The Debate on Animal Experimentation

Original text Luciano Carlos Cunha · Volume IV, p. 79 of the PDF edition · English translation

20.1. Introduction

This chapter first examines the main arguments used in defense of animal experimentation and points out the problems with those arguments. It then presents some of the main arguments against animal experimentation.

20.2. The arguments from necessity and efficacy

The argument from necessity

One of the arguments most frequently employed by defenders of animal experimentation is that it is indispensable for scientific research. For example, Beauchamp (1999, p. 472) argued that granting nonhuman animals the right to life is something that should not be done because it could have negative effects on scientific research.

One way to respond to this argument is to point out that there is a whole range of research procedures that do not use nonhuman animals, such as inanimate and computer models, cell and tissue cultures, epidemiological studies, research on microorganisms, and so on.¹⁸¹. Therefore, animal experimentation would not be indispensable for research to continue.

It could be objected that there may be no substitute method for the use of animals in certain contexts. However, this does not necessarily imply that using animals is justifiable in such contexts. For example, imagine that in certain contexts the only way to carry out a piece of research were to kidnap and kill humans. That would still be insufficient to justify adopting such a method. This is because the mere fact that a certain means is the only way to achieve a certain end does not necessarily imply that there is justification for using that means. This holds whether one adopts a deontological perspective in ethics (where, for example, individuals would have the right not to be used as mere

resources, regardless of the consequences) or a consequentialist perspective (since, according to consequentialism, it would have to be shown that adopting the means in question produces the best consequences, taking into account everyone who would be affected, and not only in this case in isolation, but also taking into account the future ramifications of the consequences, including the way such an act would influence the future decisions of other agents). This does not mean that it is impossible to justify the use of a harmful means in certain circumstances. It means only that pointing out that a certain means is the only way to achieve a certain end is insufficient to establish that we are justified in using that means. Much more would have to be demonstrated.

The argument from efficacy

One of the frequent objections to methods that do not use animals in experiments is the claim that, at least in some cases, using animals would be more efficient. This objection therefore assumes that efficacy is the fundamental criterion for choosing among the various research methods.

Are we in a position to compare efficacy?

One way to respond to this objection is to deny that we are in a position to compare the efficacy of methods that use animals and methods that do not, since not enough work has been done on the latter, given that funding for research that does not use animals is minimal¹⁸².

Thus, even if the currently existing methods that do not use animals were not as efficient as those that use animals for the same purposes, there would be no basis for concluding that, if a similar amount of resources were devoted to both kinds of methods, the methods that do not use animals could not be as efficient as (or even more efficient than) the methods that use animals.

Moreover, the companies involved in research that uses animals invest a great deal of resources in promoting the idea that animal testing is necessary. These resources are

several orders of magnitude greater than the amount of resources devoted to developing methods that do not use animals¹⁸³. Thus, paradoxically, a large amount of resources that could be used to develop methods that do not use animals is used instead to spread the idea that the use of methods harmful to animals is necessary, on the basis of the claim that there are no other methods for the same purposes.

Does being more effective automatically justify a practice?

Another way to respond to the appeal to efficacy is to point out that, if the fact that a method is the most effective automatically justified using that method, then this would be a reason to prefer testing directly on humans instead of testing on nonhuman animals, since doing so would avoid problems in the results due to the difficulties of extrapolating data between different species¹⁸⁴. That is, the appeal to efficacy cannot justify experimentation on nonhuman animals¹⁸⁵. However, even though testing on humans is more effective, most of us (including many scientists who experiment on animals) would consider it unacceptable, for moral reasons, to do so. This seems to show that such scientists do not in fact accept the criterion they claim to accept.

In short, there are two main reasons why the appeal to efficacy does not justify the use of animals. The first is that the very idea that, because a method is the most effective way of achieving a certain goal, adopting it is automatically justified is highly questionable. The second is that, even if that idea were correct, it would imply testing directly on humans, since that would be more effective than using animals.

20.3. Would giving greater weight to humans justify animal experimentation?

A moral double standard

Defenses of animal experimentation reflect a moral double standard, one that considers the use of nonhuman animals in experiments acceptable (a use that is

often defended on the grounds that it would be more efficient than methods that do not use animals) while at the same time considering the use of humans under the same conditions unacceptable, even though it would be more efficient than methods that use nonhuman animals¹⁸⁶.

In other words, it is held that experiments on nonhuman animals are acceptable (even imperative) despite the harms such experiments cause, and that the same procedures are unacceptable in the case of human beings precisely because of the harms they cause. It is, therefore, a moral double standard based on the species of the victim.

The claim that the good of nonhuman animals does not matter

One way to try to justify this double standard would be to maintain that the good of nonhuman animals does not count to any degree. However, it would have to be explained why the good of animals does not matter, rather than simply assuming it from the outset¹⁸⁷.

Even the guidelines that endorse the use of animals usually maintain that the good of nonhuman animals matters to some degree, however minimal. For example, the Declaration of Helsinki, despite completely disregarding several important interests of animals (such as not being used as resources, not suffering at all, not being killed, and so on), upholds the need to protect animal welfare, in the sense of minimizing their suffering while they are being used as resources¹⁸⁸.

It is therefore contradictory for defenders of animal experimentation to claim that the good of nonhuman animals does not count to any degree and also to claim that such experiments are justified because they allegedly comply with animal welfare guidelines.

The claim that human interests should always weigh more

Another way to try to defend the double standard is to claim that, although the interests of animals matter to some degree, human interests would always matter more, whatever they may be. One problem with this view, however, is that it would imply that a trivial interest of a human (such as avoiding a light pinch on the arm) would carry more weight than the interest of a nonhuman animal (or even of billions of nonhuman animals) in not suffering the worst possible torture for years. This would be the height of arbitrariness, to say the least. As we saw in chapter 2, it is a blatant violation of the principles of equal consideration and impartiality.

Moreover, the guidelines regulating animal experimentation do not seem to accept this view, for by maintaining that some interests of nonhuman animals matter (however minimally), they are holding that the interests of nonhuman animals outweigh at least some human interests (since respecting these interests of animals costs humans at least some effort, however minimal, that could have been devoted to other purposes).

The claim that similar human interests should count more

Faced with the implication above, someone might then argue that it is not that all human interests are more important than all the interests of nonhuman animals, but only that the relevant interests of humans weigh more than similar interests of nonhuman animals.

A first problem with this position is that, if the interests of nonhuman animals are given some weight, even if much less, when we add up the interests of a large number of nonhuman animals, this already amounts to a significant weight. Therefore, the claim that nonhuman animals should receive some consideration, but less, implies accepting that there is a certain number of animals above which their interest in not being used as resources outweighs the human interest in using them.

That is, even giving greater weight to human interests, animal experimentation could not be considered acceptable once it exceeded a certain limit in terms of the number of animals used. And in practice that limit is routinely exceeded¹⁸⁹.

Moreover, a second (and main) problem with this position is that, as we saw in chapter 2, there seems to be no justification for giving different weight to similar levels of harms and benefits depending on the species of those affected. If so, then it is unjustifiable to give less weight to harms that befall nonhuman animals than to similar harms to humans.

20.4. Kantianism for humans and utilitarianism for nonhumans?

Another way to argue that the double standard is consistent is to maintain that we should adopt a pluralist approach: Kantian principles for human beings and utilitarian principles for nonhuman animals¹⁹⁰.

Kantian principles would prohibit using human beings as mere means. Utilitarian principles (which, according to this view, should be applied only to nonhuman animals), on the other hand, could accept such use when doing so is the most efficient way of maximizing utility (understood as the total aggregate balance of well-being).

There are two fundamental problems with this position, which will be spelled out below.

The first problem is that, even if the double standard in question were acceptable, the use of animals would be justified only under the conditions in which utilitarianism approved of it. However, utilitarianism would certainly take into account the sum of the interests of animals (giving equal weight to similar interests of humans). Since the number of animals used in experiments is estimated at more than 100 million individuals per year¹⁹¹, their aggregate interests would carry an enormous weight, one that would outweigh the aggregate interests of humans in using them¹⁹².

The second problem is that the position in question offers no reason for adopting this double standard: instead, it simply assumes that it is justified. In any other context, adopting Kantianism in some cases and utilitarianism in others would be considered a paradigmatic example of incoherence. For it to be accepted in this case, a justification would have to be presented, rather than simply assuming it as though it were obvious.

Sometimes an attempt is made to justify adopting this double standard by claiming that humans have autonomy, and that this would give rise to a protection against being used as resources¹⁹³. However, there are two problems with this move, detailed below.

The first problem is that an enormous number of humans do not have autonomy. If someone is human but not yet an adult, or is already an adult but lacks the cognitive capacities typical of an adult human, then they do not have autonomy. Therefore, according to this position, children and people with some cognitive impairment should be used in experiments in the way animals are currently used.

The second problem is that this move does not explain why it would be the capacity for autonomy that gives rise to protection against being used as a resource, and not some other characteristic (such as, for example, sentience). One might reply that autonomy gives rise to this protection because autonomous agents are harmed when their autonomous will is thwarted and, for that reason, must be objects of respect. However, sentient beings in general, whether or not they have an autonomous will, are also harmed when they are used in experiments, especially if they suffer and/or are killed. Therefore, it is completely arbitrary to say that the reason autonomous agents should not be used in experiments is that they would be harmed by it, and then to say that beings incapable of autonomy should be used in experiments despite being harmed by it.

20.5. Questioning the assignment of greater weight to human interests

In the two previous sections we saw that, even if greater weight were assigned to human interests, this would still be insufficient to justify animal experimentation. Now we will look at the central critique of the very idea that human interests should count more.

If we have reasons to give moral consideration to someone because that someone can be harmed and benefited, then it follows that the strength of the reasons for avoiding a harm and for pursuing a benefit depends on the magnitude of the harm and of the benefit, and not on the species of the one who bears it. This becomes clearer with the principle of equal consideration, which we examined in detail in chapter 2 and which will be briefly recapitulated below.

The principle of equal consideration¹⁹⁴ is a safeguard ensuring that individuals are not unfairly favored or disfavored in the consideration of the harms and benefits they might receive. The principle prescribes that harms and benefits of similar magnitude must be given the same weight.

For example, suppose someone claims that the suffering of individual A matters to a certain extent, but that the equivalent suffering of individual B, who likewise has done nothing to deserve it, matters less. In that case, this person is violating the principle of equal consideration by unfairly disfavoring individual B.

In the example above, the principle was violated because similar harms were given different weight. There is, however, another, even more extreme way of violating the principle of equal consideration: when lesser harms are given greater weight than greater harms.

For example, suppose someone holds that relieving A's mild pain is more important than relieving B's intense pain. Suppose that neither of them did anything to deserve the pains they suffer. In that case, B is being unfairly disfavored in an even more blatant way, since the harm B suffers is given less weight not only than similar harms to A, but even than lesser harms to A.

The vast majority of experiments using animals are violations of the principle of equal consideration of this more blatant kind, which gives less weight to the greater harm, and those that are not still violate the principle by giving different weight to similar harms.

For example, the vast majority of experiments on animals aim to satisfy minor human interests, such as obtaining new cosmetics or new cleaning products, by sacrificing greater interests of nonhuman animals, such as their interests in not suffering and not dying. It could be objected that there is research using animals that aims to fulfill important human interests, such as curing diseases or developing vaccines. However, as stated above, the fact that one of the interests at stake is an important interest is not enough to show that the principle of equal consideration is not being violated. In the case of animal experimentation, it is violated when the similar interests of nonhuman animals (such as their interests in not suffering and not dying) are given less weight. If the roles were reversed (that is, if the proposal were to kill humans to secure equally important interests of nonhuman animals), the practice would be considered unacceptable. This shows that animal experimentation is grounded in a biased principle, one that determines who will be the victims and who will be the beneficiaries on the basis of the species to which they belong, and not on any even minimally impartial criterion.

20.6. What is really the ethical problem with animal experimentation?

The following argument, attributed to the philosopher Charles R. Magel, is often used by animal advocates against animal experimentation:

"Ask the vivisectionists why they experiment on animals, and they will answer: ‘Because animals are like us.’ Ask the vivisectionists why it is morally 'OK' to experiment on animals, and they will answer: ‘Because animals are not like us.’ Animal experimentation rests on a logical contradiction¹⁹⁵".

Below we will see some reasons to think that this is not a good argument. This does not mean that animal experimentation is justified, since, as we saw above, there are several other good arguments against it. What we intend to show here is that the ethical problem

with animal experimentation does not lie in committing a logical contradiction, but rather in treating differently cases that are similar in their morally relevant properties.

Let us begin by observing that what the argument claims about animal experimentation could equally be said of cell cultures, computer models, or any other method that substitutes for the use of animals. For example:

"Ask animal advocates why they are in favor of testing on cell cultures, and they will answer: ‘Because cells are like sentient beings.’ Ask animal advocates why it is morally 'OK' to experiment on cells, and they will answer: ‘Because cells are not like sentient beings.’ The defense of methods that substitute for the use of animals rests on a logical contradiction".

The mistake in both arguments is the failure to distinguish between physiological similarities and similarities in morally relevant properties. Two beings can be similar in one sense but not in the other.

If we reformulate the two arguments, this time taking that distinction into account, we can see that the problem with animal experimentation does not lie in committing a logical contradiction, but rather in assuming that nonhuman animals differ from humans in their morally relevant properties. Let us see:

"Ask the vivisectionists why they experiment on animals, and they will answer: ‘Because animals are like us physiologically.’ Ask the vivisectionists why it is morally 'OK' to experiment on animals, and they will answer: ‘Because animals are not like us in their morally relevant properties’".

Now we can see that there is no logical contradiction. This does not mean that animal experimentation is justified. On the contrary. Both claims made by proponents of animal experimentation are probably false. For example, several authors have pointed out that physiological differences between species make the animal model scientifically inadequate¹⁹⁶. However, the bigger problem lies with the claim that animals are not like humans in their morally relevant properties (and this problem would remain even if there were no problem at all with the claim that animals are like humans physiologically). This point will be explained in more detail below.

The reason humans need moral consideration is not that they belong to the human species, or that they possess a certain set of capacities, a certain set of relationships with one another, and so on. They need moral consideration simply because they can be harmed and benefited. However, this shows that every sentient being is similar in what is relevant for receiving moral consideration, since every sentient being can be harmed and benefited (given that sentience is what makes someone capable of experiencing suffering and pleasure, for example).

Now consider the same adaptation of the second argument:

"Ask animal advocates why they are in favor of testing on cell cultures, and they will answer: ‘Because cells are physiologically like sentient beings.’ Ask animal advocates why it is morally 'OK' to experiment on cells, and they will answer: ‘Because cells are not like sentient beings in their morally relevant properties".

Now we can see that the animal advocates' position does not commit a logical contradiction either and, more importantly, that it rests on true premises. Substitute methods are probably much more similar to the organisms of the beneficiaries (whatever species those beneficiaries may belong to) than the animal model is. But, more importantly, these substitute methods are not similar to sentient beings in their morally relevant properties since, not being sentient, they cannot be harmed or benefited.

Animal advocates' frequent use of the argument attributed to Magel stems from the mistaken belief that, for an argument to be bad, it must commit a contradiction. An argument's containing contradictory premises is one way in which an argument can be bad. However, an argument can be bad without its premises contradicting one another. For example, one or more premises may be false. As we have seen, this is certainly the case with the premise that claims that sentient animals are not similar to humans in what is relevant for someone to be owed moral consideration.

20.7. Conclusion

If the arguments presented above are sound, then two important conclusions follow:

(1) Given what we saw in section 20.5, there is no justification for giving less weight to the good of nonhuman animals.

(2) Given what we saw in sections 20.2, 20.3, and 20.4, even if there were such a justification, it would not imply that animal experimentation is justified.

Therefore, there are good reasons to reject animal experimentation. This does not mean, obviously, that scientific research should be halted. It means only that there seems to be no justification for carrying it out by means of animal experimentation.