Last update: May 2024
Contact: Fabian Fischbach

Suggested citation

Number of laboratory animals: Germany and Europe

Applicability of the results of animal experiments

It is only since the advent of the modern era that animal experiments have been conducted on a significant scale. Since then a broad-ranging debate has raged on the acceptability of animal experiments. From the very outset of the debate, opponents of animal experiments have asserted that the insights yielded by animals cannot be applied to humans and are therefore largely useless. This criticism is directed both at the findings of basic research (e.g. the "mouse model") and at the results of drug tests performed on animals (see Section II). At issue was - and still is - the question of whether different species (such as human and mouse) react to the same substances in the same way on account of the structural and functional similarity of many organs or whether the effect of substances on organisms is more heavily species-specific. Were the latter to be the case, animal testing of substances would, for example, offer only an illusory sense of safety. History lends support to both standpoints: on several occasions the results of animal experiments have misled scientists into formulating incorrect research hypotheses (e.g. in research into poliomyelitis (polio) or lulled them into a false sense of security when testing product safety (as in the case of Contergan (thalidomide)) In other cases the effects observed in animal experiments did prove to be applicable to humans. The German Research Foundation (DFG), the central self-governing organisation that promotes research in Germany, estimates that animal experimentation can predict "desirable and roughly 70 % of undesirable effects on humans" (DFG (2004): Tierversuche in der Forschung. Bonn: Lemmens Verlags- und Mediengesellschaft, 2004: 18).

Alternative methods

Animal experiments can be replaced with alternative methods in a variety of areas. For example, numerous experiments are currently performed on cell cultures. By way of differentiation from experiments on living organisms (in vivo) such "test tube" methods are referred to as in vitro. Artificially generated cells of various tissue types, so-called organoids, are playing an increasingly important role in this context. Computer simulations may also serve as a substitute for the use of laboratory animals, since they help to predict how substances will act on the body. The extent to which alternative methods may replace animal experiments in the near future is a matter of some controversy. At least in the field of cosmetics research it is envisaged that safety testing on animals will be completely replaced by alternative test methods. Researchers point out, however, that even in the future it will not be possible to entirely do away with animal experiments - especially when it comes to testing pharmaceutical products: the complexity of an intact organism is necessary in order to verify all the effects of a substance. In the fields of neurobiological fundamental research and of research in infectious diseases, for instance, research with non-human primates is still irreplaceable at the current moment, according to several researchers.

Research practices

Research practices can take different forms, each with different levels of stress for the animals used. Throughout the EU, a distinction is made between low, moderate and severe levels of stress on the animals. A fourth category comprises experiments in which 'no restoration of vital function' takes place, which means that the animals are painlessly euthanized at the end of the experiment. The classification of degrees of distress is unclear for various reasons. On the one hand, there is controversy among experts as to which interventions on animals are stressful for them and to what degree. The underlying EU directive only provides vague criteria for classification and assessments are primarily based on examples. For instance, the testing of non-invasive imaging procedures in animals under anaesthesia, the administration of substances with expected minor effects or the breeding of animals with genetic modifications that are not considered to be restrictive are considered to be low-stress, associated with "short-term mild pain, suffering or distress", without long-lasting negative effects.

Moderately stressful procedures are those in "which the animals are likely to experience short-term moderate pain, suffering or distress, or long-lasting mild pain, suffering or distress as well as procedures that are likely to cause moderate impairment of the well-being or general condition of the animals" (EU Directive 2010/63/EU). Examples include surgical interventions with persistent post-operative pain or food deprivation for 48 hours in adult rats. Severe distress refers to pain, suffering and distress of great intensity or duration. Experiments that fall into that category are, for example, toxicity tests with expected deaths, organ transplants with potential rejection, or the complete isolation of social species for long periods of time. On the other hand, what can be considered stressful also varies greatly depending on the species, the age of the animals, and their previous health condition. For example, animals that are usually embedded in complex social structures are more likely to suffer while beeing kept in isolation than those that live solitary lives in their natural environment. In principle, it is a matter of conjecture as to what can be considered stressful for which animal species and to what degree. On the one hand, humans do not have direct access to the perspective of other species and can only access this through scientifically plausible theories. Secondly, according to many approaches, suffering is experienced very subjectively and therefore cannot be easily quantified.

The moral status of animals and humans

The moral permissibility of animal experiments can be examined along several lines. Firstly, the question arises as to what role animals generally play in moral considerations and what demands can be derived from this. A central problem that plays a role here, depending on the theoretical approach, is the question of how conflicting interests between humans and other animals can be mediated or which human interests are strong enough to override non-human interests. A third problem area relates to the legitimacy of different practices in the experimental context in connection with the special needs of different animal species. In the following, different approaches concerning the moral status of animals will be listed and their respective consequences for animal experiments will be pointed out. The problem of conflicts of interest will then be addressed in a separate chapter.

The answer to the question of whether animal experiments are ethically acceptable does not rest merely in the fact that for many people (for example for consumers or patients) they are useful, possibly even life-saving. On the contrary, it is necessary to ask whether and to what extent human benefit justifies the suffering and death of animals. The moral status of animals plays a decisive role here.

The concept of moral status aims to determine the circle of living beings that must be morally taken into account when making decisions, i.e. to include their interests in one's own (human) considerations for their own sake. Taking such interests into account can, in principle, lend decisions an ethical dimension. In the debate about the moral status of non-human animals, different characteristics are cited that make it necessary to take them into moral consideration. Depending on which theory is advocated here, different assessments of the permissibility of animal experiments arise. These depend not only on whether animals are granted a moral status at all but also on whether such a status allows for different degrees and how different living beings are to be placed along such a scale.

In this respect differences in status may be presumed both within the human species (for example between an embryo in an early stage of development and a grown human being) and - as is relevant to the present context - with an eye to different species. For example, § 8 of the fifth section of the German Animal Welfare Act (module: Animal Welfare Act) makes a species-dependent distinction between experiments on vertebrates or cephalopods on the one hand (such experiments are subject to authorization) and crustaceans on the other (experiments on these are only subject to notification).

In the following, a fundamental distinction is made between approaches in animal ethics that grant animals their own moral status and those that don‘t. Scaled approaches, according to which animals have a "lower" moral status than humans, are dealt with in the following chapter in the context of dealing with conflicts of interest.

1. Anthropocentric approaches

The core thesis of moral anthropocentrism is that humans have a special status in moral terms or that humans are the only living beings with a morally binding intrinsic value. This view has various historical roots and uses different patterns of justification. These are often based on the fact that humans have special characteristics that are not or not fully developed in other animals and that are of central importance for ethical consideration. Other approaches refer, for example, to the functioning of ethics as a system based on reciprocity. The argument here is that no moral demands can be made on animals and that they should therefore not be afforded moral protection. They are excluded from the ethical system in principle. Both forms of justification are often closely linked. For example, the deficits of non-human animals can also be used to classify them as morally incompetent.

If humans are the only living creatures that "count from the moral standpoint", they are under no obligation to respect animals: their use or harm does not violate any ethical precepts. Yet even within the framework of such a radically anthropocentric – i.e. human-centred – perspective obligations may be established in relation to animals. In this case, however, obligations do not exist vis-à-vis the animals themselves (since they have no inherent moral value); rather, they are indirect or derived obligations, in other words obligations that humans have in relation to animals but which are grounded in their obligations to themselves or to fellow humans.

Justification of the prohibition of cruelty to animals without recourse to an inherent moral status of animals was provided inter alia by Immanuel Kant within the scope of his ethics. Kant's argument is to be found in a chapter of "Metaphysik der Sitten" ("Metaphysics of Morals") (§§ 16-18). Kant does not substantiate the prohibition of animal cruelty on the basis that anyone who torments animals is doing them an injustice, but rather with the assertion that the person responsible for cruelty to animals is weakening himself in his capacity to commit moral acts. He is thus in violation of an obligation that he has to himself. What is more, animal cruelty impairs the ability to empathise with the suffering of others (including other humans). Since this capacity is, however, "very useful" for man's co-existence in a society, anyone who wilfully puts it at risk is in violation of an obligation to his fellow men. Such arguments against animal cruelty are referred to as brutalisation arguments or pedagogical arguments.

For animal experiments, it could be deduced from this that they are only justified if they are not associated with unnecessary harm and are aimed at a clear goal. This could be used, among other things, to formulate an argument against basic research, especially if it involves great suffering for animals. In principle, however, against this theoretical background, human interest can also justify a high degree of animal harm in an experimental context and even present it as morally welcome, especially if there are morally commendable intentions behind it, such as helping others through the development of new medicines. For example, various directives and regulations at European regulatory level stipulate that animal testing must take place as part of substance and process testing before these substances come into contact with human test subjects or consumers. This is intended to ensure the safety of humans by using laboratory animals.

The understanding that the rough and cruel treatment of animals is not wrong per se, but only indirectly through the consequences for one's own moral character and the co-existence of mankind, was criticised early on, inter alia by Arthur Schopenhauer.

One of the objections raised against the basic argumentation strategies of anthropocentrism is that it is based on essentialist assumptions about "human nature", which cannot be upheld scientifically and would consequently lead to the exclusion of some people from the moral community. Against this background, a reciprocity argument is also difficult to defend, as the central issue for moral consideration does not seem to be whether the living being in question is morally capable, but whether it is vulnerable in a decisive respect.

Many authors now consider it more plausible to assume that harming sentient beings is morally questionable as such and in relation to them. Accordingly, an ethical demand to avoid animal harm is not about the effects that cruel behavior can have on interaction with other humans, but about the treatment of the animals concerned per se.

In ths context the preamble to the Council of Europe's 1986 „European Convention for the Protection of Vertebrate Animals used for Experimental and other Scientific Purposes“  can be cited, which is based on the belief that there is a "moral obligation" for humans to "give due consideration" to the "capacity for suffering" and "memory" of animals, but that they must also use animals in their quest for "knowledge, health and safety“.

2. Non-anthropocentric approaches

As in the above-mentioned preamble, reference is made, for example, to the sentience or interests of non-human animals to justify their moral status. The two resulting approaches, which can be described as the animal interests position and the animal rights position, currently dominate the debate on the appropriate treatment of animals. More recently, relationalist approaches have also gained relevance, which determine the morally permissible treatment of non-human animals based on the relationship we have with them.

2.1. Animal interests position
Peter Singer makes the moral status of living organisms dependent upon their capacity to have interests (for example in survival and freedom from pain). All living creatures that have interests in the same way have the same moral status.

This view results in two consequences: an upward revaluation of the moral status of animals capable of having interests and a downward revaluation of the moral status of human organisms that are unable or have a reduced capacity to have interests. Research on human embryos, for example, would no longer pose an ethical problem (since embryos do not yet have an interest in freedom from pain or survival), whereas painful experiments on mice would constitute a serious moral evil. In his book "Animal Liberation" Singer coins the phrase "speciesism" to describe the view that living creatures such as humans and mice with an equal interest in a pain-free existence could have such a different "value" that the one can be used to benefit the other. For Singer speciesism is a form of discrimination, just like racism or sexism: one group of living creatures is disadvantaged by another without the existence of any morally relevant reasons. Speciesism is thus a form of group egoism of mankind directed against non-human creatures.

What are the implications for the field of animal experiments? Singer does not support an absolute ban on animal experiments (in contrast to the proponents of animal rights). Nor does he speak out in favour of an absolute ban on experiments on humans. Due to their self-awareness and awareness of the future (most) people, in Singer's assessment, have a stronger interest than animals in not being misused as research objects. What is more, humans attach significantly greater importance than animals to their own survival on account of their orientation towards the future. Based on man's more extensive interests, Singer believes that to a certain extent the use of animals in biomedical experiments is more justified than the use of human subjects. However, Singer is of the opinion that in moral terms experiments conducted on humans who - due to a lack of cognitive and emotional faculties - have a limited capacity to have interests comparable to that of higher animals (for example infants or the severely mentally handicapped) are of an equal status to certain animal experiments.
Singer's position is pathocentric, meaning that he calls for the attribution of an equal moral status to all living creatures with a capacity for feeling. Yet it has long been disputed whether animals are capable of feeling at all and, if so, the extent to which they have this capacity. In the modern era it was widely held that animals were incapable of thinking or feeling. This can now be considered a scientifically outdated view, at least for all vertebrates. In the "Treaty on the Functioning of the European Union" (TFEU), the member states of the EU also commit themselves "[i]n formulating and implementing the Union's [...] policies, [...] [to,] since animals are sentient beings, pay full regard to the welfare requirements of animals, while respecting the legislative or administrative provisions and customs of the Member States" (Article 13 Title II TFEU).

2.2. Animal rights

In accordance with the "rights view", as it is referred to by Tom Regan, its founder, the essential quality that a living organism must exhibit in order to be a bearer of rights is its "being the subject of a life". Every living creature with its own individual wellbeing has an inherent value and is therefore not merely a means for foreign ends. To this extent the animal interests position (see above) and animal rights position are similar.
The call for animal rights, however, envisages further-reaching consequences than the call to take account of both animal interests and human interests. In contrast to the animal interests position, the rights position asserts that all living creatures that are the experiencing "subject-of-a-life" should be protected by individual rights. At stake here, then, is the question of whether the concept of rights - as it exists with respect to humans - can and should be extended to parts of the animal world. Were animals to possess individual moral rights like humans, animal experiments would then be ruled out even if they promised an outstanding benefit - just as (forcibly conducted) experiments on humans are unacceptable under all circumstances irrespective of the benefit for the common good. Proponents of a rights position therefore reject animal experiments just as they do the consumption of meat in general.

One of the objections sometimes raised against the rights position is that rights as such only exist as a consequence of their mutual recognition (which animals are not capable of). Why should animals have rights if they lack insight into their significance and the possibility of acting accordingly? Animal rights supporters counter this objection with the "argument from marginal cases": nor do human beings have to be morally capable and rational in order to be the bearers of rights (e.g. infants, severely mentally handicapped persons or coma patients). In these instances the legal profession would ensure protection of their rights.
The call for human rights for anthropoid apes is a form of minimal requirement for proponents of the animal rights position and animal interests position.

2.3. Compassion for animals

As the most prominent representative of ethics of compassion, Arthur Schopenhauer, differing from representatives of the animal interests and animal rights position, sets aside the attribution of a moral status. By doing so, compassion ethicists attempt to bypass presuppositions concerning commonly shared values such as human dignity. This is due to the fact that these presuppositions are to constitute moral statuses while they are at the same time not being accepted or cannot be accepted by everyone.

Similar to Singer, for Schopenhauer the circle of such members of a moral community who are worthy of protection is constituted of those beings capable of suffering.  By means of her compassion with other creatures, the human being recognizes the worthiness for protection of the vis-à-vis and the responsibility to protect this vis-à-vis from suffering. Living beings capable of suffering also include animals whose need for protection Schopenhauer nonetheless esteems to be lower than that of the human being. This owes to the fact that a living being’s ability to suffer is dependent upon its intelligence and that it is hence the human being who possesses the highest quality of suffering.

According to this, meat consumption and the keeping of farm animals are legitimate as long as a death free of pain is warranted and the consumption is essential for survival, and, in the second case, the keeping of the animals is not excessive. Schopenhauer generally objects animal experiments, which cause suffering for the animals. Analogously to the justification of meat consumption, he nonetheless allows for animal experiments if they cause almost no pain and are essential for the survival of the human being.  

Based on Arthur Schopenhauer’s theory, Ursula Wolf developed the approach of a generalized compassion as an extension of the ethics of compassion. Apart from values and the assignment of status, this approach is characterized by its attempt to justify rights and obligations for the reduction of suffering.  Schopenhauer had spared a justification of rights and obligations; for him, the basic rules of ethics are derived from compassion.  Wolf, by contrast, derives from compassion duties to protect concerning all living beings capable of suffering; this does, however, only apply to human beings since only human beings have at their disposal the necessary means for reflection for moral action. In comparison to Regans animal rights position, Wolf’s obligations rest on the foundation of compassion, while Regan links rights and obligations to the living beings‘ inherent value.

Those people are members of a moral community who have at their disposal the ability to suffer. The higher dimension of protection of those living beings that have an, at least basal, self-conscience needs to be kept in mind. Overall, Wolf refuses a differentiation in status between humans and animals by not considering the empirical rationales as substantive and by generally regarding all living beings capable of suffering as equally worthy of protection.

Concerning the issue of the legitimacy of animal experiments, the following standpoint arises in Wolf’s argumentation: The higher value of the human being based on empirical facts such as an allegedly higher ability to suffer due to her intelligence is rejected by Wolf as morally irrelevant; according to Wolf, they do not justify a gradation in status and hence no animal experiments which cause suffering to animals.

Similar to proponents of the animal rights position, approaches to the ethics of commission are oftentimes being criticized by means of the argument that moral consideration is only sensible concerning those living beings that are capable of moral consideration; this means that due to their capability, only human beings should be included in the circle of those that are directly worthy of protection. Wolf replies to this point that the specific capability of human beings to moral action should also be applied toward animals exactly because human beings have such capability at their disposal.

2.4. Relationalistic approaches

Compassion ethics includes some aspects that are also found in contemporary, so-called relationalistic approaches. For example, the possibility of feeling compassion for a particular animal depends not least on the relationship we have with it. It makes a difference, therefore, whether the encounter takes place in the animal's natural environment and it possibly sees us as prey or whether it is in a relationship of dependency with humans and is reliant on their goodwill. Similarly, it is often easier to feel compassion for animals that are perceived as aesthetically pleasing or cute than for those that look repulsive and frightening to many people. Both are aspects of the relationship that people can have with certain animal species. In addition, there are other aspects within the relationship between humans and animals on the basis of which responsibility or consideration for animal life can be justified. This includes the usefulness of a certain animal or animal species for humans or the common history that a person shares with another animal or a group of people with a certain animal species, as well as possible dependencies and obligations that arise due to the keeping of the animals, their living conditions (directly) influenced by humans or even their genetic origin (breeding, domestication).

Relationalist approaches attribute an intrinsic value to non-human animals on the basis of the relationships of responsibility that humans have with them. Accordingly, obligations towards individual animals or animal species arise here from the actual or potential relationship, which can be justified by opportunities for interaction or dependencies (in both directions). A fundamental distinction must be made here between personal and impersonal relationalism: Personal relationalism assumes that the moral significance of animals depends on how we as individuals can relate to a particular animal.

Theories that can be subsumed under this umbrella term, including both sentimental and feminist approaches, assume that morality is generally a matter of extending our natural attitudes such as sympathy and care towards other living beings. In contrast, impersonal relationalism and the theories associated with it are not based on forms of morality that arise from personal attitudes, but are based on duties of justice that exist due to particular types of impersonal relationships, such as the contribution to social cooperation. Based on the Rawlsian idea of justice, it is assumed that all beings (including animals) that contribute to the production of social goods through cooperation and participation are entitled to consideration under the concept of fairness and that humans therefore owe them a fair share, for example in the form of better living conditions or special rights. Such participation of animals exists, for example, when they are used for food (e.g. as farm animals), for scientific research (e.g. as laboratory animals) or for the individual support of humans (e.g. as guard dogs or guide dogs).

One theory that combines both personal and impersonal relationalism is Clare Palmer's. Here, diverse and divergent human duties towards animals are grounded in the various types of morally significant human-animal relationships, which are not limited to affection, companionship and domestication, but also include a range of causal relationships that consciously or unconsciously influence the lives of animals. In contrast to wild animals living in the wild, humans have a particular moral responsibility towards domesticated animals because, depending on the degree of domestication, they are dependent on humans and the vulnerability created by their particular living and existence circumstances demands explicit duties on the part of humans.

2.5. Biocentrism

Based on the approaches presented here, it may only be possible to argue for the protection of some animal species, while others, those that are probably not sensitive to pain in a sense that we can understand, are systematically excluded from moral consideration. For example, David DeGrazia, a representative of pathocentrism, i.e. an animal interest position, points out in several essays that there is convincing empirical evidence that insects are not sensitive to pain and therefore we have no moral duty towards them. In the context of animal experiments, this means that anything could be researched on insects in any way, and the same applies to all other creatures that are not sensitive to pain. Such an argument is based, on the one hand, on the assumption that the phenomenality of insects or other non-pain-sensitive animals is fully understood and, on the other hand, that all relevant forms of suffering express themselves in a way that is accessible to humans. From a skeptical perspective, both of these assumptions can be questioned and point to an anthropocentric orientation even of non-explicitly anthropocentric animal ethical positions.

Biocentrism attempts to escape a limited human perspective and postulates that all living beings, including those that are not obviously sentient or capable of interest, are morally worthy of consideration, i.e. should be protected for their own sake. In practice, however, this attitude is associated with major challenges. For example, it seems impossible to treat bacteria or other single-celled organisms with the same moral caution as mammals. A central problem here is to weigh conflicting interests against each other. Accordingly, advocates of biocentrism often find it difficult to call for clear and radical measures. One well-known representative of a biocentric position was Albert Schweitzer, who did not categorically reject animal experiments himself, but called for a more careful treatment of animals (and plants) overall.

Conflicts of interest and trade-offs

All approaches are capable of establishing a fundamental moral status for non-human animals. However, with the exception of animal rights positions, this says nothing about how the interests of individual living beings are to be placed in relation to one another. The problem of conflicts of interest lies in how different interests of different moral subjects are to be weighed against each other, for example in the case of moral dilemmas. If the fundamental moral status of non-human animals is assumed, a fundamental moral challenge arises for medical research, among others. On the one hand, it is important to preserve the health of other people in the long term, for example by developing treatment options for emerging diseases. On the other hand, despite the increasing number of alternatives, medical research is still dependent on research on living beings in order to test the effectiveness of its results, whereby in many cases the fundamental interests of the research subjects are violated. In other words, there is a conflict between society's interest in medical progress and an individual interest in the physical integrity of individual living beings. How to deal with such conflicts depends on various theoretical assumptions in connection with the attribution of moral status.

At this point, the distinction between scaled and non-scaled or absolute theories of moral status is crucial. While the latter rigorously determine the moral intrinsic value of a living being, often on the basis of a single, decisive criterion, such as capacity for suffering, and do not permit any gradations, scaled theories open up the possibility of establishing a hierarchy between different moral addressees. In many cases, such approaches are based on the view that the characteristic constitutive of a moral status can be more or less pronounced or that more or fewer criteria from a list of constitutive characteristics can be fulfilled. Assuming the fundamental moral status of non-human animals, there are, therefore, two further assessment scenarios for the evaluation of moral permissibility: Either the moral status of laboratory animals is equated with that of humans or, although the self-interests of animals are recognized in principle, their moral status is subordinated to that of humans. Hypothetically, it could also be assumed that the moral status of humans is subordinate to that of other animal species, but a general, species-level argument for such a position can be considered unusual.

Supporters of the moral permissibility of animal experiments generally have no theoretical means at their disposal to advocate species egalitarianism, i.e. the view that all animals or all living beings are equally morally significant, if they do not wish to contradict this very view. Ultimately, in very few cases do laboratory animals or their species benefit in any way from the experiments carried out on them and it is questionable whether they can be compensated for their suffering or whether such compensation would make a moral difference. Instead, a recourse to scaled theories is necessary, combined with the assumption that human interests are fundamentally to be weighted more heavily than those of other animals. A common argument to this effect goes as follows:

1. It is ethically imperative and appropriate to conduct scientific research to augment medical resources and treat human disease.

2. The more important it is to treat human disease, the more heavily this importance can be weighted in determining the ethical appropriateness of a research practice to study the disease.

3.  Humans are more important and of greater moral value than other animals.

4. It is sometimes appropriate to instrumentalize animals for research on human diseases.

The philosopher Jürgen Habermas submitted another suggestion, which tends more towards relationalism, of what the reasoning of such a position could be. He allows a genuine moral status to animals, which depends on the level of social interaction with humans. This seems to display well that the moral intuition of human behaviour is especially relevant towards more sophisticated animals.

Also the animal protection act, especially the obligation to produce proof of indispensability as well as the ethical tenability of the animal experiments, approximately confirms this understanding.

Suggested citation

German Reference Centre for Ethics in the Life Sciences (2024): In Focus: Animal Experiments in Research. URL https://www.drze.de/en/research-publications/in-focus/animal-experiments-in-research [date of access]