Info

a Swiss Federal Veterinary Office b Canadian Council on Animal Care a Swiss Federal Veterinary Office b Canadian Council on Animal Care c United States Department of Agriculture d Includes vertebrates except rats, mice, birds, and animals used for food or fiber research need to ask: in what way are these experiments unproblematic? The suggestion that this kind of suffering simply does not matter morally looks dubious. Surely it does matter — it is just that it carries less weight than more serious suffering. The alternative suggestion that minor suffering is justifiable is more plausible, but it is implicit in it that even minor suffering needs to be justified and hence, matters. In this sense, such suffering is not unproblematic.

The moral acceptability of animal experimentation is bound to be less questionable where animal suffering is minimized. To reduce such suffering, however, we will need to be quite clear about its nature — about the way or ways in which animals suffer. Our intuitive perceptions of what causes distress to animals are not always reliable. Procedures that appear unpleasant to human beings may cause an animal little distress. Thus, for example, no increase in levels of the stress-related hormone corticosterone was observed in mice kept in a room where other mice were being killed.5 Conversely, procedures that strike human beings as pleasant may be demonstrably stressful for an animal. Thus, the standard husbandry practice of cage-cleaning results in an immediate, although not necessarily prolonged, increase in aggression among male mice.6 Again, photographs of animals with electrodes attached to their heads may appear shocking when shown in animal protection campaigns, but cranial implants often look more inhumane than they, in fact, are. If the implants are positioned correctly and kept clean and free of infection, the most that the animal experiences in day-to-day life is some itching at the site of the implant. Brain tissue is not itself innervated, so the implant does not cause pain beneath the skull.

If we are to avoid these anthropomorphic assessments of animal distress and suffering, we will need to employ technically defined measures of the impact of experimental procedures on the subjective experience — measures defined in research on laboratory animal welfare. Over the last three decades, the study of animal welfare has become an established scientific discipline. Methods of assessing animal welfare have emerged from studies of the ways in which animals react, behave, and function in different experimental and everyday husbandry situations. Within the field, there is some variation in the approach to animal welfare, in that some animal welfare scientists emphasize health and biological functioning (e.g., Reference 7), whereas others hold that animal welfare is primarily a matter of the feelings of the animals (e.g., Reference 8). These approaches may diverge less than they appear to, because in practice, the factors they appeal to generally correspond. Nevertheless, in defining animal welfare, it may be helpful to ask why we are interested in animal welfare and suffering in the first place. The following remarks set out an answer to this question:

Animal welfare involves the subjective feelings of animals. The growing concern for animals in laboratories, farms, and zoos is not just concern about their physical health, important though that is. Nor is it just to ensure that animals function properly, like well-maintained machines, desirable though that may be. Rather, it is a concern that some of the ways in which humans treat other animals cause mental suffering and that these animals may experience "pain," "boredom," "frustration," "hunger," and other unpleasant states perhaps not totally unlike those we experience.

Marian Dawkins (1990) From an Animal's Point of View: Motivation, Fitness, and Animal

Welfare, Behavioral and Brain Sciences

This reasoning looks cogent. It suggests that any acceptable definition of animal welfare will take subjective feelings into account. If this is right, it follows that when animals are used as research models, it is the prospect of their experiencing pain or other kinds of unpleasant mental states that is the main cause for concern.

It should be acknowledged, however, that the occurrence of unpleasant mental states does not, by itself, imply that there is suffering. Such states are an unavoidable part of normal animal life and often serve as signals or behavioral prompts that help the animal to satisfy its biological needs. Sometimes, negative experiences are compensated for by corresponding positive experiences — few would argue that a hungry animal that finds food is suffering, even though the experience of being hungry is not pleasant. Unpleasant states, therefore, represent a welfare problem only when they are not compensated by corresponding positive feelings, or persist for an extended period of time, or occur frequently. Thus, a captive hungry animal that is not fed raises a welfare problem. So, too, does an animal that is strongly motivated to build a nest, or explore, but is kept in an environment where it has no opportunity to exercise these kinds of behavior.

Let us now turn to the measurement of animal welfare. Obviously, it is impossible to measure the intensity or duration of negative mental states directly. We therefore have to rely on physiological, behavioral, pathological, and other indirect parameters of feeling. Parameters of this kind include: changes in normal behavior; the occurrence of abnormal behaviors, such as stereotypies; altered activity of the hypothalamo-pituitary-adrenal axis or the sympatho-adrenal system; other hormonal changes; and modifications in body temperature, immunocompetence, plasma ion levels and body weight.9-13 These parameters are connected with the activation of various physiological and behavioral systems. It is worth pointing out that, since these systems affect not just welfare levels, but also the way an animal will react in an experimental situation, an animal's welfare status might well affect its suitability as a research model.14-16

Much animal suffering can be avoided through the proper use of anaesthetics and analgesics, and by careful handling procedures and improved housing systems. However, the fact that a great deal of pain can be controlled neither removes nor diminishes the remaining discomfort. An animal that has been operated on or is developing a medical condition, such as ascites or tumors, will inevitably experience a degree of discomfort. Equally, it is almost always necessary to deprive animals of some freedom and control over their environments to experiment upon them, and experimental interventions will, in many cases, involve a certain amount of distress even when they are carried out with care. Welfare problems caused by restrictive housing are often overlooked, but they are clearly important, because they will affect not just the experimental animals, but control groups and breeding stock as well. Behavioral restrictions imposed by standard housing might be stressful either because animals experience frustration when species-specific behaviors cannot be carried out or because animals have no opportunity to control their environments.1617 In most cases, the experimental protocol does not require restrictive housing.

In most countries, direct cruelty and the infliction of unnecessary suffering are illegal. In practice, the focus on unnecessary suffering will obviously permit some animal suffering — namely, that which secures a sufficiently important benefit for human beings. But in this formula, what is to count as a sufficiently important benefit? Is it enough that we will understand the function of all the genes in the mouse genome, or find an effective vaccine for HIV, or develop new agricultural pesticides and test their toxicity? And what should we say about the common situation in which the benefits of a line of research are, as the research gets underway, uncertain? Can a potentially large benefit that cannot be guaranteed be sufficiently important to license animal suffering?*

0 0

Post a comment