Accept No Substitutes: The Ethics of Alternatives

Any model system which moves down the phylogenetic scale from the generally acceptable animal model will be considered an alternative.
—B. Taylor Bennett^[1]

It is common to argue that animal experimentation is justified by its essential contribution to the advancement of medical science. But note that this argument actually contains two premises: an empirical claim that animal experimentation is essential to the advancement of medical science and an ethical claim that if research is essential to the advancement of medical science, then it is justified. Neither premise looks weak; the first premise is an article of faith for most biomedical researchers, and the second is usually considered so obviously true that it goes unstated. In fact, however, both are open to challenge. In the logic of the case, only one of the premises needs to be shown false or moot in order to refute the argument. A number of other commentators have questioned the first,^[2] but it is the ethical premise that I find particularly wanting.

I think there are at least two ways to question it. The first depends on articulating the ultimate point of medical science. One can plausibly maintain that the justifying purpose of medical science is the health of humanity, or even more generally, the welfare of humanity. But if so, then medicine should be concerned as much as possible with preventing disease and injury, rather than with developing treatments and cures. Furthermore, as some have argued, we already know enough about the causes of most serious diseases to go a long way in preventing them. Our finite medical resources would therefore be far better spent on basic health care and public health education and campaigns than on further medical research.^[3] Animal experimentation would likely be radically reduced, if not eliminated, in the bargain.

The second objection to the ethical premise of medical research is about whether the ends of medical science are overriding. For even if a given research regime showed great promise to eradicate cancer, society would not necessarily approve it. That is why we have animal experimentation in the first place. Cancer and all other major human ailments would probably yield their secrets far more rapidly if we performed various grisly and fatal experiments on human subjects instead of on other animals. Yet it is surely the consensus of modern medicine and society that we would never perform such experiments on humans, not even with their consent.

So then the question arises: Why do we perform them on other animals? If we would not conduct lethal experiments on humans in order to advance medical science, then there is no absolute necessity to cure or prevent cancer.^[4] Yet one hears continually that we need to use animals in biomedical experiments, that animal experimentation is necessary. It is not.^[5] It is a choice made by human beings for their own benefit.

When we think about the costs and benefits of animal experimentation, it is natural—unfortunately— to assume that only the costs and benefits for human beings are relevant. But of course, there are other sentient beings whose welfare is affected, namely, the animals on whom the experiments are performed. For all sorts of reasons, this fact has been largely overlooked for most of the history of medicine. In the starkest form of the argument in favor of animal research, even the bare capacity of nonhuman animals to feel pain has been denied by both biomedical researchers and philosophers.^[6]

Of course, that view is by no means the contemporary consensus of the medical community. In the meantime, ethologists and others—including, indeed, animal experimenters—have confirmed what any pet owner already knows: nonhuman animals experience not only pain, but also distress and many other emotions.^[7] Plainly, then, whatever benefits accrue to human beings from animal experimentation must be weighed against not only the costs to humans (including the opportunity costs of rejecting promising treatments because of their inefficacy or harm to other species and diverting scarce resources from effective preventive regimes), but also the costs to animals. Arguably, too, assessing these costs means more than just tallying up the experiences of the animals in laboratories; interests that transcend these costs are also relevant. Most salient would be the interest in continuing to live,^[8] since the vast majority of laboratory animals are killed.^[9]

In the calculation of all of these kinds of costs to these animals, we must consider not only the quality and magnitude of their suffering and thwarted interests, but also the number of animals affected. These figures are difficult to compile with precision because there are no uniform reporting requirements for all animals used in biomedical experiments. However, the number is certainly in the tens of millions, and of these a significant subset are subjected to “unrelieved pain and distress of varying severity.”^[10] Against this “cost,” therefore, we are to weigh the benefit to human (and to animal) medicine in the long haul.

How is such a calculation to be carried out? There are two main problems here. First is that an actual cost—the animals’ suffering and thwarted interests—is being balanced against a speculative benefit, such as a cure for cancer or the discovery of a new analgesic drug. The greater the gap between the certainty of the former and the uncertainty of the latter, the less does the latter justify the former. The second problem for this sort of calculation is that a common measure is needed in order to compare the various values. In particular, one wants to know how pain and distress and lost opportunities are to be assessed across species.

Here one must be alert to the inevitable bias human beings apply in their assessments of other animals’ welfare and interests. The bias sometimes works to the apparent benefit of the animals, as when we treat pets as if they were members of our own human families. But in many other cases, we completely shut out of our mind what the animals must be experiencing. When animals are utilized for human purposes, such as providing food, clothing, and medicine, we simply fail to consider their actual suffering or denied freedom and so forth; we see these effects as somehow not measuring up to what a human being would experience in similar circumstances.

Absent from this way of thinking is an appreciation of the nonhuman animal’s own valuation of his or her way of experiencing the world. If each of us, whatever kind of animal we may be, has but this one life to live, might we not conclude that a rat’s life has as much value for him or her as a human’s for him or her? Indeed, might not the much shorter life expectancy of a rat be an argument that the rat’s few remaining years have greater, rather than lesser, value?^[11] On whom does the burden of proof rest with this kind of issue? Is the issue even resolvable? And if it is not, then should we give the benefit of the doubt to those over whom we have absolute power and in whose exploitation we have a strong interest?

Thus, the various considerations that bear on the ethics of animal experimentation. Can any conclusion be drawn? I would say yes. One is that the only sure reason we can give for animal experimentation is that we have the power and the desire to do it for our own human purposes. This is not really a justification; it is that “might is right.”^[12] (Some experimentation on animals is done to promote animal welfare. However, not only does it typically “sacrifice” the individual for the sake of the species, but also typically does so in service to the broader human purpose of exploiting the animals, as when seeking a way to maintain the health of animals who are penned in close quarters and then slaughtered for food.)

The two most commonly given alternative rationales are that other animals experience less suffering or loss than we would under analogous circumstances and that their suffering or loss matters less than ours would. But any such attempt to justify promoting our good by imposing “bad” on other creatures must be immediately suspect, given that it is self-serving.

A second conclusion is that anything short of abolishing animal experimentation altogether risks leaving the status quo virtually intact. Consider the position of rodents in the laboratory. Rats and mice constitute the overwhelming majority of animals used for biomedical research—certainly over 90 percent.^[13] Yet their welfare is systematically overlooked not only by animal users,^[14] but even by some animal protectionists. The most in-your-face manifestation of the former is the explicit exclusion of research rodents from the definition of “animal” by the federal Animal Welfare Act.^[15] But we find that even the so-called alternatives movement commonly contains a fatal loophole. For while a layperson may assume that the term “alternative” refers to the use of some wholly nonanimal method of research, testing, teaching, or training, in fact, it often means an animal “down the phylogenetic scale.”

Thus, developing alternatives to the use of animals can mean simply using a different animal. (To stretch this point to absurdity, all animals used in animal experimentation can be thought of as alternatives, since they are alternatives to human animals.) Furthermore, the characterization of the other animal—usually a rodent—as “lower” on a phylogenetic “scale” is arbitrary and disputed.^[16] The alternatives movement is therefore at risk of becoming a bait-and-switch con.

And it is even worse than that, for the very same animal (both species and individual) can be used as an “alternative.” This is due to two additional ambiguities. One of them is between an experiment on a whole animal and an experiment on tissue taken from an animal of the same species. The latter can be considered an “alternative,” but of course the animal is still bred, confined, and subject to various procedures. The other ambiguity is that “alternative” can refer to any attempt to reduce the number of animals in research, refine the procedures performed on them, or replace an animal subject with some other model.^[17]

But it is really only full replacement of animals in biomedical research that merits the name “alternative.” Any alternative to that understanding of “alternatives” is unjustified, not only in word but in deed.

Joel Marks is professor emeritus of philosophy at the University of New Haven and a bioethics center scholar at Yale University. His books include Ethics without Morals (Routledge, 2013) and Ought Implies Kant (Lexington Books, 2009). His main areas of scholarly interest are theoretical and applied ethics, which have most recently converged on animal ethics.