As is the case with all laboratory animals, it is imperative to carefully consider the ethical implications of acquiring marmosets, keeping and breeding them in captivity, and using them for research. Workshop presenters and attendees discussed theoretical and practical aspects of these considerations, as well as concrete steps the research community can take to fully explore the ethical implications of working with marmosets and guide care and research activities.
Adam Shriver, a research fellow at the Oxford Uehiro Centre for Practical Ethics and the Wellcome Centre for Ethics and Humanities, discussed how an understanding of ethical theories can inform approaches to working with research animals.
What Is Ethics?
Ethics is the process of determining which course of action has the best moral reasons for being undertaken. These reasons should be universal, meaning that anyone in a similar situation would have the same reasons to take the same action (Hare 1981). Ethics is not about self-interest, emotion, or what is legal or customary. Like good science, good ethics analyzes competing views using arguments based in reason and evidence. Moreover, Shriver said, the best ethical arguments are those that take competing
theories seriously and offer detailed responses rather than simply ignoring alternative value systems.
People’s conceptions of “ethics” can vary quite widely but ethicists have developed helpful distinctions that can greatly improve the clarity of ethical discussion. The Harvard philosopher John Rawls wrote the following: “Moral Theory is the study of substantive moral conceptions, that is, the study of how basic notions of the right, the good, and moral worth may be arranged to form different moral structures” (Rawls 1974). The “right, good, and moral worth” mentioned by Rawls refer to what moral philosophers call deontic concepts, value concepts, and concepts of moral worth and can be used to evaluate different aspects of morality (Rawls 1974). Deontic concepts such as “right” or “wrong” refer to the evaluation of actions. We use such concepts when we say “keeping a promise is right” or “lying is wrong.” Value concepts such as “good” or “bad” refer to evaluations of states of the world; a tragic mining accident is bad, but having a high gross domestic product (GDP) is good. Concepts of moral worth refer to evaluations of character traits like virtues and vices such as honesty or greed.
As Rawls suggests, there are different ways of thinking about the relationships between these types of concepts, Shriver continued. Keeping these distinctions in mind can greatly help the clarity of ethical debate. Different views on the significance of value concepts, deontic concepts, and moral worth concepts form the basis for the three most prominent types of ethical theory: consequentialism, deontological ethics, and virtue ethics. These theories can inform approaches to assessing the ethics of animal research.
Consequentialism is the view that whether an action is right or wrong depends entirely on how good or bad the consequences are that result from the action (Scheffler 1988). On the simplest version of consequentialism, when you undertake an action, if the overall consequences that result from the action were good, then the action was right. If the overall consequences were bad, the action was wrong. We can think of the harms and benefits in harm–benefit analysis (or cost–benefit analysis) as referring to bad and good consequences. Thus, in the context of animal research, this simple form of consequentialism would entail that in order for animal research to be morally justified, any harms done to animals must be outweighed by the potential benefits the research provides.
However, many prominent consequentialist views say that right actions are not just those that have consequences where the good outweighs the bad even by a very small amount, but instead are only the actions that produce the best possible ratio of good and bad consequences. In particular, the
prominent consequentialist view known as utilitarianism suggests that all good and bad consequences are ultimately reducible to pleasures and pains, and this theory holds that right actions are therefore those that maximize pleasures and minimize pains (Mill 1863; Sidgwick 1907).
Weighing the harm to animals and the benefits to humans using consequentialist theories may sound relatively simple, but in fact requires consideration of many deeper questions. One important set of questions relates to which harms and benefits are considered. How should we think of potential benefits to humans when their likelihood is uncertain? Should gaining knowledge be considered valuable in itself even if the benefits of the knowledge remain uncertain? Do we include the harms animals suffer from procurement, transport, and housing in our harm–benefit analysis or the ecological harms that occur when animals are taken from their native habitats?
Another important question is how we determine which beings have moral standing. What capacities must beings have in order to deserve moral standing? A baseline that most agree on is sentience—whether a being can feel pleasure or pain. Working with sentient beings imposes the obligation to minimize negative experiences. This question of sentience and moral standing is particularly important for work involving chimeras and genetic modification because those processes can alter neural systems and produce new capabilities that might change a being’s moral standing.
In addition to sentience, moral standing can also be based on other cognitive capacities, such as an organism’s capacity for forward-looking desires or a sense of self. The fact that primates have these capacities makes working with them both appealing for researchers trying to understand human behavior and ethically problematic. Concern is warranted when studies produce the type of suffering in animals that we aim to avoid in humans.
Complicating these considerations is the fact that animals cannot speak for themselves, so it is humans who must evaluate animal sentience and higher capabilities. To address this, a commonly accepted idea called the precautionary principle states that the absence of full scientific certainty should not be used as a reason to avoid taking steps to prevent potential seriously bad outcomes (Birch 2017). As such, even in the absence of certainty, we should assume animals in research can suffer and should take appropriate steps to prevent this.
According to all consequentialist theories, the best actions would avoid causing any unnecessary harms. Thus, the Three Rs Principles (3Rs), developed by Russell and Birch (1959), emphasizing reduction, replacement, and refinement can be seen as a requirement of all ethical research. Nevertheless, minimizing unnecessary harms is not the same thing as weighing harms against benefits, so the fact that the 3Rs has been followed during research is not sufficient to show, on consequentialist analysis, that the research is
justified. It would additionally need to be determined that the overall benefits of the research sufficiently outweigh the harms.
Finally, weighing harm to nonhuman animals against benefits to humans requires making judgments about human interests versus nonhuman animal interests, and there is widespread disagreement about how these interests should be weighed against one another. At one end of the spectrum, a view sometimes referred to as lexical superiority holds that human interests always have higher value than nonhuman interests. At the other end of the spectrum, the Principle of Equal Consideration of Interests (PEC) holds that animal and human interests deserve the same standing (Singer 2011). While the PEC does not state that human and animal interests are the same, because humans and animals have different capacities, it posits that when experiences are similar they should be similarly weighed, such that a mouse’s pain, for example, would be given equal weight to a human’s pain.
Deontological ethics is an alternative school of thought that can be thought of as the denial of consequentialism (Scheffler 1994). According to these views, actions are not right or wrong because they have good or bad consequences; other considerations are relevant (Kant 1785; Ross 2002). These considerations could include the motivations of the person performing the action, or the relationship the person has to those harmed or benefit-ted by the action. On deontological theories, some actions are wrong even if they result in overall positive consequences. For example, when we say that humans have a right to life, liberty, and the pursuit of happiness, this means that it is wrong to deprive people of these, even if doing so in particular circumstances would result in overall positive consequences. The concept of rights (human or animal rights) is therefore a deontological concept; rights entail that some things are prohibited regardless of consequences.
Bioethics regulations related to human research are strongly deontological in flavor. For example, if research on human subjects was conducted without gaining proper consent or without properly informing people of the potential risks involved, it would be considered wrong, even if it ultimately contributed to a medical advancement. As a society, we think that there are certain restrictions on how research can be conducted on humans that always apply.
Similar to questions about moral standing, it is necessary to define what qualities impart human rights: is sentience alone sufficient (Regan 1986), or is it necessary to have high-level rational capacity that only humans possess (Cohen 1986)? The answers are important for determining the rights held by nonhuman primates (NHPs) or even non-primate animals.
When most people think of animal rights in context of research, they tend to envision strict prohibitions on the use of animals in research. With animals as cognitively sophisticated as NHPs, there certainly are a number of ethicists who would hold that they should not be used in any invasive research (Regan 1986). However, deontological theories are also compatible with more limited claims. For example, setting an upper limit on the amount of pain or distress that can be inflicted on an animal in research can be thought of as a deontological rule. Similarly, some would argue that primates have a right to be housed in conditions that provide them with adequate space and opportunities for social engagement. And some argue that primates of a sufficient degree of cognitive sophistication should be provided an opportunity to dissent to procedures (Fenton 2014).
Deontological ethics can also involve looking beyond outcomes and considering the potential for a variety of harms during the process of an animal’s care and use, including harms to its dignity, bodily integrity, or autonomy. For example, while research involving genetic modification may not induce pain, it may change a being in a way that could affect its bodily integrity or autonomy (Bovenkirk 2002).
Robert Nozick introduced the idea of “Utilitarianism for animals, Kantianism for people,” suggesting the deontological rules apply only to humans. This idea seems to be roughly reflected in biomedical rules governing research on humans and nonhumans. Even though people disagree about the specific range of rights conferred to animals, most agree that there is a threshold beyond which it is unethical to cause animals pain. To the extent this is true, their views would be consistent with some deontological ideas.
Finally, virtue ethics is an assessment of the character traits of a person, rather than his or her actions (Annas 2009). According to these views, morality goes beyond evaluating the rightness of particular actions. It is instead based on an assessment of behavior as a whole. Because virtue ethics has a different target than consequentialist and deontological ethics, it also can be seen as compatible with one or the other of these views (Driver 2001). However, some virtue ethicists have argued that these other approaches are wrong and at odds with humanity’s historical conceptions of morality (e.g., that of ancient Greek philosophers and Confucianism), which did not provide decision procedures for deciding what to do in every individual circumstance but instead emphasized the character traits of virtuous individuals and their sensitivity to context (Anscombe 1958).
Virtue ethics can be relevant for research on animals in a number of ways. First, it may suggest that a one-size-fits-all approach to evaluating
research may not sufficiently take into account subtle context that is relevant for evaluating particular experiments. Second, virtue ethics may suggest that researchers and animal caretakers take on special obligations due to their professional roles. And finally, many virtue ethics supporters would hold that approaches toward animals ought to be consistent across domains of life in order to count as truly ethical behavior. For example, if one claimed to be consequentialist or utilitarian toward animals when discussing research, but did not take into account possible harms to animals in decisions about food choices or purchasing behavior, it would be difficult to claim that the person truly possessed the praiseworthy character trait of showing compassion toward animals.
Shriver concluded that there are many under-discussed issues related to NHP research and that it would be important to engage in a full discussion of both the ethical commitments and alternative perspectives. Importantly, he stressed, there needs to be a space where these debates can occur without people feeling personally attacked. Just as with the best science, the strongest ethical positions are those that take into account and respond to the ideas raised by others.
Steven Niemi, the director of the Office of Animal Resources at Harvard University, spoke to the practical aspects of current approaches to ethics in animal care and potential paths for improvement. These considerations involve not only how people conceive of animal research ethics from a philosophical standpoint, but how those conceptions are upheld (or not) in practice. Other considerations relate to how research practices are perceived by different stakeholders, such as the research community, policy makers, and the general public.
People today view animal ethics differently than they did in the past. Studies suggest that an increasing number of people believe that animals have the same rights as humans, and that the public is finding it less and less morally acceptable to use animals in medical research (Jones 2017). At the same time, communication has become more global and instantaneous, though not necessarily more reliable; it is often difficult to separate fact from fiction. As societies change, values will follow. It is vital for those involved in animal research to respond to changing values with proactive engagement and changes in practice when appropriate. The use of any primate species raises particular ethical questions that must be carefully considered.
To make value judgments and guide ethical practices, it is vital to examine what is known about marmosets. Marmosets are intelligent, have complex social needs, and are fragile in captivity. Organizations including
the Institute for Laboratory Animal Research (ILAR), the Association of Primate Veterinarians (APV), and the American College of Laboratory Animal Medicine (ACLAM) offer information that can increase understanding of these animals and guide ethical practices.
Pitfalls of Current Practice
There are important gaps in the mechanisms set up to review and enforce protocols for working with marmosets and other laboratory animals. Most current review and assessment protocols are either internal to an institution, occur infrequently, or remain opaque to the public.
Institutional Animal Care and Use Committees (IACUCs) are a primary mechanism to review protocols, but they are internal bodies dependent on institutional expertise, raising questions about the potential for subjectivity or biases. External mechanisms, such as the U.S. Department of Agriculture inspections, may diagnose problems but come too late to protect animals. While researchers with federal grant money are obliged to self-report lapses to the National Institutes of Health’s (NIH’s) Office of Laboratory Animal Welfare, this mechanism does not prevent lapses from occurring in the first place.
AAALAC International (formerly known as the Association for the Assessment and Accreditation of Laboratory Animal Care International), which is both voluntary and confidential, merely ensures compliance with existing policies and laws (in the United States these are the Animal Welfare Act; the Guide for the Care and Use of Laboratory Animals; and the Public Health Service Policy), while the 3Rs were designed to minimize harm in animal use, not animal care, and they may be an insufficient framework in that context.
While ethically caring for animals is inherently important, there are also significant political consequences when mistakes are made. Negative publicity can harm an institution’s reputation and the entire field of animal research. Consideration for such risks goes beyond personal feelings about particular institutional practices, with implications for the ethics of the research community as a whole.
Potential Paths to Improvement
A potential alternative to the current oversight mechanisms is an external independent and expert confirmation of an institution’s competence, that could involve the creation of a peer exchange network, and an independent assessment of the level of readiness of an animal care and use program. In addition, competence assessments could be shared with the regulators in writing so that they are attuned to current practices.
Open sharing of results, successes, and failures would create an environment that better supports continuous improvement. For example, a professional repository hosted by an expert, neutral organization could provide a robust body of metrics (i.e., evidence-based performance standards) for assessing and comparing inputs and outputs of marmoset research. This would improve knowledge and transparency and potentially lead to innovative study designs. An important consideration, however, is that “best practices” can be problematic because people often assume that those are the only practices allowed. It may be more useful to create a catalog of successful applications or specific circumstances.
As part of any new tracking, evaluating, or reporting mechanism, it is necessary to consider what could go wrong, for instance, if an institution fails its competency assessment. The potential consequences of such a failure should serve as a motivator to be carefully prepared before new types of research begin, and to anticipate adverse publicity, with a plan to address not only the facts but the emotions involved.
Creating “starter kits” also could help newcomer institutions establish marmoset research with appropriate practices from the beginning. For example, such a kit could describe the recommended minimum number of animals or families, as well as their housing, social, and dietary needs. It could also include the institution’s minimum diagnostic capabilities and the required level of expertise to administer medicine, vaccines, or screenings.
Another possible improvement is to conduct marmoset research in the natural habitats where marmosets live. U.S. researchers are not limited to working in facilities within the United States. If the animals could be housed closer to their home, perhaps in a more natural environment, it could relieve some of the stress they experience during transport and research could potentially take less time and impose fewer expenses.
A more open, inclusive, and equitable approach to engaging with interested parties would help the research community better explore the nuances of working with laboratory animals. In addition to the public, vendors of laboratory equipment, caging-enrichment devices, substrates, and diet could be involved in such discussions. Without involving a broader array of stakeholders in such discussions, it is impossible for researchers to know their motivations or level of expertise and the research community is prone to the pitfalls of making decisions in a vacuum.
Workshop participants discussed the different lenses through which animal research is viewed and raised ideas to address concerns and guide practice for the ethical care and use of marmosets. Several participants emphasized that, like good science, ethical positions are strongest when
they consider different perspectives, respond to alternative ideas, and minimize personal attacks or emotional reactions. While discussions about the ethics of research practices can be challenging to have—and to have publicly and transparently—it is crucial for the scientific community to have these discussions, especially when embarking into new research directions.
Participants highlighted the role of professional societies in guiding the field through these discussions. Examples include the American Society of Primatologists and the International Primatological Society. Many researchers who work with marmosets are immunologists or geneticists, not primatologists, and they may need the guidance or expertise these organizations can offer. In addition, a representative from APV expressed the organization’s commitment to addressing all the questions brought up at the workshop and contributing resources to resolving them.
Understanding the Nature of Basic Research
Participants noted that when considering the potential harms and benefits of animal research, and especially when communicating about them with the public, one important distinction is the one between basic research and the development of actual medical therapies. Fundamental misunderstandings regarding the nature and purpose of basic research, or a lack of awareness of how medical advances depend on many thousands of basic science studies at the molecular, cellular, and physiological levels, can lead to confusion and conflicting assumptions about what harms and benefits are being weighed. It is also difficult to account for surprises, in which basic research investigations uncover unexpected new findings that significantly advance research toward beneficial applications. Conversely, research avenues thought to have great benefits may not pan out as expected.
A number of participants pointed out that despite the difficulty of drawing a line from basic research to human benefits, being more proactive and transparent about this could increase the public’s awareness of these issues, especially that there are not always direct, obvious, or immediate payoffs to research. Despite these nuances, it is nonetheless possible to make an estimation of how the expected benefits balance against the expected harms of a research undertaking.
Accounting for the Full Spectrum of Harms
Several participants noted that another challenge is that there are no procedures in place to account for all possible harms animals may suffer as a result of animal research that arise from factors other than the research use itself. For example, there are often clear restrictions on the amount of stress or distress that is permissible during research activities, while harms
that may arise as a result of transportation or over-production are not included in these restrictions and may not be adequately addressed. While researchers are accountable for the full spectrum of harms that may be associated with an animal’s use in research, participants discussed the importance of ensuring the well-being not only of marmosets that are used in research, but also those that are not used (perhaps due to over-production) and those that may have been genetically modified.
Relationship Between Researchers and the Public
A number of audience participants agreed that the nature of the discussion about animal research ethics changes markedly when conducted in the political sphere. When politics are involved, discussions often become about winning or losing instead of accuracy or finding common ground. In Europe, animal rights activists and politicians have successfully campaigned toward a phasing out of NHP research. As researchers in the United States and other places grapple with the ethics of using animals in research, it is valuable to keep these experiences in mind.
One participant explained that in Europe, scientists, faced with increasing regulatory constraints on permissible research activities, have to provide compelling reasons to use animals in research. To do this more effectively, the participant suggested, it would be valuable to include in conversations regarding the use of animals in research the perspectives of animal rights activists, animal protection agencies, and researchers who use alternatives to animals. More collaboration among these groups can increase mutual respect and enable scientists to demonstrate why animal research is sometimes essential. Expanding on this idea, other participants added that it would also be valuable for such discussions to include the benefits animal-based research can yield for animals themselves—not only humans—for example, to enhance veterinary care, improve the welfare of pets, or inform conservation efforts.
A participant noted that before discussions about ethics in animal research even become part of the public debate, it is helpful to have more conversations regarding these issues in academic settings and the relative benefits of alternative options, one participant suggested. Such discussions could help chart a path toward cost-effective, reliable performance standards to measure data such as behavior, clinical methodologies, growth rates, and activity level. Such standards could bolster confidence in evaluations and strengthen the evidence base for research practices.
One goal all who use animals in research can agree on is the need for continuous improvement of their care. A participant suggested the research community needs to improve its animal care practices and communicate about those improvements with the public to justify assertions that the
field is both doing the right thing and also moving in the right direction. For example, medical advancements have made animal procedures far less invasive, which most stakeholders would view as positive progress.
Beyond just communication, outreach programs can also improve the relationship between the research community and the public and demonstrate that researchers care for the animals in ways that go beyond just their use in research use. For example, a participant from the Wisconsin National Primate Research Center (WNPRC) described how the WNPRC hosts K–12 and undergraduate school visits, which expose children to the idea of animal care as a career, enhance science majors’ understanding of primate research, and give both audiences a more thorough explanation of the role of animal research in biomedical discovery. By visiting research facilities, the participant noted, students can see for themselves what the research entails, and positive outreach can make a big difference in how the next generation of scientists works with animals. Center researchers also do outreach activities with patient support groups to explain how animal research is being used to develop better therapies for human diseases, the participant added.
Another participant noted that, in engaging with the public, more successful organizations tend to follow a “pull” approach that attracts people to play an active role in a topic or discourse, rather than pushing people away from the discussion. Many organizations are using the Internet to pull the public into their community or business by crowdsourcing solutions or offering prizes, for example. A corollary in animal research, the participant suggested, could be to issue a challenge for students to design a better marmoset house. Such an approach would pull students and their teachers into the issue from the perspective of the research community and provide an opportunity for a first-hand look at the nuances involved.
Annas, JE. 2009. Virtue Ethics. In The Oxford Handbook of Ethical Theory Oxford University Press. doi: 10.1093/oxfordhb/9780195325911.003.0019.
Anscombe GEM. 1958. Modern moral philosophy. Philosophy 33:1–19.
Birch J. 2017. Animal sentience and the precautionary principle. Animal Sentience 16(1).
Bovenkerk B, Brom FW, van den Bergh BJ. 2002. Brave new birds: The use of “animal integrity” in animal ethics. Hastings Cent Rep Jan–Feb; 32(1):16–22.
Cohen C. 1986. The case for the use of animals in biomedical research. N Engl J Med 315(14):865–870.
Driver J. 2001. Uneasy Virtue. New York: Cambridge University Press.
Fenton A. 2014. Can a chimp say “no”? Reenvisioning chimpanzee dissent in harmful research. Camb Q Healthc Ethics 23(2):130–139.
Jones JM. 2017. Americans hold record liberal views on most moral issues. Gallup News. Available at: https://news.gallup.com/poll/210542/americans-hold-record-liberal-viewsmoral-issues.aspx.
Hare RM. 1981. Moral Thinking: Its Levels, Method, and Point. Oxford, England: Oxford University Press.-
Kant I, Wood A. 1996. Groundwork of the metaphysics of morals (1785). In M. Gregor (Ed.), Practical Philosophy (The Cambridge Edition of the Works of Immanuel Kant, pp. 37–108). Cambridge, England: Cambridge University Press. doi: 10.1017/CBO9780511813306.007.
Mill, JS. 1863. Utilitarianism. London, England: Parker, Son, and Bourn.
Rawls, J. 1974. The Independence of Moral Theory. Proceedings and Addresses of the American Philosophical Association 48:5–22. doi: 10.2307/3129858.
Regan T. 1986. A case for animal rights. In Fox MW, Mickley LD (Eds.), Advances in Animal Welfare Science 1986/87 (pp. 179–189). Washington, DC: The Humane Society of the United States.
Ross D, Stratton-Lake P. (Ed.). 2002. The Right and the Good. Oxford, England: Oxford University Press.
Russell WMS, Burch RJ. 1959. The Principles of Humane Experimental Technique. London, England: Methuen and Co. [Reissued: 1992, Universities Federation for Animal Welfare, Herts, UK].
Scheffler S. (Ed.). 1988. Consequentialism and Its Critics. Oxford, England: Oxford University Press.
Scheffler S. 1994. The Rejection of Consequentialism: A Philosophical Investigation of the Considerations Underlying Rival Moral Conceptions. Oxford, England: Oxford University Press.
Sidgwick H. 1907. The Methods of Ethics, 7th ed. London, England: Macmillan and Co. Available at: https://www.gutenberg.org/files/46743/46743-h/46743-h.htm.
Singer P. 2011. Practical Ethics, 3rd ed. New York: Cambridge University Press.