of gene drives in low- and middle-income countries could have a significant impact on human welfare. If it were technically feasible, a gene drive that limited the germination of witchweed (genus Striga) could boost the production of rice, corn, millet, and other cereals in developing countries. Crop damage from Striga, a parasitic plant that penetrates the roots of the host plant and saps nutrients, is particularly extensive in Africa and Asia. In Africa, one species (Striga hermonthica) alone is responsible for $10 billion per year in crop losses (Pennisi, 2015a). Alternative solutions may be possible, including the development of witchweed-resistant crops, but the economic effect of witchweed remains extensive.

The Value of Science and Innovation

Because research into gene drives is still at a very early stage, a definitive account of the benefits they might generate is not yet possible. The benefits envisioned so far may not yet been adequately understood, and the technology might, as it develops, lead eventually to uses that cannot yet be foreseen. In discussing the technology’s likely effects, it is therefore important to be cautious about any one way of articulating and framing its likely outcomes. In science, one line of research tends to lead to still other possible lines of research. The work that goes into developing one technology can present possibilities for yet other technological developments. This is part of the potential benefit of developing organismal models, such as the zebrafish (see Case Study 7) to study gene drives, and to explore their applicability to other vertebrates. The possibility that research will tend to foster further, as-yet-unknown scientific advances is itself a significant category of benefit.

The benefit of facilitating science raises some issues that are different from those of public health and agricultural applications. Like those applications, the benefit of basic science may be ultimately grounded in a belief that the work will lead to tangible improvements in public health, agriculture, or other areas. But the benefit would be indirect, open-ended, and hypothetical.

Additionally, the capacity of research on gene drives to foster advances in science and technology might also be considered valuable for a more immediate and less tangible reason. It may be rooted, to some degree, in an intrinsic value sometimes given to knowledge, understanding, and innovation. To possess knowledge is to have a belief that is not only true but justified by evidence and reason. To gain understanding is to develop an overall picture of the thing one understands, putting different pieces of knowledge together and critically reflecting on their relationship to each other. Innovation is valuable in good part, of course, because it often leads to economic benefits, but it may also be valued in itself: innovation puts understanding to work in the world in ways that may reflect creativity, diligence, planning, and leadership. Knowledge, understanding, and innovation therefore require and display capacities and virtues that are sometimes considered to make humans special, and they may also give one a special power in relation to the world. Finding intrinsic value in knowledge is also very much part of the tradition of science: Although this view of the value of science often goes unspoken, its significance is readily

apparent (Sarewitz, 1996). It is probably the chief argument in support of sending probes to distant parts of our solar system and searching the galaxy for other solar systems. In biology, too, value is often attached to relatively arcane investigations that are unlikely to have an immediate impact on human welfare—such as trying to learn how life formed, how different living things came to be, and how long-extinct living things once lived.

The value that many people find in knowledge, understanding, and innovation is not always an overriding consideration in deciding whether to conduct research. That value may be outweighed by concerns about potential harms. However, it is a significant consideration, both in private life and in public decision making. From the standpoint of a scientist who decides to pursue the work described in Case Study 7, at least part of the rationale is likely to be a belief that it is intrinsically worthwhile. If the risks of research are minimal, then the perceived intrinsic value of the research, together with the possibility that it will lead to as-yet-unanticipated benefits, is likely to provide a very strong rationale for proceeding with basic research.

POTENTIAL HUMAN HARMS OF GENE DRIVES

Many of the possible harmful effects of gene drives have to do with environmental outcomes, which are considered in the next section. However, some gene drives pose potential harms to human well-being if they do not function in field release as expected. Additionally, human harms might result from accidents in the laboratory (concerns about biosafety) or from any potential that gene drive research might have for deliberate misuse (concerns about biosecurity).

The release of gene-drive modified organisms has the potential to generate public health harms. One theoretical example is a mosquito modified so that it could not host the dengue virus that becomes a more susceptible host to another existing or new virus that harms human health. Another hypothetical outcome of this scenario is that the dengue virus might evolve a new phenotype that poses a slightly different hazard from the one that the gene drive was meant to suppress. A gene drive that suppressed rather than modified the host organism might have other effects. The removal of an entire species, such as a mosquito, could have effects on other organisms in the ecosystem, which could in turn lead to unwanted changes, such as an increase in the population of another insect disease vector as it fills the ecological niche opened by suppression of mosquito populations.

Gene drives developed for agricultural purposes could also have adverse effects on human well-being. Transfer of a suppression drive to a non-target wild species could have both adverse environmental outcomes and harmful effects on vegetable crops, for example. Palmer amaranth in Case Study 6 is a damaging weed in the United States, but related Amaranthus species are cultivated for food in in Mexico, South America, India, and China.

Deciding whether to go forward with a field release of a gene-drive modified organism will require a reasonable level of assurance that the possible harms have been identified and studied and that they are outweighed by the potential benefits, where the characterization of the potential outcomes involves both their significance (or severity) and their likelihood. The likelihood may depend not only on technical aspects of the gene drive and how it is expected to function within the organism, but also on environmental and societal issues. A positive balance of potential benefits over potential harms might mean that the harms are not very severe, that their likelihood of occurring is tolerable, that a reliable mitigation strategy can address potential harms, or perhaps that the potential harms are non-negligible but are still outweighed by the possible benefits. There are also trade-offs to consider (Finkel, 2011): The potential outcomes of a release will need to be weighed against the potential outcomes of alternative solutions to the problem for which the release is proposed, and also against the outcomes of doing nothing—which could amount to very great harm given an enormous, immediate, and highly certain public health problem. A gene-drive modified organism may offer a technological way of addressing a problem that was initially generated by larger societal and environmental problems, and if the technological solution provides a way of

avoiding the larger issues, it may have the effect of perpetuating them. On the other hand, if the immediate problem is very serious, then a comparatively quick, targeted solution to it might be attractive anyway. Identifying the potential harms of a proposed field release will require case-by-case analysis and include use of a structured, systematic, and reasoned methods to investigate and model the possible outcomes, making use of everything known about the relevant species and ecosystems. Cost-benefit analysis may also be useful for modelling the possible outcomes of regulatory or policy decisions about gene drive research and use.

Although structured decision making tools for examining and modelling outcomes can provide useful guidance, they may not always be decisive given the questions of value on which they depend. While the outcomes might be tangible human interests, identifying them, articulating their significance, and determining the tolerable level of uncertainty about them are matters of value and may remain contested. The probabilities assigned to outcomes may also leave some uncertainty about how a proposed release will go. Moreover, some theoretical harms—such as the possibility that a pathogen might adapt to a gene drive and produce a new and worse phenotype—are hard to predict. How much certainty is needed in order to declare that the outcomes have been adequately studied is a further question of value. Resolving uncertainty takes time, and prolonging the analysis can sometimes prolong the problem. A society might opt for a more or less precautionary position with respect to uncertainty, declaring either that the uncertainties must be minimized as much as possible or that some uncertainty is acceptable when there are significant potential benefits (Kaebnick et al., 2014).

Some of the outcomes about which people may express concerns may be scientifically implausible. This can be a result of the complex ways in which technical information is generated and communicated in a society, particularly when it is connected to difficult value questions, and because of challenges of perception that are associated with some kinds of risks. Some kinds of potential harms are likely to be seen as more alarming than others for reasons that are independent of the degree or likelihood of damage (Slovic, 1987). Structured decision making tools may not assess outcomes in a way that is satisfying to those who are particularly alarmed by those outcomes.

The possibility that public attitudes about harms may seem irrational at times does not mean that public attitudes can be set aside. Both humility and prudence require deference to the public perceptions and understanding of research. Since benefits and harms are matters of value, it is impossible to say exactly which outcomes should be considered benefits, which outcomes should be considered harms, and how much weight they should be given without incorporating the publics’ own views. Different publics may identify and gauge relative benefits and harms somewhat differently. Some members of the public believe that scientists irrationally overestimate their ability to produce the benefits they propose. There is likely to be broad agreement that eliminating malaria and dengue would be good, but there might be differences of opinion about how that benefit compares against potential harms of gene-drive modified organisms, either to humans or to the environment. Moreover, a society that is affected by a disease may place a much greater value on eliminating that disease than would a society where the disease does not occur. There could also be reasonable differences of opinion about how much confidence we need in predictions about outcomes in order to decide whether to pursue a potential benefit (and incur some potential harms), or to take precautionary measures against the potential harms (and constrain progress toward the benefit). Issues of risk assessment, risk, perception, public engagement, and precaution are addressed further later in this chapter, as well as in subsequent chapters.

Dual Use Concerns

Research that might be put to deliberately malicious uses is sometimes known as dual use research (NSABB, 2007). The dual use potential of gene drives is not the same as that of other lines of research in synthetic biology. In principle, synthetic biology techniques can be used to

synthesize pathogens or modify them in ways that make them more dangerous, and gain-of-function research on influenza viruses and other pathogens can be used not only to learn how to defend against those pathogens but also to create more potent ones (Presidential Commission, 2010). Gene drive technologies would be inapplicable to bacteria and viruses (because they are limited to organisms that reproduce sexually), would not be effective on humans (because of humans’ long generation times), and might be of limited effect on crops and livestock (because their reproduction is sometimes controlled in ways that would hinder propagation of a gene drive). Dual use potential is not necessarily a reason not to pursue the research. One common argument for pursuing research into the synthesis or modification of pathogens is that the best defense against dual use is a good offense: the research provides a basis for defending against those pathogens (Fauci et al., 2011). Dual use concerns about gene drives are also discussed in Chapter 8.

VALUES RELEVANT TO POTENTIAL ENVIRONMENTAL EFFECTS

There is a widespread sense among researchers and commentators that the capacity of gene drives to genetically alter a wild population, and potentially an entire species, represents a new type of ethical environmental challenge (Esvelt et al., 2014; Caplan et al., 2015; Charo and Greely, 2015). There are significant potential environmental benefits but also legitimate questions about potential environmental harms. The values attached to the potential environmental outcomes may be understood in different ways, some of which are not universally accepted. As a result, how they are to be weighed against each other and alongside public health and agricultural outcomes is very complicated.

Potential Environmental Benefits

Applications of genetic technologies in agriculture can lead to the accidental alteration of wild populations (Lai et al. 2012; Ellstrand et al., 2013). To date, no agricultural application has incorporated a mechanism specifically designed to force a change through a population as would a gene drive. The closest analog to what gene drive technologies can accomplish in the shared environment is the use of genetic engineering to confer beneficial traits to threatened species, with the hope that, if genetically altered organisms were released in the environment, the engineered traits would drive through the population under the “natural” pressure of evolution. This kind of application is known as “facilitated adaptation.” One example of facilitated adaptation is the effort now under way to impart resistance to chestnut blight to the threatened American chestnut through the transferring of genes from wheat, grape, Asian chestnuts, and other organisms (Newhouse et al., 2014).

Case Study 3, which describes a gene drive to prevent mosquitoes from transmitting avian malaria, highlights considerations for conserving threatened or endangered species. Avian malaria occurs throughout the world and on almost every continent, impacting several hundred species of birds. Parasites of the genus Plasmodium are responsible for pathogenicity, mass mortality, population declines, and even extinctions of many bird species (van Riper et al., 1986; Valkiūnas, 1993). In Hawaii, the fossil record shows that many events in the past have affected the size and diversity of populations of native birds. Hawaii’s native birds live in a fragile habitat where any disturbance, from human settlement and hunting to diseases, leads to a drastic reduction of the species diversity. Avian malaria, caused by Plasmodium relictum and transmitted by Culex quinquefasciatus mosquitoes, is widely recognized as the greatest current threat to the Hawaiian avifauna, especially honeycreepers (Warner, 1968; Freed, 1999; van Riper and Scott, 2001). A wave of extinctions of native birds during the 1920s and 1930s has been attributed to avian malaria, and today native birds living at elevations below 1,500 meters continue to be at risk from malaria (van Riper et al., 1980; Goff and van Riper, 1981). In contrast, malaria has minimal impact on the survival of non-native birds, and because mosquitoes are rare at altitudes

above 1,500 meters, higher elevations are hypothesized to be protective to native forest birds (Samuel et al., 2015). If a gene drive were developed either to reduce populations of the mosquito vector, or to make them refractory to infection with the malaria parasite, the susceptible birds might begin to repopulate the higher altitudes and reintroduce themselves into original ecosystems of lower elevations.

Aiding the threatened honeycreeper species through introduction of a gene-drive modified mosquito, for example, could potentially prevent the bird’s extinction; however, such an intervention could also be expected to have unintentional impacts on the ecosystem as well as on the human population. For example, since the honeycreepers are nectar-feeders, there may be shifts in plant species biodiversity if the bird population is reintroduced into areas where they are currently not found. Competition with other birds for similar nesting and feeding sites could also occur, thereby modifying the diversity of other fauna.

Similar environmental benefits are at play in Case Study 4, which describes gene drives to suppress non-native rodent populations on remote islands such as are found in the Pacific. Mice and rats have been inadvertently introduced to these islands by maritime travelers with frequently catastrophic effects on native species and ecosystems. These effects are sometimes a result of direct predation by the rodents on the various native species, but they may also result from habitat alteration, competition for food, and other ecosystem interference.

A gene drive to control nonindigenous rodents is attractive in part because of the many challenges to control them using alternative methods. Initial efforts at population control involved the use of rodenticides, usually anti-coagulants. First-generation compounds, such as warfarin, had to be administered in high concentrations over multiple doses. They have now been replaced by second-generation compounds such as the odorless and tasteless toxicant Brodifacoum (Mensching and Volmer, 2008). The cost of administering these compounds is estimated to be in the millions of dollars due to expenses associated with their regulation, dispersal method, and actual inherent cost of the toxicant (Meerburg et al., 2008; Williams, 2013). Rodents can sometimes evade the chemicals. Moreover, the chemicals can result in a comparatively painful death for the affected rodents (Gould, 2015) and they may adversely affect the health of humans, other animals, and the overall ecosystem (Lorvelec and Pascal, 2005; Witmer et al., 2011).

Mechanical control methods, such as trapping, are not considered suitable to eradicate a rodent population, although they can be useful in conjunction with other methods. Two types of traps currently exist and are categorized based on the outcome to the rodent (Hygnstrom and Virchow, 1992; Witmer and Jojola, 2006). Kill traps such as snap traps are effective only on a small scale, while the effectiveness of glue traps and snares is questionable given the animal’s ability to avoid them (e.g., jumping over them) (Witmer and Jojola, 2006). Kill traps also call into question the welfare of the animal and whether this method is in fact humane. Live traps are a non-lethal, arguably more humane, but expensive alternative to kill traps. While live traps tend to be successful for capturing rodents, the trapped rodents must then be relocated, which poses a further set of problems (Hygnstrom and Virchow, 1991; Witmer and Jojola, 2006). Collectively, these mechanical methods cannot discriminate between target and non-target organisms (Lorvelec and Pascal, 2005), and so their use raises similar issues to that of chemical toxicants. In addition, traps require considerable human labor and monitoring, and may cause injury to the workers who place them. Finally, animals are able to adapt to these traps, which can be damaged easily by people or animals (Witmer et al., 2011).

Biological controls of invasive rodents include predators, parasites, and other disease-causing agents that act to limit the population. One of the considerations in using this type of method is whether the introduced organism would itself become invasive following its placement in an environment to which it is not endemic. Several unsuccessful applications of this method have taken place in the past. The introduction of rabbits into Australia in the late 1800s (Garden, 2005) required subsequent efforts to control their substantive, unexpected, population growth (Fenner, 1983; Saunders et al., 2010). The introduction of the cane toad to control agricultural pests of Australian sugar cane (Weber, 2010) had a similar, unexpectedly complicated outcome.

The cost of this type of intervention will vary depending upon the targeted organism of interest and the biological control agent being introduced.

Other methods currently being explored to control non-native rodent populations take advantage of the process of RNA interference (RNAi), in which double-stranded RNAs might be delivered to the rodent to silence the expression of genes essential for life (Gao and Zhang, 2007). Technical issues associated with this technique include actual delivery of double-stranded RNAs, their inherent stability and thus persistence of inhibition, the concentration required to eradicate a species, their mechanism of spread, and their potential biosafety risks. Proof-of-concept, however, has been demonstrated with sea lampreys (Heath et al., 2014). Another possible method is the induction of autoimmune infertility, achieved through the introduction of a virus expressing proteins that elicit an immune response, and therefore target the fertilization process and prevent formation of the zygote (Chambers et al., 1999). This technique would reduce the target population, but challenges would remain with respect to the administration of the virus at the appropriate time in the rodent’s life-cycle and the numbers of rodents to be infected (Jacob et al., 2008). It would also be necessary to ensure that infected rodents mate with one another as opposed to untreated rodents (Biotechnology Australia, 2001). Finally, in some instances it may not be possible to eradicate an invasive rodent population because doing so is cost-prohibitive, because of the location and topography of the land limit access, because the presence of humans would damage the ecosystem, or because of others harms posed to the area.

In short, there are many ways to try to rid an island of a nonindigenous rodent population and many reasons those methods are likely to fail (Gould, 2015). A gene drive that successfully affected the entire population may then appear particularly attractive. A gene-drive modified rodent could be released on an affected island with relatively little other human labor required, and perhaps at relatively low cost.

Potential Environmental Harms

The potential environmental release of gene-drive modified organisms will raise questions about possible harmful environmental outcomes. Case Studies 1 and 2, for example, the potential consequences for other species of reducing the mosquito population may need to be considered, especially given the large geospatial scale at which the gene drive would likely be implemented. Some highly valued species may depend on the mosquito population, even in places where the targeted mosquitoes are nonindigenous. As previously noted, a gene drive to modify or eliminate Palmer amaranth in the American South, considered in Case Study 6, could affect closely related wild species as well as to food crops in other parts of the world. Spotted knapweed, the target of a gene drive considered in Case Study 5, is pollinated by insects, including butterflies; so as a result, there may be unintended environmental consequences that would require further research before such a gene drive is pursued.

Restoring a bird species as in Case Study 6, may also have unexpected environmental consequences that need to be considered. An ecosystem can sometimes adapt to human alterations in ways that cannot be reversed without bringing about still more unwanted changes.

Using gene-drive modified organisms to bring about environmental changes is analogous in some respects to the past attempts to use biological controls to fight pests. As the history of unfortunate experiences with biological controls suggests, adequate assessment of the environmental harms of a proposed release will require careful, case by case analysis. Structured assessment tools for carrying out this analysis are discussed at length in Chapter 5. One example of complex considerations that must be examined is whether the invading species plays a critical role in the ecosystem. For example, Tamarix (salt cedar) species have overtaken many riparian communities in the American Southwest, often as hybrids that are not found in their native ranges (Schaal et al., 2003). In the process, Tamarix has displaced native plants as the breeding habitat for approximately 50 native bird species (Sogge et al., 2008); and hence suppression of this invasive species could have unintended consequences for native birds. Remarkably, Tamarix

also alters the salinity of soil, which negatively affects the ability of native plants to re-colonize (Zavaleta et al., 2001), so sites must be restored prior to reintroduction of native species. Assuming that the technological obstacles of transformation and targeting could be overcome, gene drives to suppress Tamarix populations would likely spread slowly, because they are long-lived perennials, commonly spread vegetatively as well as sexually, and may have substantial population substructure, as is typical of asexually spreading organisms (Sakai et al., 2001). Tamarix nonetheless illustrates a long-standing complication: the eradication of an invasive plant species may lead to unexpected consequences, such as the loss of habitat for native species or even the establishment of a second, more resilient invasive species (Zavaleta et al., 2001).

Adequately assessing the environmental harms of a proposed release of a gene-drive modified organism also requires extensive engagement with those who might be affected by the release. As with the potential benefits, the harms cannot be adequately identified and weighed without that input. If the release is contemplated for a low- or middle-income nation, it is very important that people in developed countries avoid imposing their own views about what the benefits and harms are and how they should be weighed.

Intrinsic and Anthropocentric Values

Similarly, the public must be engaged in order to identify and weigh relevant environmental outcomes appropriately. In the applications described in Case Studies 3 and 4, for example, it would be important for researchers and project organizers to ask exactly why and in what way it is a benefit to rid an island of avian malaria or nonindigenous rodents and thereby try restoring a native population. Similarly, it is important to think about how the environmental harms should be understood. Different people may understand and value environmental outcomes in very different ways. Some people evaluate environmental outcomes in terms of human outcomes: An environmental harm is an environmental effect that has negative repercussions for human health and welfare, and an environmental benefit is an outcome that fosters desirable human outcomes. This way of thinking about environmental outcomes is at work when people speak of “ecosystem services,” for example. Ecosystems perform a wide variety of functions that are vital to humans, communities, and societies, ranging from generating food to cleaning water to providing opportunities for recreation.

On the other hand, some people evaluate environmental outcomes not only in terms of outcomes for humans but also in terms of their effects on the environment itself—for example, the effects on biodiversity or on the richness and resilience of ecosystems, aside from ways in which biodiversity and ecosystem resilience are beneficial to people. This way of thinking about environmental outcomes is often at work when people express concern about endangered species. For example, although endangered species are sometimes valued for their ecosystem services, or for their economic or medical usefulness, they may also be considered valuable in and of themselves, because they are part of the shared environment. To see environmental outcomes as valuable in and of themselves is to think of naturally occurring environmental phenomena as intrinsically valuable and to adopt a preservationist stance toward those phenomena. Views about the intrinsic value of the natural world probably also play a role in efforts to protect “wild” places, such as through the US Wilderness Act, the Wild and Scenic Rivers Act, and the national park system and other federal and state preserves, and such views may also have some role in the efforts to pass the US Clean Air and Clean Water Acts.

Gene drives’ unique mode of altering the shared environment poses special challenges, and perhaps also special opportunities, for those who take a preservationist stance toward the natural world. Genetic engineering techniques in general are sometimes perceived as intrinsically unnatural (President’s Commission, 1982; Nuffield Council on Bioethics, 2015). Aside from whether the gene drive itself is perceived as unnatural, gene drives could have significant effects on particular organisms and ecosystems, such that the perceived naturalness of those phenomena, and of the places where they are found, could be substantially changed. More broadly, gene drive

technologies raise special questions shared by many environmentalists (although not all) about the ever greater powers that humans are developing to alter the natural world. From this perspective, gene drive technologies might be seen as shifting the balance of power in significant new ways insofar as they may let humans overrule some “natural laws,” such as Mendelian rules of inheritance and Darwinian conceptions of survival of the fittest. They may appear, to some people, to reflect the same human hubris, the same overeagerness to control nature and the same overconfidence that we could succeed at it, that have created many environmental problems. In the case studies considered above, the clearest human benefits have to do with such human needs as avoiding disease and providing food, but perhaps, at some point in the future, gene drives could be developed in which the benefits involve human preferences and fancies. Perhaps gene drives could be used to suppress or modify populations of insects merely on the grounds that they are nuisances, for example. Following the news in 2016 that Zika virus, transmitted by the mosquito Aedes aegypti, might present a significant public health threat, some discussion appeared in the popular media about whether mosquitoes in general should be eliminated—those that are annoying as well as those that pose public health threats. In principle, some might also propose to use gene drives to make wild species more aesthetically pleasing. Zebrafish genetically engineered to be fluorescent are now sold as pets, and kits are available on the Internet that allow customers to produce mustard plants engineered to glow faintly in the dark.¹ In theory, gene drives could allow individuals to propagate such traits in wild populations.

Questions about how to define “nature” and how to understand the value attached to nature raise a number of difficult philosophical and social problems (Cronon, 1995; Soper, 1995; Sagoff, 2003; Thompson, 2003; Marris, 2013; Kaebnick, 2014; Nuffield Council on Bioethics, 2015). Skeptics of concerns about nature argue that no entirely natural phenomena exist any longer, for example, and that human intervention into nature is already common and sometimes (in medicine, for example) widely accepted. In the long-running debates about genetically engineered crops and livestock and about the use of genetic technologies to treat or perhaps even to enhance human beings, skeptics have also argued that concerns about nature are based on religious, superstitious, or personal psychological reactions that are not easily defended in the kind of public discourse that should support public policy making. Similarly, skepticism about “nature” might itself reflect corporate and other interests in the activities and technologies that are sometimes seen as unattractive alterations of nature.

These debates about nature will continue, and gene-drive modified organisms may be a significant new moment in them. In a survey of the use of new genetic technologies on nonhuman organisms, bioethicists Alta Charo and Henry Greely have observed, for example, that some people “decry the ‘end of nature’ and the loss of the sense of a reality outside ourselves, whether created by God or by nature, [and] feel impoverished by the increasing human footprint on the world…. Even those not reflexively against ‘unnatural’ changes through biotechnology might find something unsettling about altering the biosphere with uses that are recreational, whimsical, or even Disneyfied” (Charo and Greely, 2015). On the other hand, those who resist genetic engineering because they see it as “unnatural” have to confront the possibility that gene drives might sometimes be very valuable tools for conservation, as illustrated in Case Studies 3, 4, and 5 (Jennings, 2015; Webber et al., 2015).

The intrinsic value that many find in the natural world presents an interesting comparison to the value that many find in knowledge, understanding, invention, innovation, and industry. In some ways, these two stances may be similar. Like the value found in knowledge, understanding, and innovation, concerns related to the intrinsic value of nature, and how to compare those concerns to more tangible human benefits and harms, will be contested in debates overpublic policy. The two kinds of value also contrast with each other to some degree; finding value in nature

___________________

¹Experience the Glo!, https://www.glofish.com; Natural lighting without electricity, http://www.glowingplant.com.

seems to call for adjusting human activity in order to accommodate nature, while finding value in knowledge, understanding, invention, innovation, and industry seems to celebrate the alteration of nature to support human activity. On the other hand, it may be possible for an individual, community, or society to share both values to some extent. Perhaps, each stance even implicates the other: Preservation of natural phenomena can be aided by appropriately directed efforts to understand and intervene in the world, and human activity in the world depends on trying to accommodate the natural world.

This report does not side with any particular way of understanding these issues and does not resolve them. They are left here as open questions, and are part of a growing and heated debate among environmentalists about the values that underpin environmentalism. Historically, in the United States, some environmentalists have leaned toward preservationism, tracing their thinking back through Aldo Leopold’s “land ethic” to John Muir’s call to protect Yosemite and Henry David Thoreau’s celebration of wildness and of places that exhibit untrammeled wildness and limited human impact. Others have leaned toward thinking of natural phenomena in terms of ecosystems services—a stance that is often called conservationist and traces back to Gifford Pinchot and the creation of the US Forest Service (Rich, 2016). Recently, some environmentalists have proposed that these two sides could be and should be bridged with a third, middling position, perhaps a “gardening ethic” that values alteration of nature and accommodation of nature simultaneously (Pollan, 1991; Marris, 2013; Rich, 2016). The evolving debate about the desirable human relationship to nature is also reflected in the idea that the earth has entered the Anthropocene, defined as an epoch in which human influence in nature will leave a geologic record (Waters et al., 2016). Passing this boundary is seen sometimes as evidence of the need for greater restraint toward nature, and sometimes as showing that humans should accept a strongly interventionist role in nature, for they are in that role whether they like it or not. However these questions about the value of nature and the proper human relationship to nature are understood, they are likely to be very important in the public’s response to gene drive technologies and in decisions about how those technologies should be developed and used, given the prospect that gene drives could be a tool for modifying wild species to suit human needs, perhaps to bring about their extinction, perhaps to alter them to suit aesthetic preferences. Moreover, different publics will undoubtedly frame these questions differently. The views about nature that have been described here are found predominantly in Western cultures, and probably particularly in the United States, since European views of “nature” are more likely than American views to see natural phenomena as part of agricultural contexts—and to see agricultural phenomena as part of “the shared environment” (Soper, 1995).

CONCERNS ABOUT JUSTICE

In addition to questions about various kinds of potential benefits and harms, research on gene drives presents questions of justice. Questions of justice differ from questions about potential benefits and harms in that they are more about who than what: They are about who would be affected by the benefits and harms, who will be able to conduct research into gene drive technologies and study the release of gene-drive modified organisms, and who will make the decisions about whether to pursue the benefits and risk the potential harms. They are questions about the distribution of potential benefits and harms, about liberty, about the nature of legitimate decision making for matters affecting the public. They are about how communities and nations are affected by gene drive technologies, the ability of scientists and funders to undertake the research, and the relationship of citizens to nations and of nations to each other.

Some of the envisioned uses of gene drives are motivated in large part by concerns about justice. Part of the value of Case Study 2, for example, is that the people who are most seriously affected by malaria are in low income countries whose health (and other) needs have often been overlooked by wealthier, more developed countries. Cures for malaria have been available for a long time, but they are seldom available to the people who need them most. The most at-risk

countries, where malaria is a very significant burden for communities and governments, often have limited health care systems and little capacity to fund or conduct medical research. In sub-Saharan Africa, where the burden is greatest, diagnosis and treatment alone, excluding prevention strategies, are estimated to have cost about $300 million per year since 2000 (WHO, 2014).

In several of the case studies, concern about the distribution of benefits, set against the history of the relationships between high-income countries and lower-income countries, is part of the reason to move forward with the research. However, concerns about justice can also present reasons to be particularly cautious about a gene-drive modified organism. In Case Study 6, the gene-drive modified Palmer amaranth envisioned to suppress the population might be beneficial in the United States, where Palmer amaranth is a pest, but be harmful if it were to make its way to Mexico, South America, India, and China, where related Amaranthus species are cultivated for food. In such a case, a comparison of the benefits to the harms involves not only an understanding of their magnitude and likelihood, but also of the relative life circumstances of the people who would experience them and perhaps even of the histories and relationships of the countries in which those people live. Similarly, some societies could be understandably cautious and give researchers little latitude to proceed considering release of a gene-drive modified organism that has been developed by researchers from high-income countries, that would be proposed for release in a low-income country, and whose benefits and harms cannot be fully known in advance of the release. For Case Studies 1 and 2, any harms from the release of gene-drive modified mosquitoes are likely to be borne disproportionately by low- and middle-income countries. If the research in those cases is driven by researchers and funders from wealthy countries, researchers and other decision makers may tend to underestimate or discount the risks. On the other hand, as noted earlier, the people who are immediately affected by a disease are the most likely to understand its true burden. Those from wealthy countries may tend to discount the benefits that others value.

These questions about disproportionately distributed benefits and burdens highlight the importance of the relationship between researchers and funders from wealthy nations and those in poorer countries who must live with the consequences of research in their environs. If an environmental release of gene-drive modified organisms leads to unanticipated public health or environmental harms and for which no mitigation strategy has been put in place, the researchers and funders bear a responsibility not to abandon the people enduring those harms. Withdrawing from the community can give rise to feelings of abandonment and a sense of loss (Lavery et al., 2008). In short, a strong and long-term relationship between communities and researchers is deeply important (Brown et al., 2014; King et al., 2014).

Another set of concerns about justice centers on who is involved in decisions about the development and use of gene drives. People hold a wide variety of views about justice, especially if the scope of inquiry is not limited to Western democracies, and these different views could lead to different expectations about the roles of research in society and how research should be conducted. There may be a loose consensus that benefits of research should not all accrue to the wealthy while all the harms are borne by people who are poor and powerless, but there is also some general agreement that scientists should have liberty to pursue their research as long as they do not cause harm to others. This loose consensus leaves room for meaningful disagreement, where different people could largely agree on the likely outcomes of releasing a gene-drive modified organism into the environment, but still come to different conclusions about whether the release is a good idea.

In the absence of any strategy for resolving such questions, the best course of action is to ensure that the people who could be affected by a proposed project or policy have an opportunity to have a voice in decisions about it. Experts acting alone will not be able to identify or weigh the true costs and benefits of gene drives (Kaebnick et al., 2014; Sarewitz, 2015). In other words, justice require procedures that allow both broad public decision making about the development and use of gene drives and local community decision making about specific proposed releases of gene-drive modified organisms. The ability of people in low-income countries to participate

meaningfully in decision making would be supported best not by merely engaging them in decision making but by building the capacity in those countries to conduct research that is locally valuable, regulate and provide oversight of gene drive research generally, and carry out their own decision making about its application. To ensure that capacity-building activities are not just a guise for off-loading expensive and risky research—perpetuating rather than addressing injustice—such activities need to include the development not just of technical capacity to do research but also of capacity to oversee safe and responsible research practices and decide how best to use research findings. Genuine capacity building must be understood as empowerment, and empowerment must mean that a community or country is able to act on its values rather than merely relying on values imported from elsewhere.

Selecting Sites for Field Tests or Environmental Release of Gene-Drive Modified Organisms

A special issue that arises in research involving genetically modified organisms is the selection of sites for conducting confined field trials and perhaps for releasing the organism into the environment. A variety of research publications address site selection for release of mosquitoes that have been genetically modified in ways that do not involve gene drives (Lavery et al., 2008; Brown et al., 2014). Researchers working on gene-drive modified mosquitoes and other organisms should bear in mind the recommendations from these publications, not only for guidance on matters of justice, but also for practical guidance. Site selection should be guided by many considerations, including the balance of benefits and harms, both in terms of public health and the environment and as understood in collaboration with the stakeholders in the community (as discussed above); the feasibility of examining outcomes through structured tools such as risk assessment (as discussed in Chapter 5); the feasibility of community engagement (as discussed in Chapter 6); and appropriate governance structures within the host country (as discussed in Chapter 7). It is important to be able to establish a relationship with the community stakeholders (Brown et al., 2014; King et al., 2014), learn about the community’s own understanding of its interests, establish trust, navigate the regulatory structure, and follow through on commitments made to the community (Lavery et al., 2008; Brown et al., 2014).

Environmental release of gene-drive modified organisms also raises issues that go beyond the selection of a specific location for the release. While some kinds of genetically modified mosquitoes are likely to disappear from the environment unless they are released repeatedly, gene drives are designed to propel a trait through an entire population, moving beyond any single community and crossing national boundaries as well. Deciding when and where to release a gene-drive modified organism requires attention to national, regional, and perhaps even global concerns in addition to the concerns of the local community.

Other Analyses of Gene Drives and the Issues They Raise

There is no well-developed public debate yet about gene drive research, as there is about genetically engineered organisms in agriculture. In the academic literature to date, only a few analyses have addressed at length the ethical issues raised by gene-drive modified organisms.

Commentators have been nearly unanimous that gene drive technologies might have very significant, tangible benefits in a variety of contexts, especially public health, agriculture, and environmental conservation, and they also agree that there are a variety of questions about the potential harms of gene drive technologies, both to humans and to the environment. Questions have been raised, for example, about whether engineered gene drives will have the intended effects on target organisms (Oye et al., 2014; Caplan et al., 2015), and, in particular, whether the transmission of disease might be worsened when the target organism is a vector (Benedict et al., 2008); whether gene drives might spread to other organisms (Oye et al., 2014); what effects gene-drive modified organisms might have on humans who consume them (Caplan et al., 2015); what effects they might have for other populations of organisms and for ecosystems (Oye et al.,

2014; Caplan et al., 2015; Webber et al., 2015); and what dual use potential they might have (Gurwitz, 2014; Oye et al., 2014). These concerns are most significant for possible field releases of gene-drive modified organisms, but scientists engaged in gene drive research have also recognized the importance of ensuring that laboratory work is conducted safely (Akbari et al., 2015). These concerns have not yet led any scholarly commentators to call for a halt to research on gene drive technologies, but they have led to many recommendations that would constrain and guide such research.

A number of analyses address several broad themes. One concerns uncertainty: The outcomes of gene drives are, for the time being, highly uncertain because of unresolved questions about how a given gene drive will function (for example, whether there will be off-target or pleiotropic effects, the nature of potential gene—environment interactions, and whether the gene drive could create selective pressure for yet other undesirable effects), about whether the gene drive will be transmitted to other, unintended populations of similar or different organisms, and about the overall effects of engineered gene drive mechanisms on ecosystems and humans. Recognition of this uncertainty has led commentators to recommend that research and related applications proceed only if a number of precautionary measures are in place. Among the recommendations that have been advanced are that research should be made public, with concepts and intended applications published in advance of construction and testing (Oye et al., 2014); that risk assessment should be conducted on a case-by-case basis to examine the possible outcomes of any release (Benedict et al., 2008; Oye et al., 2014); that research on a possible environmental release should occur in stages, from laboratory through preliminary trials, with each stage providing opportunities for feeding data back into decision making (Benedict et al., 2008; Oye et al., 2014; Caplan et al., 2015); and that a drive should not be developed unless mitigation methods or so-called immunizing or reversal drives are also developed (Oye et al., 2014; Caplan et al., 2015). The constraints appropriate for gene drive research are discussed in Chapters 2, 5, and 6.

Such recommendations appear to endorse a moderate degree of precaution about gene drive technologies, although the concept of precaution in scientific research is understood in various ways and is hotly contested. Often, precaution is understood as a single general principle. One widely cited formulation holds that, if preliminary scientific evidence suggests that a proposed activity poses “threats of harm to human health or the environment,” then measures should be taken to forestall the possible harms, and the activity’s proponent or proponents shoulder the burden of proof in establishing that the activity should proceed.² Some critics of synthetic biology have endorsed this formulation of precaution (FOE, 2012). Others argue that a precautionary principle could be specified in a variety of ways, giving different policy responses to the proposed action and identifying different conditions that would warrant the response (Parke and Bedau, 2004). Precautionary principles could therefore vary both in the stringency of the restraints they impose on an action and in the sensitivity of the trigger. Other commentators describe precaution not as a principle but as an “attitude” or approach that is characterized by asking that a stronger case be made for an activity, and more assurances provided about it, than in a “proactionary” approach to proposed activities (Wolf, 2014). In a similar vein, the Presidential Commission for the Study of Bioethical Issues recommended that synthetic biology be approached with “prudent vigilance,” which the commission saw as a middle-of-the-road position between a strong precautionary stance and a strong proactionary stance. In a discussion of research on genetically modified mosquitoes, El-Zahabi-Bekdash and Lavery (2010) conclude that the goals of a precautionary “mindset” can be achieved in part through community engagement, since the community may be able to provide critical insights about potential harms. Strong formulations of precaution have come under a variety of criticisms, most notably that precaution will lead to inaction (Sunstein, 2005); however, by specifying constraints that allow research to continue, the commentary to date on gene drives deflects

___________________

²Wingspread Statement on the Precautionary Principle. 1998. The precautionary principle. Available at http://www.sehn.org/state.html.

such criticisms. Further details on ways to incorporate precautionary steps into the conduct of gene drive research are discussed in Chapter 5.

Structured tools for modeling outcomes play an important role in decision making about how to use gene-drive modified organisms. As noted above, risk assessment is important in considering proposed environmental releases, and cost-benefit analysis may be helpful for informing regulatory and public policy decisions. Public examination of the costs and benefits will be particularly important if the development and use of gene-drive modified organisms depends primarily on public or philanthropic funding. Using cost-benefit analyses in a way that can support anticipatory governance presents challenges. At an early stage in a technology’s development, there may not adequate information available to compare the potential benefits and harms of using that technology or to compare those outcomes to other possible strategies for addressing a given problem. In addition, highly formal cost-benefit analysis, in which benefits and harms are estimated as sums of money, is criticized on grounds that it distorts or omits some of the public’s values (MacLean, 1998; Mandel and Gathii, 2006; Kysar, 2010; Sinden, 2015). Any intrinsic value that is assigned to wild species or to the natural environment, for example, may not be easily monetized.

The existing scholarly commentary is in agreement that gene drives might have broader environmental harms that need assessment, but the language used to express this concern varies. As discussed above, Charo and Greely consider that the environmental harms might in part reflect concerns about the extent of human impact over the natural world; indeed, what count as environmental benefits from a human perspective might nonetheless raise objections from some quarters (Charo and Greely, 2015). In examining the potential for gene drives to advance the conservation of ecosystems by eliminating invasive species, Webber et al. (2015) express the underlying value as a question of national biosecurity that should be addressed by the countries where the species in question are found. Oye et al. (2014) argue that the effects of gene drives on genetic diversity warrant consideration, although they do not discuss whether genetic diversity is valued because it may produce human benefits or for its own sake. Caplan et al. (2015) ask whether using a gene drive to eliminate a species would “upset the ecological balance,” which they suggest might override potential human benefits of the drive.

Perhaps precisely because the appropriate language for identifying, expressing, and weighing these value considerations is unclear, the scholarly commentary calls for public discussion of gene drive technologies, and it holds that this discussion should occur both at a broad, societal level and at a local, community level corresponding to the site at which a gene-drive modified organism might be released. Public engagement is usually understood in these works not merely as a process of informing the public about gene drive technologies, nor merely as a process of winning the public’s acceptance, but as a process in which the public has meaningful opportunities to deliberate and contribute to decisions about whether and how to use gene drive technologies. Public engagement therefore also provides an opportunity for public consideration and input as what constitutes beneficial and harmful outcomes, how to deal with uncertainty about those outcomes, what level of precaution to endorse, and how to understand the human relationship to nature. Public engagement is taken up in detail in Chapter 7. Public engagement in order to undertake a risk assessment is discussed in Chapter 6.

Engaging with members of the public is complicated by variations in the perception of risk. In risk assessment and in this report, risk is understood to involve measurable parameters—the statistical likelihood and the severity of a given harm. A considerable body of psychological research attests, however, that how people perceive and evaluate risks involves more than these measurable parameters. The risk of a harmful outcome is likely to be perceived as greater for some types of harm than for others (Slovic, 1987). Those risks of harm likely to be seen as greater are distinguished in psychometric research as being unfamiliar, uncontrollable, imposed rather than voluntarily accepted, associated with a sense of dread, and catastrophic (Slovic, 1987).

Genetic technologies rank high on these measures (Slovic, 1987). Gene drives might rank particularly high if their capacity to alter shared environments is associated with a marked sense of dread and unfamiliarity and if their capacity to be “invasive” is seen as a lack of controllabil-

ity. The issues raised by attempts to release genetically engineered mosquitoes in the Florida Keys in order to drive down populations of dengue-transmitting mosquitos may illustrate the challenge confronting the use of gene drive technologies (Alvarez, 2015). Public distrust of genetically engineered crops and livestock may encourage a similar distrust of gene drives. The fact that gene-drive modified organisms would be deliberately introduced into wild populations and comparatively less managed environments may cause some members of the public to see them as even more unattractive than other genetically modified organisms. On the other hand, gene drives systems might turn out to be less threatening than other genetic technologies if they can be put to significant conservation purposes and if they are not seen as reflecting corporate interests and a disregard for the environment. Such considerations show the importance of being wary about any one way of framing gene drive technologies, and they also reveal some challenges to be addressed in public and community engagement.

Finally, the scholarly commentary raises questions about existing governance. Oye et al. (2014) suggest that US regulations may be inadequate for gene-drive modified organisms in general and may not apply to insects at all (see Chapter 8). Others have raised the question of whether US regulations would apply to drives designed to be inserted into plants without using a plant disease vector (Caplan et al., 2015). Oye et al. (2014) also suggest that both US and international security regulations may not apply to drives that raise dual use concerns because those regulations rely on lists of agents and may not include gene drives. Webber et al. (2015) hold that the decision of whether to use a gene-drive modified organism to try to eliminate an invasive species requires a regulatory framework that provides a mechanism for working through the relevant concerns. Governance of gene drive research is discussed at length in Chapter 8.

CONCLUSIONS

Questions about responsible science and applications of gene drive technology rest on values at every step, from why and how research should be conducted to whether and where a gene-drive modified organism could be released into the environment. Values are also implicit in the development of appropriate governance for this new field.

Key value-based questions concern the determination of the potential benefits and harms of gene drives to humans and the environment. There are also questions about who would benefit, who would be harmed, and who would make decisions about gene drive technologies. A third area concerns the place of humans in ecosystems and our larger relationship to nature. Some of these questions echo considerations in debates about genetic engineering.

Considerations regarding the potential benefits and harms of gene-drive modified organisms will be central in deciding whether to allow field testing or open environmental release. Understanding and comparing potential outcomes involves a number of challenges. Benefits and harms can be identified and assigned appropriate weight only case by case and only with the input of the people who will be affected by the release. Perceptions of outcomes may also be affected by a range of cultural and psychological factors in addition to the statistical likelihood and the quantifiable severity of a given harm.

Not everyone will be affected by gene drive research and applications in the same way. When selecting sites for field trials or open environmental releases, it will be important that researchers consider the values of the publics affected by the release and their understanding of the balance of benefits and harms. The expectation that people should have a voice in fundamental decisions that affect their health and their environment is particularly important and may generate additional guidelines for the release of gene-drive modified organisms. Approaches to ensure that communities participate meaningfully in decision making about the use of gene-drive modified organisms will be essential, particularly in low- and middle-income countries where power differentials may affect such participation.

Perspectives on the place of human beings in ecosystems and our larger relationship to nature—including human impact on and manipulation of ecosystems—have an important role in

the emerging debate about gene drives. The increased power that gene drive technologies might give human beings to alter, and perhaps eliminate, wild species, thereby altering the shared environment, will be intrinsically objectionable to some people. An increased ability to conserve species and ecosystems or protect public health through gene drive technologies may be intrinsically attractive to other people.

Developing public policies for gene-drive modified organisms will require careful attention to the human relationship to nature, a need that is amplified for proposals to use gene drives in ways that could lead to the extinction of species or significantly alter the environment.

Some of the fundamental reasons to conduct gene drive research include widely shared commitments to fighting human disease, promoting human welfare, and protecting and restoring the natural environment. In addition, research on gene drives aligns with the intrinsic value that many people find in the pursuit of knowledge, understanding, and innovation. However, widely shared commitments to protect human welfare and the environment also provide reasons to develop public policy guidelines that may constrain research on gene drives or the releases of gene-drive modified organisms. Integrating precautionary measures into the research process can help to balance these potentially conflicting commitments—for example, by using structured tools to assess potential benefits and harms, by providing ample opportunities to gather further information about potential outcomes and revisit decisions about how to proceed, and by ensuring that people who will be affected by a proposed release are integrated into the decision-making process.

REFERENCES

Singleton, L.A. Hinds, H. Leirs and Z. Zhang, eds. Canberra, Australia: Australian Centre for International Agricultural Research.

Charo, R.A., and H.T. Greely. 2015. CRISPR critters and CRISPR cracks. Am. J. Bioeth. 15(12):11-17.

Crampton, J.M., S.L. Stowell, M. Karras, and R.E. Sinden. 1999. Model systems to evaluate the use of transgenic haematophagous insects to deliver protective vaccines. Parassitologia 41(1-3):473-477.

Cronon, W. 1995. The trouble with nature; or, getting back to the wrong nature. Pp. 69-90 in Uncommon Ground: Rethinking the Human Place in Nature, W. Cronon, ed. New York: W.W. Norton.

Dobson, S.L., C.W. Fox, and F.M. Jiggins. 2002. The effect of Wolbachia-induced cytoplasmic incompatibility on host population size in natural and manipulated systems. Proc. Biol. Sci. 269(1490):437-445.

Elliott, C. 1992. Where ethics comes from and what to do about it. Hastings Center Report 22(4):28-35.

Ellstrand, N.C., P. Meirmans, J. Rong, D. Bartsch, A. Ghosh, T.J. de Jong, P. Haccou, B.-R. Lu, A.A Snow, C.N. Stewart Jr., J.L. Strasburg, P.H. van Tierderen, K. Vrieling, and D. Hooftman. 2013. Introgression of crop alleles into wild or weedy populations. Annu. Rev. Ecol. Evol. Syst. 44:325-345.

El-Zahabi-Bekdash, L., and J.V. Lavery. 2010. Precaution through effective community engagement in research with modified mosquitoes. AsPac. J. Mol. Biol. Biotechnol. 18(2):247-250.

Engber, D. 2016. Let’s kill all the mosquitoes. Slate, February 1, 2016 [online]. Available at http://www.slate.com/articles/health_and_science/science/2016/01/zika_carrying_mosquitoes_are_a_global_scourge_and_must_be_stopped.html [accessed April 22, 2016].

Esvelt, K.M. 2016. A Proposal to End Schistosomiasis [online]. Available at http://www.sculptingevolution.org/genedrives/schistosomiasis [accessed April 19, 2016].

Esvelt, K.M., A.L. Smidler, F. Catteruccia, and G.M. Church. 2014. Concerning RNA-guided gene drives for the alteration of wild populations. eLife 3:e03401.

Fauci, A.S., G.L. Nabel, and F.S. Collins. 2011. A flu virus risk worth taking. The Washington Post, December 30, 2011 [online]. Available: https://www.washingtonpost.com/opinions/a-flu-virus-riskworth-taking/2011/12/30/gIQAM9sNRP_story.html [accessed March 17, 2016].

Fenner, F. 1983. Biological control, as exemplified by smallpox eradication and myxomatosis. Proc. R. Soc. Lond. B. Biol. Sci. 218(1212):259-285.

Finkel, A.M. 2011. “Solution-focused risk assessment”: A proposal for the fusin of environmental analysis and action. Hum. Ecol. Risk Assess. 17(4):754-787.

FOE (Friends of the Earth). 2012. The Principles for the Oversight of Synthetic Biology. Washington, DC: Friends of the Earth.

Freed, L.A. 1999. Extinction and endangerment of Hawaiian honeycreepers: A comparative approach. Pp. 137-162 in Genetics and the Extinction of Species, L.F. Landweber, and A.P. Dobson, eds. Princeton: Princeton University Press.

Gao, X., and P. Zhang. 2007. Transgenic RNA Interference in mice. Physiology 22(3):161-166.

Garden, D.S. 2005. Australia, New Zealand, and the Pacific: An Environmental History (Nature and Human Societies), M.R. Stoll, ed. Santa Barbara: ABC-CLIO.

Goff, M.L., and C. van Riper, III. 1980. Distribution of mosquitoes (Diptera: Culicidae) on the east flank of Mauna Loa Volcano, Hawaii. Pac. Insects 22:178-188.

Gould, F. 2015. General Overview of Gene Drive Research in Different Organisms. Webinar, October 15, 2015. Available at: http://nas-sites.org/gene-drives/2015/10/02/webinar-gene-drive-research-indifferent-organisms/ [accessed March, 17, 2016].

Gurwitz, D. 2014. Gene drives raise dual-use concerns. Science 345(6200):1010.

Heath, G., D. Childs, M.F. Docker, D.W. McCauley, and S. Whyard. 2014. RNA interference technology to control pest sea lampreys—a proof-of-concept. PLoS ONE 9(2):e88387.

Higgs, S., K.E. Olson, L. Klimowski, A.M Powers, J.O. Carlson, R.D. Possee, and B.J. Beaty. 1995. Mosquito sensitivity to a scorpion neurotoxin expressed using an infectious Sindbis virus vector. Insect Mol. Biol. 4(2):97-103.

Hygnstrom, S.E., and D.R. Virchow. 1992. G92-1106 Controlling Rats. Historical Materials from the University of Nebraska-Lincoln Extension Paper 1512 [online]. Available at: http://digitalcommons.unl.edu/cgi/viewcontent.cgi?article=2508&context=extensionhist [accessed March 17, 2016].

Jacob, J., G.R. Singleton, and L.A. Hinds. 2008. Fertility control of rodent pests. Wildlife Res. 35(6):487-493.

Jennings, B. 2015. Science, Ethics, and Governance Considerations for Gene Drive Research. Webinar, October 28, 2015. Available: http://nas-sites.org/gene-drives/2015/10/03/second-public-meeting/ [accessed March 17, 2016].

Joubert, D.A., T. Walker, L.B. Carrington, J.T. De Bruyne, D.H. Kien, T Hoang Nle, N.V. Chau, I. Iturbe-Ormaetxe, C.P. Simmons, and S.L. O’Neill. 2016. Establishment of a Wolbachia superinfection in Ae

des aegypti mosquitoes as a potential approach for future resistance management. PLoS Pathog. 12(2):e1005434.

Kaebnick, G.E. 2014. Humans in Nature: The World As We Find It and the World As We Create It. New York: Oxford University Press.

Kaebnick, G.E., M.K. Gusmano, and T.H. Murray. 2014. The ethics of synthetic biology: Next steps and prior questions. Hastings Center Report 44(S5):S4-S26.

Kamrud, K.I., K.E. Olson, S. Higgs, A.M. Powers, J.O. Carlson, and B.J. Beaty. 1997. Detection of chloramphenicol acetyltransferase in the saliva of Culex pipiens mosquitoes. Insect. Biochem. Mol. Biol. 27(5):423-429.

King, K.F., P. Kolopack, M.W. Meritt, and J.V. Lavery. 2014. Community engagement and the human infrastructure of global health research. BMC Med. Ethics 15:84.

Kysar, D. 2010. Regulating from Nowhere: Environmental Law and the Search for Objectivity. New Haven, CT: Yale University Press.

Lavery, J.V., L.C. Harrington, and T.W. Scott. 2008. Ethical, social, and cultural considerations for site selection for research with genetically modified mosquitoes. Am. J. Trop. Med. Hyg. 79(3):312-318.

Lorvelec, O., and M. Pascal. 2005. French attempts to eradicate nonindigenous mammals and their consequences for native biota. Biol. Invasions 7(1):135-140.

MacLean, D. 1998. The ethics of cost-benefit analysis: Incommensurable, incompatible, and incomparable values. Pp. 107-121 in Democracy, Social Values, and Public Policy, M.M. Carrow, R.P. Churchill, and J.J. Cores, eds. Westport, CT: Praeger.

Macrina, F.L. 2014. Scientific Integrity: Text and Cases in Responsible Conduct of Research, 4th Ed. Washington, DC: ASM Press.

Mandel, G.N., and T. Gathii. 2006. Cost-benefit analysis versus the precautionary principle: Beyond Cass Sunstein’s laws of fear. U. Illinois Law Rev. 5:1037-1080.

Marris, E. 2013. The Rambunctious Garden: Saving Nature in a Post-Wild World. London: Bloomsbury.

Meerburg, B.G., F.W.A. Brom, and A. Kijlstra. 2008. The ethics of rodent control. Pest Manag. Sci. 64(12):1205-1211.

Mensching, D., and P. Volmer. 2008. Rodenticides. Pp. 1191-1196 in Handbook of Small Animal Practice, 5th Ed., R.V. Morgan, ed. St Louis, MO: Saunders Elsevier.

National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research. 1978. The Belmont Report: Ethical Principles and Guidelines for the Protection of Human Subjects of Research. US Department of Health, Education and Welfare Publication No. (OS) 78-0014. Washington, DC: US Government Printing Office.

Newhouse, A.E., L.D. Polin-McGuigan, K.A. Baier, K.E.R. Valletta, W.H. Rottmann, T.J. Tschaplinski, C.A. Maynard, and W.A. Powell. 2014. Transgenic American chestnuts show enhanced blight resistance and transmit the trait to T1 progeny. Plant Sci. 228:88-97.

NSABB (National Science Advisory Board for Biosecurity). 2007. Proposed Framework for the Oversight of Dual Use Life Sciences Research: Strategies for Minimizing the Potential Misuse of Research Information [online]. Available at: http://osp.od.nih.gov/sites/default/files/biosecurity_PDF_Framework%20for%20transmittal%200807_Sept07.pdf [accessed April 22, 2016].

Oye, K.A., K. Esvelt, E. Appleton, F. Catteruccia, G. Church, T. Kuiken, S.B. Lightfoot, J. McNamara, A. Smidler, and J.P. Collins. 2014. Biotechnology. Regulating gene drives. Science 345(6197):626-628.

Parke, E.C., and M. Bedau. 2004. The precautionary principle and its critics. Pp. 69-97 in The Ethics of Protocells: Moral and Social Implications of Creating Life in the Laboratory, E.C Parke, ed. Cambridge, MA: MIT Press.

Pennisi, E. 2015a. How crop-killing witchweed senses its victims. Science 350(6257):146-147.

Pennisi, E. 2015b. Gene drive turns mosquitoes into malaria fighters. Science 350(6264):1014.

Pollan, M. 1991. Second Nature: A Gardener’s Education. New York: Grove Press.

President’s Commission for the Study of Ethical Issues in Medicine and Biomedical and Behavioral Life Sciences. 1982. Splicing Life: A Report on the Social and Ethical Issues of Genetic Engineering with Human Beings. Washington, DC: President’s Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research.

Presidential Commission for the Study of Bioethical Issues. 2010. New Directions: The Ethics of Synthetic Biology and Emerging Technologies. Washington, DC: Presidential Commission for the Study of Bioethical Issues.

Rich, F. 2016. Getting to Green, Saving Nature: A Bipartisan Solution. New York: W. W. Norton.

Sagoff, M. 2003. Genetic engineering and the concept of the natural. Pp. 11-26 in Genetic Prospects: Essays on Biotechnology, Ethics, and Public Policy, V.V. Gehring, ed. Lanham, MD: Rowman and Littlefield.

Sakai, A.K., F.W. Allendorf, J.S. Holt, D.M. Lodge, J. Molofsky, K.A. With, S. Baughman, R.J. Cabin, J.E. Cohen, N.C. Ellstrand, D.E. McCauley, P. O’Neil, I.M. Parker, J.N. Thompson, and S.G. Weller. 2001. The population biology of invasive species. Annu. Rev. Ecol. Syst. 32:305-332.

Samuel, M.D., B.L. Woodworth, C.T. Atkinson, P.J. Hart, and D.A. LaPointe. 2015. Avian malaria in Hawaiian forest birds: Infection and population impacts across species and elevations. Ecosphere 6(6):1-21.

Sarewitz, D. 1996. Frontiers of Illusion: Science, Technology and the Politics of Progress. Philadelphia: Temple University Press.

Sarewitz, D. 2015. Science can’t solve it. Nature 522(7557):413-414.

Saunders, G., B. Cooke, K. McColl, R. Shine, and T. Peacock. 2010. Modern approaches for the biological control of vertebrate pests: An Australian perspective. Biol. Control 52(3):288-295.

Schaal, B.A., J.F. Gaskin, and A.L. Caicedo. 2003. Phylogeography, haplotype trees, and invasive plant species. J. Hered. 94(3):197-204.

Sinden, A. 2015. Formality and informality in cost-benefit analysis. Utah Law Rev. (1):93-172.

Slovic, P. 1987. Perception of risk. Science 236(4799):280-285.

Sogge, M.K., S.J. Sferra, and E.H. Paxton. 2008. Tamarix as habitat for birds: Implications for riparian restoration in the Southwestern United States. Restoration Ecol. 16(1):146-154.

Soper, K. 1995. What Is Nature? Culture, Politics, and the Non-Human. Oxford, UK: Blackwell.

Sunstein, C. 2005. Laws of Fear: Beyond the Precautionary Principle. Cambridge, UK: Cambridge University Press.

Thompson, P.B. 2003. Unnatural farming and the debate over genetic manipulation. Pp. 27-40 in Genetic Prospects: Essays on Biotechnology, Ethics, and Public Policy, V.V. Gehring, ed. Lanham, MD: Rowman and Littlefield.

Valkiūnas, G. 1993. Pathogenic influences of haemosporidians and trypanosomes on wild birds in the field conditions: Facts and hypotheses. Ekologija 1:47-60.

van Riper III, C., and J.M. Scott. 2001. Limiting factors affecting native birds of Hawaii. Stud. Avian Biol. 22:221-233.

van Riper III, C., S.G. Van Riper, M.L. Goff, and M. Laird. 1986. The epizootiology and ecological significance of malaria in Hawaiian landbirds. Ecol. Monogr. 56(4):327-344.

Warner, R.E. 1968. The role of introduced diseases in the extinction of the endemic Hawaiian avifauna. Condor 70:101-120.

Waters, C.N., J. Zalasiewicz, C. Summerhayes, A.D. Barnosky, C. Poirier, A. Gałuszka, A. Cearreta, M. Edgeworth, E.C. Ellis, M. Ellis, C. Jeandel, R. Leinfelder, J.R. McNeill, D.d. Richter, W. Steffen, J. Syvitski, D. Vidas, M. Wagreich, M. Williams, A. Zhisheng, J. Grinevald, E. Odada, N. Oreskes, and A.P. Wolfe. 2016. The Anthropocene is functionally and stratigraphically distinct from the Holocene. Science 351(6269):aad2622.

Weber, K. 2010. Cane Toads and Other Rogue Species. New York: Public Affairs.

Webber, B.L., S. Raghu, and O.R. Edwards. 2015. Is CRISPR-based gene drive a biocontrol silver bullet or global conservation threat? Proc. Natl. Acad. Sci. 112(34):10565-10567.

WHO (World Health Organization). 2009. Dengue in the Western Pacific Region. Geneva, Switzerland: World Health Organization.

WHO. 2014. World Malaria Report 2014. Geneva, Switzerland: World Health Organization.

Williams, T. 2013. Poisons used to kill rodents have safer alternatives. Audubon Magazine, January-February 2013 [online]. Available at: http://www.audubonmagazine.org/articles/conservation/poisons-used-killrodents-have-safer-alternatives?page=3 [accessed March 17, 2016].

Witmer, G., and S. Jojola. 2006. What’s up with house mice? A review. Pp. 124-130 in Proceedings of the 22nd Vertebrate Pest Conference, R. M. Timm, and J. M. O’Brien, eds. Davis, CA: University of California, Davis.

Witmer, G., J. Pierce, and W.C. Pitt. 2011. Eradication of invasive rodents on islands of the United States. Pp. 135-138 in Island Invasives: Eradication and Management, C. R. Vietch, M. N. Clout, and D. R. Towns, eds. Gland, Switzerland: IUCN.

WMA (World Medical Association). 2013. Declaration of Helsinki: Ethical Principles for Medical Research Involving Human Subjects [online]. Available at: www.wma.net/en/30publications/10policies/b3/ [accessed April 22, 2016].

Wolff, J. 2014. The precautionary attitude: Asking preliminary questions. Hastings Center Report 44(S5): S27-S28.

Zavaleta, E.S., R.J. Hobbs, and H.A. Mooney. 2001. Viewing invasive species removal in a whole-ecosystem context. Trends Ecol. Evol. 16(8):454-459.