and Peter Sandøe2
Laboratory Animal Science, IBMC – Instituto de Biologia Molecular e Celular, Universidade do Porto, Porto, Portugal
Danish Centre for Bioethics and Risk Assessment, Faculty of Life Sciences, University of Copenhagen, Frederiksberg C, Denmark
This chapter aims to encourage scientists and others interested in the use of animal models of disease – specifically, in the study of dementia – to engage in ethical reflection. It opens with a general discussion of the moral acceptability of animal use in research. Three ethical approaches are here distinguished. These serve as points of orientation in the following discussion of four more specific ethical questions: Does animal species matter? How effective is disease modelling in delivering the benefits claimed for it? What can be done to minimize potential harm to animals in research? Who bears responsibility for the use of animals in disease models?
Key wordsContractarian viewutilitarian viewanimal rights viewprinciple of 3Rs
Contemporary research in the life sciences, particularly in biomedicine, involves experimentation on large numbers of live animals. Many of the animals are used for research aiming to discover new ways to prevent, cure, or alleviate human diseases. Some research animals are used as disease models. That is, conditions are artificially induced in them, which in some relevant respects resemble the conditions that we want to prevent, cure, or alleviate in humans.
The process of altering an animal so that it can serve as a disease model sometimes involves distressing or painful interventions; and the conditions induced in the animals may give rise to anxiety, pain, and other forms of suffering. Moreover, animals are often housed in ways that limit their freedom, and they are invariably killed when the experiment comes to an end. The overwhelming majority of animals used are vertebrates with highly developed nervous systems. They cannot, of course, consent to their own participation in research. Nor do they stand to benefit, as individuals, from such participation.
These facts present both the scientific community and society in general with a question. In pursuit of, for the most part, admirable and understandable goals, scientists carry out experiments causing discomfort, pain, and distress to animals. Often, they limit the freedom of the animals they are dealing with, and eventually, in most cases, they kill them. Are they morally justified in acting in this way? We might also ask whether, where it exists, societal approval of this kind of scientific activity is warranted.
The answer to this question will clearly depend on one’s general view of human duties to animals. In this chapter, we will not defend a single view of this kind. Rather, we shall present three ethical perspectives. We urge the reader to reflect on her own stance with these perspectives in mind. Following this, we will go through a series of further questions relating to the use of animals as models of dementia, which we think each researcher should ask herself; and we will describe what we take to be the key ethical issues raised by these questions. The questions are: Does animal species matter? How effective is disease modelling in delivering the benefits claimed for it? What can be done to minimize potential harm to animals in research? Who bears responsibility for the use of animals in disease models?
Although we do not advocate a specific ethical stance, but rather urge the researcher to make up her own mind, we do have an ethical agenda. We think that, at the very least, people who make use of animals should be prepared to devote the time and effort it takes to think through their choices from an ethical perspective. This would not only have a positive impact on the animals used in research, but also improve the credibility of animal research across society as a whole.
2 Should One Use Animals in Research?
When it comes to consideration of the right way to treat animals, there is no single, unanimous view in our societies. Even within families, for example, when the issue of using animals for research is brought up at dinner, there are strong disagreements. Some of us are outraged by the idea of exposing innocent animals to painful research. Others will take a more moderate view, arguing that as long as the research is vital and everything possible is done to protect the animals from unnecessary suffering, it is acceptable to use animals in research. Yet others will say that they do not care about the plight of rats and mice, and that we should be free to use animals to make discoveries in medicine as we wish.
The same thing happens when philosophers meet to discuss the ethics of animal use. They divide, with various groups favoring different ethical theories about human duties to animals. Such theories can be useful vehicles through which to articulate the principles underlying disagreements about animal experimentation. For a fuller outline of the ideas presented in the rest of this section, see (1).
Here, we shall present and briefly discuss three ethical views, or theoretical approaches. According to the first, the contractarian view, it is only one’s own long-term interests that count from an ethical point of view. Since we depend for our own well-being on collaboration with other human beings, duties governing our dealings with fellow humans become established. However, no such duties exist in regard to animals (2).
According to the utilitarian view, on the other hand, what matters is the impact of what we do on the well-being of those affected by our actions. Here, the basic principle is that what entitles me to moral consideration, or what gives me moral status, is my capacity to suffer as a result of, or benefit from, your actions. Since not only other human beings but also other sentient animals have this capacity, we should be concerned about the welfare of both humans and animals. Of course, it is impossible to cater to the interests of every individual potentially affected by a course of conduct, so we should focus on the interests that are most dramatically served by the type of action we are considering. In essence, we should seek to produce the greatest total fulfilment of interests (3).
Finally, according to the animal rights view, we should distinguish between interests and rights. Rights must be respected. One should not allow interests to overrule them. In the case of human rights, this means that we do not allow an innocent being to be sacrificed for the sake of the common good. Advocates of animal rights expand this approach and apply it to all sentient animals. They therefore object to the idea of sacrificing animals for the sake of the good of others (4).
For those who adopt the contractarian view, the way animals are treated is not always irrelevant: once people are emotionally attached to certain kinds of animals, for example, and dislike or feel outraged by the practice of using them in painful experiments, this becomes an ethically relevant concern. For example, because most people like cats and dogs more than they like rats and mice, causing suffering to the former is likely, in the contractarian picture, to be a more serious problem than causing the same amount of suffering to the latter. Likewise, nonhuman primates will probably receive more protection than other animals, because their plight is of very considerable concern to many people.
What matters, on this view, are the feelings and beliefs of fellow humans on whose collaboration one depends to gain a licence to operate. On this approach, then, setting ethical limits to the use of animals for research boils down to the task of defining a publicly acceptable framework that allows humankind to harvest the potential benefits of animal-based research. One specific reason for looking after the welfare of animals involved in research is the avoidance, wherever possible, of experiments that are likely to cause public concern.
According to the utilitarian approach, the interests of every individual affected by an action deserve equal consideration. This means that for the utilitarian–unlike the contractarian–the impact of procedures, housing facilities, and the like, on the well-being of the laboratory animals must be taken into consideration in its own right. The only justification that can be given of animal use in research is that the cost to the animals used is outweighed by the benefits of the research.
On the utilitarian approach, then, ethical decisions require us to strike the most favorable balance of costs and benefits for all the sentient individuals affected by what we do. But doing the right thing, according to the utilitarian, is not just a matter of doing what is optimal. It is also essential to do something rather than nothing: if something can be done to increase well-being, we have a duty to do it. This utilitarian duty to act, proactively, so as always to bring about improvements has important consequences for society.
In the case of laboratory animals, a pragmatic utilitarian might be willing to apply something called the “Principle of the 3Rs.” This principle requires researchers, where possible, to replace the existing live-animal experiments with alternatives, reduce the number of animals used, and refine methods so that animals are caused less suffering (5). It is not hard to see that less invasive sampling techniques, improved housing systems, and more precise models requiring fewer animals to be used are likely to be viewed as morally attractive developments within the utilitarian perspective.
Utilitarianism, as described above, suggests that animal interests can be justifiably sacrificed where that leads to the protection or satisfaction of vital human interests – as happens in much biomedical research. But is that an acceptable view? A more radical variety of utilitarianism might be worth exploring. Animal experimentation sometimes means sacrificing vital interests an animal has in continued life and the avoidance of suffering. Insisting firmly that human and animal interests deserve equal consideration, the utilitarian philosopher Peter Singer has concluded that the sacrifice of such vital animal interests is acceptable only where the benefits are extraordinarily important:
[I]f a single experiment could cure a disease like leukemia, that experiment would be justifiable. But in actual life the benefits are always much, much more remote, and more often than not they are nonexistent (6, p. 85).
It is evident, then, that within the utilitarian approach a wide range of views are represented. Some utilitarian observers accept animal experiments when there are no alternatives and as long as we do our best to prevent and alleviate animal suffering. Others, like Singer, setting the demand for human benefit higher, would prefer to see nearly all such experiments abolished.
What all utilitarians agree on, however, is the methodological precept that ethical decisions in animal research require us to balance the harm we do to laboratory animals against the benefits we derive for humans and other animals. This precept – the notion that we can work out what is ethical by trading off one set of interests against another – is precisely what is denied by advocates of animal rights.
On the animal rights approach, it is always unacceptable to treat a sentient being merely as a “means to an end” – to use a sentient creature as a tool, or instrument, in pursuing one’s goals. On a radical version of this view, no benefit can justify violation of the rights of an individual, whether human or animal; so, where an experiment violates an animal’s rights, there is no reason to look for its expected benefits to humans or other animals. To find out whether an experiment is morally justified, we need only ask whether it respects the animal’s rights and preserves its dignity. The implications of this way of looking at matters are radical. Tom Regan and many other adherents of the animal rights view argue in favor of an abolitionist position.
On this version of the animal rights view, experimentation on animals should simply stop. It matters not that an experiment will cause only minor harm to the animals it involves. It matters not that this experiment is of extraordinary importance to humanity at large. The thing that matters is that every time an animal is used for an experiment, it is treated as a mere means to an end. This being so, animal experiments are unacceptable, period.
It is possible to imagine a less uncompromising, more moderate advocacy of the animal rights approach. The right to life – or more accurately, the right not to be killed – is regarded as basic by some influential proponents of animal rights. But one might be doubtful about this, partly because animals have a much more limited perspective on the future than we have. What matters to animals is that, here and now, they are well off, whereas a human has aspirations and worries that reach across his or her entire potential life-span. In light of this, one might suggest that animals have something like a right to protection from suffering, or certain levels of suffering. It could then be argued, perhaps, that all animals should be protected from suffering if this involves intense or prolonged pain or distress, which the animal cannot control.
The key idea of the animal rights approach is that there are absolute, nonnegotiable limits to what can be done to animals. Certain things should not be done to animals even if this means we are prevented from doing things that would have clear benefits outweighing any pain and suffering caused along the way. If the rights approach is characterized more loosely in this way, bans on certain kinds of experiment–like the one introduced in Danish legislation outlawing experimentation that causes strong pain or other forms of intense suffering to animals – look like an indication that the legislators have adopted a moderate animal rights view.
So the question, raised at the start of this section, about whether animals should be used in research, has no single answer that all will agree on. There are basically different views, of which we have here distinguished three prominent types. In the rest of this chapter, we shall not attempt to adjudicate between these views, but rather discuss the issues raised by animal disease models in the light of all three.
3 Does Animal Species Matter?
Animals of very different species are used in research. The choice of animal depends on the research in question, of course. It is also affected by the experience and expertise of the researcher, the facilities of the institution, by legislation, and sometimes by public discussion in the country where the work is carried out. Let us assume that, in the case of dementia research, if the matter being investigated requires an in vivo approach, it will require use of an animal that has a complex enough nervous system to actually have mechanisms for learning or memory formation. Even so, there are many species that fulfil that requirement, ranging from nematodes to chimpanzees.
Does the choice of species matter when it comes to ethical evaluation of the research? It does, as we will now discuss, but quite how it matters will depend on one’s ethical position. Those taking a contractarian approach will be primarily concerned with differing kinds of public sensitivity to different species. Those taking a utilitarian stance will focus on the capacity of animals of different species to suffer. But before taking a closer look at the different types of animals that can be used in dementia research (invertebrates, fish, rodents, and nonhuman primates), we need to introduce two topics that will be important in any discussion of the moral status of different species: sentience and the socio-zoological scale.
Sentience is the capacity to perceive or feel things. A sentient being is one that has its own experience of life, meaning that “there is something it is like” (7) to be that being. Scientific understanding of sentience (human and animal) is still limited. At present, neurobiology does not explain consciousness in terms of material mechanisms of the nervous system. We assume that other individuals are sentient, because they are behaviourally and physically similar to us. While this assumption is uncontroversial for adult human beings, when we extend it to nonhuman animals, the issues become more complicated, for here verbal evidence is unavailable, and the behavioral and physical similarities are smaller. Although common sense may posit sentience in many species, a scientific argument for attributing it must be based on systematically collected evidence.
Such a systematic approach is suggested by Smith and Boyd (8). To determine whether an animal has mechanisms similar to those that we know are essential for human subjective experiences, and whether an animal behaves in similar ways to sentient humans, we can consider a checklist of neuroanatomical/physiological and behavioral criteria determining the capacity to experience pain, stress, and anxiety in nonhuman animals. For any of the relevant experiences, this checklist will include the possession of higher brain centers and evidence of behavioral reaction to potentially nociceptive, anxiogenic, or stressful experiences. Further evidence will be added if these behavioral reactions are modulated by drugs, which have a known anxiolytic or analgesic effect in humans, and if there are peripheral nervous structures (including receptors, signal substances, and hormones) for each specific type of reaction, and a connection between these and the higher brain centers. The more of these criteria that are fulfilled, the stronger is the evidence that an animal is indeed sentient.
Looking at the way in which different taxonomic groups of animals fare on such a systematic analysis, there are two important lessons to be drawn. The first is that all vertebrate animals meet the criteria for sentience. When Smith and Boyd’s original analysis was published, such evidence existed only for mammals and birds, but over the last decade evidence of fish sentience has accumulated (9,10). The second lesson is that, for many of the invertebrates, we still know too little to be able to give a useful answer.
However, there is another way to approach the question whether and how animals matter. Thus, there is clearly a hierarchy of animals – a moral ordering that has been called the socio-zoological scale (11). The basis of the scale is that people rate animals as morally more or less important, and therefore more or less worth protecting, according to a number of factors. These include how useful the animal is, how closely one associates with the individual animal, how cute and cuddly the animal is, how harmful the animal can be, and how “demonic” it is perceived to be.
Today, in western societies, some companion animal species, notably dogs and cats, seem to be at the top of the scale. Among other animals, large carnivores and primates are at the top end of the scale. In the middle are large farm animal species such as cattle and pigs. At the bottom of the scale are pests or vermin such as rats and mice. Fish, which are cold and slimy, also appear to be quite low down the scale. Thus, among the animals used for research, there will be a hierarchy with nonhuman primates, dogs and cats at the top, pigs (etc.) in the middle, and rodents and fish near the bottom. Below rodents and fish, one finds insects and other invertebrates.
The socio-zoological scale is in many ways based on tradition and prejudice, and its use as a basis of animal protection can be criticized from both scientifical and ethical point of view. From both the utilitarian and the animal rights perspective, it is bound to seem unfair to discriminate animals solely on the basis of the scale – an unfairness comparable to racist treatment of humans. On the contractarian view, on the other hand, there is nothing problematic about treating animals in line with the scale, and thus giving more protection to primates and dogs than one does to rodents and fish. This is because, on the contractarian view, animals matter only to the extent that they matter to humans.
Whatever one’s ethical view, it is important to be aware that the socio-zoological scale is part of social reality. This reality is, among other things, reflected in legislation that has been introduced to protect animals.
To start at the very bottom of the socio-zoological scale, the fruit fly Drosophila melanogaster shows a number of age-related functional types of decline that are also evident in humans, including deficiencies in learning and memory (12). Parallels with human cognitive decline related to age and oxidative stress can also be demonstrated in Caenorhabditis elegans, making this even simpler organism (a nematode) a potential candidate for dementia research (13). The short lifespan and the ease with which these animals can be kept, in combination with the well-developed knowledge and technology deployed in manipulating the Drosophila genome, means that there are clear practical advantages in using these invertebrates as research models. Their use is generally perceived as less of an ethical issue than the use of vertebrates.
In fact, from the contractarian point of view, the use of invertebrate organisms in research does not seem to be an ethical issue at all. When fruit flies appear in the kitchen most people readily kill them without thinking further; and the fact that they have been widely used in laboratories since the beginning of the twentieth century does not seem to have caused much, if any, public discussion – this, despite the fact that genetic research on Drosophila involves major alterations of the bodily integrity of the flies. From the utilitarian point of view, the important question is whether invertebrates such as Drosophila and C. elegans are sentient. Given the difficulty of proving sentience, the question is perhaps better expressed thus: whether these invertebrates are likely to possess the capacity for suffering and pleasure. C elegans has very simple nerve organization; it also lacks one of the most important components for sentience: a central nervous system.< div class='tao-gold-member'>