One of the main challenges that animal ethics has to confront is the question of how to deal with harm faced by non-human animals in nature. The fact that there is a great deal of suffering in the wild, which surpasses that caused by humans leads not only to attempts to rise to the challenge but also to dismissals of the moral obligation to alleviate non-human animal suffering altogether. The main question of this paper is concerned with whether the reductio of the natural-harm-argument can be avoided by disvaluing non-human suffering and death. According to the natural-harm-argument, alleviating the suffering of non-human animals is not a moral obligation for human beings because such an obligation would prescribe human intervention in nature for the protection of non-human animal rights and/or interests.Footnote 1 This, the natural-harm-argument claims, is absurd.

I shall first discuss the natural-harm-argument in some detail, concentrating especially on the predation argument, which is the version in which the natural-harm-argument appears most frequently. I shall go on to discuss three of the most common responses to the natural-harm argument, according to which human intervention in nature to alleviate harm to non-human animals is not morally obligatory and in some instances not permissible: The Moral Agency Argument, the Argument from Hubris and the Implementation Argument. These reject the conclusion that humans should intervene in nature for the sake of non-human animals. By adapting these arguments, I shall consider two ways of retaining the moral obligation to intervene in nature while avoiding the reductio of the natural harm argument by formulating two constraints to the moral obligation of human intervention in nature: The practicability constraint and the hubris constraint.

In the second part of the paper I shall consider an alternative attempt at retaining the obligation to alleviate non-human suffering while avoiding the reductio of the natural-harm argument. This approach takes the reductio seriously and considers it to be a challenge to generally accepted views on suffering and death. The obligation to intervene in natural processes to alleviate suffering and death to non-human animals exists only if these are considered to be bad enough to create a moral demand towards its prevention. If suffering and death were not as bad as to create this demand, then they would not provide moral reasons to alleviate or prevent them to the same degree. Disvaluing non-human suffering and death can be accomplished along two lines. The first line questions whether human and non-human suffering and death are different in relevant respects to require different types of responses; i.e. intervention in nature in the former case but not in the latter. The second line questions how bad suffering and death are as such.

I shall end by evaluating the first of these ways of responding to the natural-harm-argument. I shall conclude that the attempt to avoid the reductio of the natural-harm-argument by dismissing death can only work with an anthropocentric bias, which accords to non-human suffering and death a fundamentally different value than to human suffering and death. I shall argue however that death and incapacitating suffering may indeed be more or less bad according to certain capacities of the individual in question. While this difference by itself is not enough to prescribe non-intervention, taken together with the limits to human agency laid by the hubris and practicability constraints, I shall conclude that an obligation to intervene in nature for the sake of non-human animals exists as yet in particular cases but not globally.

1 The Natural-Harm Argument and Some Responses

The allegedly absurd results of considering the combat of non-human animal harm as a moral obligation comes out best in the Predation Argument. In an influential formulation, the reductio of the Predation Argument is as follows:

  1. 1)

    „Suppose humans were obligated to alleviate avoidable animal suffering

  2. 2)

    Animals suffer when they are preyed upon by other animals

  3. 3)

    Therefore, humans would be obligated to prevent predation

  4. 4)

    But such an obligation would be absurd

  5. 5)

    Therefore, contrary to the hypothesis, humans are not obligated to alleviate avoidable animal suffering“(Sapontzis 1984, 27)

We can easily extend this argument to include not only predation as a source of harm but other natural causes as well:

  1. 1)

    Suppose humans were obligated to alleviate avoidable animal suffering

  2. 2)

    Animals suffer due to natural causes, e.g. predation, famine, disease and natural catastrophes

  3. 3)

    Therefore, humans would be obligated to intervene in nature, e.g. by preventing predation

  4. 4)

    But such an obligation would be absurd

  5. 5)

    Therefore, contrary to the hypothesis, humans are not obligated to alleviate avoidable animal suffering

Line four of the argument is what turns it into a reductio and requires some clarification. The sense of absurdity is presupposed and seems to stem from strong intuitions. One reason for the apparent absurdity of the suggestion of human intervention in nature to alleviate avoidable animal suffering is the fact that its scope and therefore the area of necessary intervention is unimaginably great: „The total amount of suffering per year in the natural world is beyond all decent contemplation. During the minute it takes me to compose this sentence, thousands of animals are being eaten alive; others are running for their lives, whimpering with fear; others are being slowly devoured from within by rasping parasites; thousands of all kinds are dying of starvation, thirst and disease” (Dawkins 1995, 131). An indicator for the rarity of living healthily till beyond a successive reproductive age in the wild is the discrepancy between the average number of offspring that adults of various species produce and the fact that their population remains fairly constant over countless generations – as long at least, as their ecosystems remain intact. While it is common to glorify nature as a place of perfect harmony and equilibrium, it can, in fact, be considered as the cause of a great deal of suffering, in which the total amount of suffering outweighs the total amount of well-being (Dawrst 2009; Clarke and Ng 2006).

Nevertheless, although the predation argument is used as a reductio against equal concern for non-human animal interests, the evoked image of Sisyphean human agents, diving in oceans trying to prevent the big fish from eating the little one; endlessly labouring in broom cupboards and jungles to keep spiders from trapping insects and other nonsensical images contribute to overlooking the fact that humans already are and have for a long time been intervening in nature for the sake of non-human animals.Footnote 2 This does not happen only in the case of domesticated animals, in which intervention is often only a means to the goal of realising human interests. It is also not the case that it is out of purely environmental concerns for, say the good of a species that individual animals are rescued from a variety of life-threatening situations they find themselves in. In fact, human intervention in nature for the well-being of individual, non-human animals is so common, that it passes without much comment and is not handled as an extraordinary, supererogatory act which could be used in a reductio. Human agents adopt abandoned animals, save them from drowning, starving and injuries, come together spontaneously or found organisations to protect individual animals from harm or to save them from imminent threat, as the same matter of course as they would do for human animals. This is not a romantic distortion of the world but simply an appreciation of the frequency and normality with which intervention in nature for the sake of non-human individuals occurs. The question is not, whether humans should or in fact do intervene in nature to alleviate non-human harm or how absurd such intervention is. The question is rather, if intervention is morally obligatory and if yes, what such an obligation entails.

If intervention in nature for the sake of non-human animals is so common, why then is the predation argument considered to be a reductio of the view that humans are morally obligated to intervene in nature to prevent non-human suffering and death? What is it about the argument that seems to capture an intuitive sense that the conclusion is too absurd to be considered and even more: That it should be absurd enough to help dismiss even the otherwise intuitively more appealing dictum that humans are morally obligated to alleviate avoidable non-human suffering? I shall now turn to three of the most common responses to the natural harm argument, which reject the obligation of intervening in nature for the sake of non-human moral patients. These arguments agree with the principle that avoidable, non-human animal suffering should be alleviated but they disagree with the conclusion that humans are morally obligated to intervene in nature in order to hinder or alleviate harm caused by natural processes.

1.1 The Moral Agency Argument

The first argument to be evaluated in this context is the Moral Agency Argument.Footnote 3 According to the Moral Agency Argument, humans don’t have a moral obligation to intervene in nature because most animals who cause harm in nature are not moral agents.Footnote 4 This argument is based on the intuition that if an event is caused by something other than a moral agent, then there is nobody responsible for the event. As many animals (including many humans) are not moral agents who can be called to responsibility or expected to be accountable for their actions, their actions fall outside the scope of accountability. One version of the Moral Agency Argument is the Moral Obligation Version: Animals cannot be obligated not to be predators. There is nothing morally right or wrong about their being predators. Therefore there are no grounds for human intervention. This response presumes “that the moral value of an action derives entirely from the agent’s responding or not responding to moral rules” (Sapontzis 1984, 28). Another version is the Autonomy Version, according to which humans ought not to intervene because they must honour the competence, autonomy or sovereignty of animals.Footnote 5 This argument is in a particularly precarious position, for while it wants to reject the obligation to intervene by using the principle of respect for autonomy, it wants to retain the idea that other animals are not (autonomous) moral agents who can be held accountable.

One reason for the popularity of versions of the moral agency argument might be found in the hold of contractarianism on moral imagination as well as in an otherwise commendable precautionary principle when it comes to attributing blame and responsibility to possible agents of morally blameworthy acts. In contractarianism the moral community is occupied by moral agents while moral patients are either actually or potentially moral agents. If a given entity is not a moral agent, it can have neither rights nor duties. The implications of this position have been criticised by a number of philosophers, famously by Jeremy Bentham, who has argued that the condition for being a moral patient is sentience and has pleaded for the dismissal of many a quality irrelevant for moral considerability (Bentham 1996, 283).

The attraction of the precautionary principle is echoed in a long tradition of legal philosophy, according to which a suspect is to be considered innocent unless proven guilty and a lack of mental competence excludes them from legal accountability altogether. Bizarre examples of historical cases, in which individual animals or families of animals were taken to court for harming humans or their property show that animals were indeed at times considered as moral and/or legal agents and thus deemed to be suitable candidates for punishment (Evans 1906). When today the Moral Agency Argument rejects the idea of morally culpable non-human animals, then part of its attraction seems to lie in developments not only in the field of animal cognition but also in legal principles meant to protect the innocent from miscarriages of justice.

Despite such attractions however, the moral agency argument fails. One reason for this failure is that intervention in nature does not require the causes of harm to be moral agents. It is not the case that humans ought to intervene to hinder or alleviate harm only in cases when this harm is caused or threatened by moral agents. When human patients are under threat, such as from natural disasters like earthquakes or tornados or from (non-human) predators, intervention is considered to be morally obligatory regardless of whether the source of threat is a moral agent or not.Footnote 6 Additionally, even if such a distinction were acceptable, the moral agency response would be relevant only for the case of predation, not for other causes of harm in nature. Harm in nature does not stem from non-human predators alone. Hunger or thirst, excessive cold or heat as well as natural disasters are all causes of harm, which remain untouched by the moral agency argument, as in these cases there is not even the possibility of considering whether the causes might be moral agents, as it might be the case with some non-human animals.

1.2 The Argument from Hubris

The second argument against human intervention in nature that I shall evaluate is the Argument from Hubris, which is based on the worry that such intervention would be an indicator of human megalomania and an overestimation of human capacities with unpredictable consequences, which might range from bad to disastrous. Such consequences may include the welfare of the very animals that intervention means to protect from harm. Peter Singer’s position on the elimination of carnivorous species in Animal Liberation is an example: “Judging by our past record, any attempt to change ecological systems on a large scale is going to do far more harm than good. For that reason, if for no other, it’s true to say that, except in a few very limited cases, we cannot and should not try to police all of nature”Footnote 7 (Singer 2002, 226). Sue Donaldson and Will Kymlicka similarly approve of a general “hands off” policy with regard to wild animals due to human fallibility, i.e. the danger of producing more harm than good (Donaldson and Kymlicka 2011, 186) while retaining a duty of positive assistance as far as it respects the sovereignty of wild animals.

The argument from hubris is based on a sceptical appraisal of human competence. Elements of such competence are the ability to reach right judgements in moral questions and the ability to implement these, i.e. the ability to perform morally right actions. An individual or collective moral judgement could be admirable in some sense but it might lack phronesis; i.e. it might not be practicable or even desirable in the world as it is. Even to contemplate an extensive intervention in nature demonstrates a lack of humility and a desire to play God.Footnote 8 In short, human views on morality may be faulty and in cases they are not, humans may simply be incapable of implementing morally praiseworthy maxims without causing more harm than good. To ignore these possibilities, i.e. to ignore, neglect or reject the fallibility of human agents is what prompts accusation of hubris.Footnote 9

Furthermore, human beings’ certainty concerning their moral stance as well as their position of agency in morality, can lead them to act in ways that are paternalistic or that impose anthropocentric values on others. Such sceptical views on human intervention can also fall back on the history of near-catastrophic effects of human activities on nature. The loss of a vast number of species, the even greater number of species under threat of extinction, depletion of resources and of the wilderness, the thinning of the ozone layer and last but not least on-going climate change threaten to irrevocably change the planet in a way that much of the diversity and possibly the very conditions of life will be lost.Footnote 10 Concern for the planet and for the living beings on it can arguably be better catered for not by human intervention but on the contrary by limiting the sphere of human influence to as little as possible and preserving greater areas of wilderness.Footnote 11

This position seems to be strong in having an empirical record of near-catastrophic human activities to point to. Nevertheless, it is not obvious that it is right. For one, such an apocalyptic vision of (further) human activity is not unavoidable. Until recently, human intervention in nature did not take place with the goal of securing ecosystems for non-human living beings, for protecting species or individual, non-human animals. The danger that the planet is faced with is not the result of misguided human intervention for the sake of the environment or non-human animals, but the consequence of agricultural, military, commercial and industrial enterprises, which were ignorant of or disregarded their effect on nature. It is not a foregone conclusion that humans are incapable of correctly judging and rightly implementing policies, which protect non-human interests on earth, even if it seems prudent not to expect too much.

Proponents of this position might also counter and criticise the hubris argument for having a romantic nature of nature, which considers nature to be a source of unfailing wisdom. Any interference in the delicate harmony of nature must then be necessarily for the worse and humans should maintain a proper deference towards nature. Such a stance can be criticised for either being ignorant of the vast amount of suffering on earth, or for disregarding it for the sake of what is thought to be the greater good.

1.3 The Implementation Argument

The Implementation Argument is related to the practical branch of the Argument from Hubris but it is not specifically concerned with whether one ought to intervene in nature for the sake of non-human interests. According to the implementation argument, the principle that avoidable non-human suffering should be alleviated is a valid principle. What the implementation argument calls into question is the practicability of the obligation derived from such a principle. Demanding equal moral consideration for non-human animals is pointless, the argument goes, when it’s impossible practically or too costly to be contemplated.

The response to the argument from hubris that well-informed and non-anthropocentric intervention need not be catastrophic applies also to the implementation argument. Intervention need not be dismissed as a megalomaniac fantasy of policing the very last nook and cranny of the world, where an insect is under the threat of being devoured by a spider. Similarly, if human intervention aims at the extinction of a species, this must not necessarily be considered to be a bad thing. It would require an extreme form of conservationism to lament the extinction of, say, smallpox and the extinction of certain bird, reptile or mammalian predators might similarly lead to more gain than harm, even if the cost is an otherwise lamentable decline in biodiversity. The argument from hubris and the implementation argument, in their most palatable formulations, should be considered as restrictions to the scope of intervention and an espousal of the principle of erring on the side of caution, rather than as reasons for a wholesale rejection of the obligation of intervention in nature for the sake of non-human animals.

It is thus possible to avoid the reductio by formulating the moral obligation to alleviate non-human suffering and death in the wild in a way as to incorporate two constraints: The first one concerns the practicability of intervention and establishes a moral obligation to intervene only in cases where this is humanly possible. The second constraint acknowledges that the lack of relevant information or competence in humans can risk producing more harm than good. Intervention to avoid non-human animal suffering would therefore be a realistic option only in cases, in which it can reasonably be estimated not to result in more harm than good. Such a reformulation acknowledges the element of hubris and thereby counteracts an intuitive resistance to the idea of extensive human intervention in nature.

2 Reconsidering the Disvalue of Suffering and Death

We saw that the reductio of the natural harm argument is avoidable in certain reformulations of the principle of alleviating non-human harm and that the challenge of the moral obligation to intervene in nature with this purpose is by no means discredited by arguments attempting to brand it as absurd. I shall now consider a third way of attempting to evade the moral obligation posed by suffering and death in the wild. Such an obligation exists only if we consider these to be sufficiently bad as to create a moral demand for its alleviation. If we reject the premise that death and suffering, as the prime cases of harm, are bad, then the moral obligation to intervene in nature in order to alleviate such harm disappears.

How then can we disvalue suffering and death? The method I shall consider here questions whether there is a relevant difference between the suffering and death of humans as compared to those of non-humans. This move would keep in place the moral obligation to intervene in nature in order to alleviate human suffering and death. We can assume that in the case of humans, there is a wide consensus on the worth of human life and the disvalue of human suffering and death. Both precautionary and rehabilitating interventions are generally considered to be not only morally permissible but morally obligatory in cases, in which humans are threatened or harmed by natural causes. In this attempt to evade the moral obligation put by the natural harm argument, the reason why humans would not be morally obligated to intervene in nature in order to alleviate non-human harm would therefore not be based on considerations of ignorance, hubris or practicability but simply on the reason that harm suffered by non-humans is not as bad and therefore not as morally demanding as harm suffered by humans.

2.1 Is Non-Human Animal Suffering Not That Bad?

The widespread conviction that there is a moral obligation to intervene in nature in order to alleviate human harm thus opens up one way of considering the question whether non-human animal suffering is bad enough to obligate intervention. If humans and other animals are similar in all relevant respects and one is considered to be the legitimate recipient of moral consideration, which obligates intervention in nature, does the principle of equality found a corresponding obligation to intervene in nature for the sake of the other as well? Considering the question of equality here is part of our endeavour to see whether human and non-human suffering and death are too different as to make the principle of equality misplaced in the context of suffering in the wild.

The method I shall follow takes as a starting point the principle of equality as a normative ideal and discusses not whether human and non-human animals are sufficiently alike to deserve the status of moral patiencyFootnote 12 but rather whether the differences between human and non-human suffering and death are sufficient to call for different treatment in the particular question of whether human intervention in nature is a moral obligation. If there is a strong case to be made for a differentiated dealing with human and non-human moral demands in the question of intervention, then it requires a complex and pluralistic approach (comp. Williams 1973; Walzer 1983). To develop such an approach, I am going to make us of some elements, with which Walzer develops an account of justice as to reign in different spheres and adapt them to the question of harm suffered by non-human animals in the wild (Walzer 1983). Justice, in Walzer’s view, is served when the valid good in one sphere cannot be converted into advantages in another sphere. Disregarding this principle, i.e. converting one good into another is unjust (Walzer 1983, 19). How we treat a moral patient needs to follow the right reasons in a given, particular case. Walzer’s “doctrine of right reasons suggests that we pay equal attention to the ‘different qualities,’ and to the ‘individuality’ of every man and woman, that we find ways of sharing our resources that match the variety of their needs, interests, and capacities” (Walzer 1996, 661). If we add non-human animals to the Walzer’s men and women, the question that follows is, what obligations such an understanding of justice puts on humans as moral agents when it comes to the question of non-human harm in nature.

2.2 Different Values of Life

Subscribing to the principle of equal consideration in the treatment of human and non-human animals, the question turns essentially to one of a correct differentiation and evaluation of types of harm with which different types of moral patients are afflicted or threatened. In this context, I shall primarily consider death and whether non-human life is essentially less worth than human life to justify different responses to threats in nature. Such a move leads the question to better established life-boat thought experiments, in which one has to choose between the loss of a human being or another animal in extreme situations of threatened survival (Regan 1983, 324; Francione 2000, 151).

If we consider death to be the end of life and life to be the condition for everything else that can be valuable for its subject, then death is not only a loss of life but of everything else of value attached to it. The death of one individual can therefore differ from the death of another individual in the quantity and quality of what is actually or potentially lost. Some intuitive sense of this difference may be part of the explanation of why the death of a child is generally considered to be tragic in a way that the death of a centenarian is not. Although both individuals lose the same thing, i.e. life and although this means the end of their time on earth for both, there seems to be a clear difference in what each could have expected out of their future lives. If we understand life not simply as the opposite of death, but as the necessary basis for a number of capacities, experiences and expectations along with their corresponding values, then the loss of life of one individual will bring about a loss of certain other things, which will necessarily differ from those of another individual. Prominent among these is consciousness: Consciousness is as yet necessarily coupled to biologically intact life-functions and most things of value, all possibility of harm and gain for the individual in question is coupled to consciousness.

If we work with a pluralist and aggregative system of value, the distinction between life and consciousness already divides entities into more basic forms of life, which we may assume to have no consciousness and more complex ones with consciousness. The usual approach to the question of different values of life is to consider value to increase with the number and nature of capacities. In the approach I am suggesting however, having consciousness is not something that makes the life of an entity valuable. Rather, it constitutes a separate sphere which is valuable in itself. The connection between life and consciousness is necessary only because of the present biological links. Consciousness independent of organic life is not unimaginable and may yet come to be in future.

Other qualities might not be as fundamental as consciousness but still important enough to consider them as establishing a sphere of their own. Even if far from being implemented universally, the idea that e.g. access to education is a sphere of justice in which social origin should have no effect, has become mainstream. According to another similar principle, financial power should be irrelevant for access to health care. In short, “we owe different duties to different qualities” (Pascal, quoted in Walzer 1983, 18). The exclusion of women or slaves from the franchise was one of many cases, in which a quality, e.g. being a woman, was used as means of exclusion from the exercise of rights granted in another sphere, e.g. self-determination. Denying a woman the right to vote on grounds of her sex is a paradigm example of dismissing the autonomy of the spheres in question, while denying a dog the right to vote is not.

A differentiated and complex understanding of equal consideration requires a determination of the spheres and additionally a determination of the entities, which can be considered to be agents and patients in these spheres. While the cognitive and political capabilities of dogs do not seem to be sufficient for them to be considered as agents in the field of representative democracy, there is no doubt that their capabilities for decision-making in other regards make them agents, the preferences of which should be taken into account. As agency and patiency are conceptually independent, a dog can be a patient in a sphere in which she is not an agent. In the example above, although it is no injustice to deny a dog the vote in parliamentary elections, she should be considered as an agent in spheres, in which she has and expresses her preferences and she should be considered to be a patient in other spheres and decisions taken in the sphere of politics should consider the interests/rights of dogs (Donaldson and Kymlicka 2011).

What relevance does this adaptation of a complex principle of equal consideration have for how we treat other living beings in general and other animals in particular? If we agree that animals have inherent valueFootnote 13 then the fact that there are different spheres of value to be taken into account in an individual’s life enables us to determine whether more or less harm is suffered by an entity by death according to the number of spheres it can be considered to occupy. The injustice of a woman being excluded from standing for parliament on the basis of her sex for example, can be multiplied if she is also excluded from the spheres of health care and higher education but inclusion in and exclusion from these spheres follow their own rules. Victories may be gained in one sphere – she may be considered as eligible to profit from health care in the same way male citizens can – without this inclusion affecting her position in other spheres.

In the question concerning positive duties for protecting the interests of non-human animals in the wild, this differentiation could be used to circumvent the reductio of the natural harm argument: Moral agents are not morally obligated to intervene in nature for the sake of non-human animals in the same way that they are obligated for the sake of human animals because, usually, the loss of life, health or the free and unencumbered use of the faculties of a non-human animal does not affect as many spheres as that of a human animal. In other words; since humans are able to occupy more spheres, in which they can be harmed or benefited, their loss of life, health or capabilities excludes them from a greater number of spheres and is therefore a greater evil. In short, human death and suffering leads to a greater loss of opportunities, valuable experience and activity in a greater number of spheres than non-human animal death and suffering. Therefore, while it is morally obligatory to intervene in nature to protect the rights/interests of humans, it is not obligatory to do the same for the sake of non-human animals. The difference between what suffering and death means for a human being is essentially different than what it means for a non-human being and this allows us to avoid the reductio of the natural harm argument.

2.3 Back to Square One

There are – at least – three reasons why this attempt at avoiding the reductio fails. The first reason is simply a recapitulation of the well-known argument that even if there might be differences on average, there seems to be no quality, which differentiates all non-human animals from all human animals. The attempt at defining human beings by what distinguishes them from all other animals has a long history. From the most familiar versions, in which humans are considered to be the only animal that can think, use tools or use language to more modern candidates of distinction like empathy, recreational sex or music, increasing sophistication in research methods reaffirms that differences between humans and other animals are differences in degree rather than in kind. Arguing for intervention in nature for the sake of all humans but not for the sake of other animals therefore falls into the speciesist trap. Differentiating the amount and nature of the spheres an individual is active in from those of another can therefore work only in individual cases and cannot found a principle of non-intervention in nature.

The second reason is concerned with the method of establishing the worth of human lives and non-human lives by using thought experiments of extreme cases. The difference in the approach between such cases and the present question of intervening in nature is the fact that while thought experiments such as life-boat cases require one sacrifice for the sake of the wellbeing of another, intervention in nature does not necessarily require such a choice. Methods of intervention can aim at minimising overall harm without gravely harming innocent threats like predators, endangering the conditions for life on earth or harming human beings. While it may be the case that a differentiation of spheres helps to enable better decisions in lifeboat cases, it does not help us in finding a blanket solution to this problem.

The third reason why disvaluing non-human suffering and death fails as a method for avoiding the reductio of the natural harm argument is concerned with the legitimacy of using different spheres in this context. The pluralistic approach to complex equality may in fact only re-establish the usual hierarchy of the worth of living beings with humans accorded the highest value. By doing so, it threatens the sanctity of life and the dignity of individuals and re-opens the door for disvaluing the value of life and suffering of beings according to how much they deviate from the standard of the healthy, able, young and male human.

The pluralistic approach to different spheres in the context of non-human harm in the wild does indeed seem to make the same mistake it sets out to avoid. It does this, because the interests of non-human animals in the wild that we are concerned with are set in the most fundamental of spheres: Life, health and the freedom to pursue their own life. While it may be true that justice requires a pluralistic approach in which different spheres are treated as autonomous and that an increasing number of capacities in an increasing number of spheres leads to a greater loss coupled to the loss of life or incapacitating suffering, it does not follow that it is (more) permissible to value the life, health and freedom of individuals according to the length and breadth of other spheres they are able to occupy. For while it seems correct to argue that there is a greater loss involved in one case than in the other, this method is actually just another way of making irrelevant qualities bear on the most basic interests. The loss of life for a healthy, active human may mean a loss of activity in a greater number of spheres than for a healthy, active dog. But it can do this only by considering the value of non-human animal life and health as lesser than that of humans based on the latter’s competences in spheres such as politics, art or science. Arguing that the loss of life of a dog is a lesser evil than that of a human due to the amount of spheres they can occupy, thereby makes a full circle and returns to the very position it set off to reject: Considering irrelevant qualities as relevant for moral considerability in the most fundamental interests. Using a differentiated account of spheres occupied by human and non-human animals would therefore enable us to argue that human death and suffering are evil in a way as to make it a positive duty to intervene in nature for their sake but that non-human animal death and suffering are not evil in the same way only at the unacceptable cost of anthropocentricism.Footnote 14

3 Conclusion

The harm faced by non-human animals in the wild continues to pose a challenge to animal ethics. I have argued that the most common responses to the natural harm argument do not provide reasons to reject the obligation to intervene in nature. Rather, the points they raise can be accommodated in the formulation of such an obligation as constraints and can serve to help establish the scope of intervention in a given case. In the case of suffering for example, we as human beings are not yet in the position to draw a fully exhaustive picture of what type of suffering is adaptive, good, acceptable or positively harmful under which conditions. An avowal of our limitations however need not prescribe a blanket policy of non-intervention. We already know some things, we are adding to our knowledge of animal lives and to the sophistication of our ethical tools and we can do more still.

The attempt at rejecting the obligation to intervene in nature by disvaluing non-human death and suffering fails likewise. I took as my starting point the wide-spread consensus about the obligation to assist human beings when facing harm from natural causes and considered the use of different spheres in order to make a case for a justifiably differentiated treatment of human and non-human animals. I followed Walzer in his characterisation of these spheres as having their own autonomy and that it is unjust to convert the currency valid in one sphere to the currency valid in another. I then suggested that some individuals are able to be active in a greater number of spheres than others and that a loss of life, health or capacities may therefore mean the additional loss of a greater or lesser number of things of value according to the number, variety and amount of unhampered activity in these spheres. As non-human animals are more restricted in the number, variety and amount of activity in their spheres than human animals, their loss of life, health or capacities is not as great an evil as that of human animals and therefore does not obligate human moral agents to the positive duty of intervening in nature for their sakes in the same way. This attempt at rejecting the obligation fails, mainly because it is based on speciesist generalisations and makes irrelevant competences bear on the fundamental spheres of life, health and freedom.

The challenge of harm in the wild thus remains in place. Although considering different spheres helps us develop a more differentiated approach to evaluating the capacities, needs and preferences of individuals, what we need to counter the challenge is an even more differentiated approach: “[…] the idea that we could have a single simple rule – whether an interventionist commitment to ‘supplant the natural by the just’ or a non-interventionist commitment to ‘let them be’ – seems implausible” (Donaldson and Kymlicka 2011, 179). We need to find individual answers to individual problems while circumventing a variety of moral pitfalls from paternalism to speciesism, from hubris to indifference. We also need to consider more factors than only capacities and spheres, e.g. by following the suggestion of taking into account concrete relations between agents and patients (Palmer 2010; Donaldson and Kymlicka 2011; Ebert and Machan 2012).Footnote 15

One thing we seem to be able to establish however, is that it is not absurd to try to rise to the challenge even if we seem to be faced with the impossibility of adequately responding to conflicting demands and even if the charge of absurdity seems to be unavoidable: “Were we to carry out our duty to assist the sheep against the wolf’s violations of the sheep’s rights, we could only succeed in violating the wolf’s rights, whereas if we allow the wolf to violate the sheep’s rights, we fail to do what we should to assist the sheep. Whatever we do, therefore, we stand to be morally condemned, which is absurd” (Regan 1983, 285). The absurdity proclaimed here by hypothetical critics assumes that there are no situations, in which moral condemnation can be avoided. There is no need to assume this: We know that there are situations in which a moral tragedy is unavoidable. In fact, the question of intervention in nature for the sake of non-human animals might just be such a case, where we are faced with moral tragedy, whatever we do. Problems with ignorance, hubris and implementation may draw limits to what we can achieve but nevertheless “[w]e have to reach conclusions based on glaring inadequacy of available empirical knowledge and a clear norm system” (Naess 1991, 135). Moral failure in this question is not the inevitable inadequacy with which we must face the problem of harm in the wild but rather the attempt to dismiss the moral obligation altogether.