Document View

Animal ethics has presented convincing arguments for the individual value of animals. Animals are not only valuable instrumentally or indirectly, but in themselves. Less has been written about interest conflicts between humans and other animals, and the use of animals in practice. The motive of this paper is to analyze different approaches to interest conflicts. It concentrates on six models, which are the rights model, the interest model, the mental complexity model, the special relations model, the multi-criteria model, and the contextual model. Of these, the contextual model is the strongest, and carries clear consequences for the practical use of animals. [PUBLICATION ABSTRACT]

INTRODUCTION

The past few decades have seen the growth of animal ethics, and the rise of numerous theories concerning the moral value of animals. Different ethical traditions (rights theory, utilitarianism, virtue ethics, and even contract theory)1 have been applied to animals, and if these traditions are consistently read, they imply that animals too have individual value, as "ends in themselves." Because of differences in their background, these theories obviously rest on different and sometimes contradictory arguments. The basic claims are, however, the same. First, the value of animals should be analyzed without species-centered assumptions, which state that moral value is inherently linked to human species. Rather, we should look for a neutral ground for moral value, and apply the criterion individually to different beings (see Rachels 1990, and the idea of moral individualism). second, most of the theories argue that the neutral criterion for moral value is consciousness in the phenomenal sense. Autonomy, prepositional language, rationality, and moral agency have been discredited as suitable criteria, and "experiencing the world as something" has been made the basis of moral value.2 Since many animals fall into this category (Dawkins 1998; DeGrazia 1996; Bekoff 2002), they too have individual value.

However, much less attention has been given to investigating what exactly are the consequences of such value as it comes to interest conflicts between human beings and other animals. In the absence of a clear value difference it remains uncertain how to prioritize human and animal interests-what exactly can be done to other animals in the name of human interests? This paper tries to shed some light into the matter by analyzing six different approaches to the question. These are the rights-, the interest-, the mental complexity-, the special relations-, the multi-criteria-, and the contextual model.

The Rights Model

Perhaps the most famous, or at least the most definitive, branch of animal ethics claims that individual value leads to moral rights. Therefore, animals too have moral rights, and these rights consist of claims (Regan 1983)3 that we, as moral agents, must respect. Although in the past the idea of animal rights has been at best marginalized and at worst rejected as absurd, today it is increasingly accepted, and in some jurisdictions it has even been awarded constitutional recognition.4 A traditionally human-based concept has finally been applied successfully to animals, in both the moral and the legal sense-overall, animals are becoming part of politics.5

Rights are often considered the primary tool to solve interest conflicts (see Regan 1983). Still, it remains an open question just how effective these tools actually are. First, it seems evident that they do not guide all our actions: sometimes other considerations overtake rights. Even the most famous exponent of animal rights theory, Tom Regan, offers detailed accounts of the circumstances where rights can be violated (Regan 1983, 301-12)6 and in some contexts even utilitarian goals may overweigh rights (see Frank 2002). second, rights do not tell us what to do when the similar rights of two individuals are in conflict. Therefore, the notion of rights may be important, but in itself is not enough to solve interest conflicts.

A further, but related problem emerges from semantic ambiguity. It is important to note the difference between theoretical rights and rhetorical rights. The former are based on theoretical claims, and the latter are a method of pointing attention toward moral problems, but do not necessarily entail a belief in rights theory, and can actually exist in conflict with that theory.7 The theoretical rights suffer from the traditional difficulties associated with rights theory, such as overt objectivism and intuitionism, stemming from the at times obscure bases of rights (Rowlands 1998, 110-15; Pluhar 1995, 233-40). The rhetorical rights, on the other hand, have been criticized on three counts: vagueness, over-use, and lack of necessity. It has been claimed that they often confuse the moral and legal aspects of rights so that the two become indistinguishable; that they have suffered from overuse up to a point of inflation, as many groups argue for rights without actually being able to stipulate the theoretical basis of such rights; and that they are not necessary for ethics (Frey 1980, 4-7; 1983, 46-52; Midgley 1983, 61-64; Singer 1995). Moreover, often these two types are confused with each other when theoretical rights are mistakenly criticized for being mere rhetoric.

In defense of theoretical rights it has to be noted that some rights theorists have tried to correct the problem of overt objectivism by basing rights on something more localized, such as human epistemology (see Pluhar 1995), and that even the standard accounts do not rely on notions of utterly neutral, ontologically independent rights. Similarly, rhetorical rights may be defended, for despite their shortcomings (as seen in the course of history), they do serve as important tools for political and moral change (Almond 1991; Sapontzis 1987, 82-84). The only thing necessary is to pay attention to exactly what type of role rights in different contexts are meant to play. However, even with these corrections, rights theories often remain too abstract. Rhetorical rights do not so much solve interest conflicts as simply point them out. This means that while the language of rights is both useful and important (obviously animal rights rhetoric, and also philosophers such as Regan, have contributed a great deal to "animal liberation"), it has limits when it comes to addressing interest conflicts at large. Rights are necessary from the point of view of individual value, but not enough from the point of view of conflict between individuals.

The Interest Model

The most obvious way to approach interest conflicts is to consider the nature of the interests involved. Different interests are compared to each other in order to determine their importance and priority. This interest model is favored by many utilitarians, such as Peter Singer (Singer 1975; 1993). As a general principle, it is common (see Taylor 1986), and found in different variations, such as the non-loss of goods principle (formulated in Rolston 1988; see also Moriarty and Woods 1997), which states that we should not take away a good unless this is done to achieve an equal or higher good. The way in which the interests are compared is usually through a two- or three-level approach. For instance, Donald VanDeVeer talks of primary interests, which are necessary for well-being; non-primary but significant interests; and secondary interests, which may enhance our well-being but are not vital (VanDeVeer 1979; 1995).8 The consequences of the model are definite. We are not morally entitled to harm the interests of other animals if they are more important than our own interests. For example, I cannot hunt down a bear simply to experience the thrill of hunting; neither can I kill it for mere culinary purposes and the desire to eat bear soup.

Again there are problems, the first one of which is the comparison of interests. The model works best when we compare interests that are clearly different. We can, without much bias, claim that the interest to survive is more important than the interest to taste exotic meat. Yet, many interest conflicts are not as clear as this, and it can be difficult to decide to which category given interests belong. For example, is it more important to cut down a possibly hazardous tree, or to leave it as the regular nesting place for squirrels; what sort of an interest is the need to engage in a certain cultural tradition? The basic problem is the categorization of interests. As Joshua Frank states: "No matter how large or fine the categories are, the problem remains the same as long as ethical treatment is discrete based on category" (Frank 2002, 53). Interest categories do not do justice to the particularities and nuances of practical life. We cannot reduce the complexities of everyday life to fit categories, no matter how broad or specific the categories.

This leads to a further difficulty when we recognize that we should not only categorize interests on an objective level, but that we also should take into consideration the subjective level. Interests ultimately serve wellbeing, and the full definition of "well-being" is impossible without the personal point of view of the subject. It is difficult to assess just how important certain practices are to different individuals. We may state that playing banjo is a secondary interest at best, but for a player it can be not only an important interest, but also a primary one-she simply cannot think of life without it. The difficulty becomes more serious in relation to different individuals and cultures: it is possible to claim that we can't compare the interests of two individuals because the subjective importance cannot be measured; furthermore, we can claim that the measures themselves are different depending on cultural context (on the relation of this to hunting, see Aaltola and Oksanen 2002). This problem also touches animals directly, for it is often stated that we cannot for certain know what their interests are. That is, since we cannot enter the minds of other animals, we cannot make decisions concerning their interests.

The interest model then faces two problems: how to categorize interests on the objective level, and how to do it on the subjective level. The problems are not, however, quite as damaging as they first seem. First, it should be acknowledged that difficulty does not equal impossibility. Objective ethical categorizations are notoriously difficult, and have been criticized extensively in the recent years for not giving enough room for the complexities of practical life (see Williams 1985; McNaughton 1988; Dancy 1993; Nussbaum 1990; and Maclntyre 1998). What the criticism often ignores is that despite these difficulties we do need some theoretical guidelines, unless we want to sink into relativism (Tànnsjo 1995; Callicott 1990; Little 2000), and the fact that ultimately even the critics themselves rely on some type of generalizations (Crisp 2000). It is also paradoxical to criticize categorization, and at the same time criticize "codifying" (systematic use of categories) for not being clear enough-categorization is blamed for not being sufficiently categorical. It is difficult to categorize all interests "neatly," but this does not mean that we should forget about all efforts of categorization: there are actual differences in importance, even when the precise importance is difficult to determine. That is, there is a need and a basis for categories, even though the categories are not clear-cut. We should be careful as to not emphasize the complexities of the practical level so much that we abandon moral theory altogether, and end up reducing "ethics of practicality" to mere "practicality." Practice should not be abandoned for categories, but neither should categories be abandoned for practice.

Second, like elsewhere in ethics, subjectivity does not necessarily lead to subjectivism. We cannot conclude that since subjective experiences of moral matters are important, moral matters can be reduced to them. It is crucial to take into account the subjective meaning of interests, but this does not entail abandoning interest comparisons altogether, or adopting the idea that everyone decides for her- or himself. We can take a viewpoint into account, and still compare it to another one. Here the Nagelian understanding of "objectivity" is fruitful: despite the fact that neutrality as such remains an abstract ideal, we may still be "reflective" and strive to escape the most obvious of our personal or cultural prejudices (Nagel 1989; see also Elliot 1985). The reason for this moderate objectivism is that in order to make just assessments in interest-conflicts, we have to step outside the framework of one particular party-be it an individual or a culture. What is needed is a common criterion for evaluating different interests, and the basic values presented in animal ethics (well-being, capacity to experience, equality, respect for others) offer such a criterion. Finally, when it comes to making judgments concerning the interests of animals, it should be noted that understanding their interests and personal viewpoints is not categorically different from understanding those of other human beings. Rather, it is a matter of degree (Dawkins 1998, 10-14), and not only something that can be achieved, but also something that the individual value of animals requires us to do.

In this light, the interest model seems like a strong alternative. Yet, it faces two further difficulties. The hierarchical system of interests may well work when we are referring to different kinds of interests, but what if the interests are of the same type, for instance, primary? This is one of the most difficult questions in animal ethics. Most philosophers argue for the right to self-protection, and claim that humans are justified to favor their own interests; even Regan grants that they may favor themselves in "lifeboat situations"9 (Regan 1983, 324). Therefore, if it is absolutely necessary for our own primary interest that we step over the interests of others, we may do so. We may eat meat, hunt, and wear fur coats if no other means are available. Still, it is important to note that the justification for favoring human beings is lacking. We can justify favoring ourselves individually by claims of self-protection, but it is less clear why we would be entitled to favor all human beings in general. That is, someone else's survival isn't our own self-protection: Liz can't choose John instead of Scruffy or Matilda the cow on the basis of self-protection alone. Moreover, just as she may protect herself at the cost of Scruffy, she may do so at the cost of John.

Another difficulty is claimed absurdity of equality. Tony Lynch and David Wells (1998) claim that it is absurd to ask why we ought to prioritize human beings. Instead of theory, we should pay attention to the practical, and to the opinion of the majority. However, this argument is weak. It rests on the naturalistic fallacy, and history shows that it would have made many efforts toward equality come to nothing. Also, the approach is very close to dogmatism: we are simply to accept a certain "truth" as given. Another version argues that rather than the idea itself, it is the consequences of equality that are absurd. According to Mary Anne Warren, a farmer could not harvest a field occupied by insects, since the survival of thousands of insects counts for more than the need of the farmer to fulfill his occupational duties (Warren 1997, 71-84). In order to avoid human-centered assumptions, careful judgment is called for concerning what counts as "absurd" (for example, not long ago veganism was commonly deemed to be "absurd"). Furthermore, Warren's example is lacking, as ultimately harvesting is a matter of survival, and hence selfprotection. Still, it is true that in some cases following only the importance of interests is not enough. This applies not only to other animals, but also to interest-conflicts between humans. The interest model is on the right track in emphasizing the role of interests, but like the rights model, it has its limits. Besides rights and interests, something more is required.

The Mental Complexity Model

The third option argues that mental capacities ought to be taken into account when the rights or interests are equal. Donald VanDeVeer claims that when the interests are of the same weight, we have to compare the cognitive complexity of the beings involved and favor the more complex ones. This approach has been favored also by Raymond Frey (1980; 1996), Jon Wetlesen (1999), and Louis Lombardi (1983). Even animal ethicists such as Peter Singer and Tom Regan advocate it in certain instances-Singer by claiming that mental capacities can affect the existence and weight of specific interests,10 and Regan by arguing that in certain extreme situations of life and death, mental complexity does matter.11 This mental complexity model rests on the perfectionist tradition, which argues that certain "perfected" skills are at the core of moral status (see Bernstein 1998). Besides perfectionist, the model is also quantitative, for it claims that moral status comes in degrees according to the state of complexity.

Like the previous ones, this model presents difficulties too. The most obvious one is that it is unclear why mental skills have relevance-why should we accept the perfectionist approach? As has been pointed out, the fact that Liz is mentally complex does not mean that she has a right to cut down an old elm tree to satisfy her curiosity (Anderson 1993; Attfield 1983). The point is simple: even if a is more mentally complex than b, this does not mean that a ought to be always prioritized in interest conflicts against b. The relation between the weight of interests and mental complexity remains unjustified. To appreciate this, we only need to consider the consequences of the model. If we are to concentrate solely on mental skills, we are entitled to use not only other animals, but also other human beings to further our own benefit. Ultimately we could have a hierarchical society, where each individual's moral status would be proportional to how much she scores in a cognitive skills test. The proponents of the mental complexity model are not unaware of this difficulty. VanDeVeer admits: "I am suggesting that in fact we do suppose widely diverse amounts of intrinsic value attach to different lives, even within the category of human beings" (VanDeVeer 1995, 8) and recognizes the possible consequences.

The proponents have concentrated on the intra-human consequence, and offered two solutions. VanDeVeer states that after a certain "threshold" of mental capacity all human beings are equal. This reply does not suffice, for VanDeVeer seems to be unsure of whether complexity matters and moral status comes in degrees, or whether moral status ultimately is categorical and equal. He solves this problem by applying the former option to animals, and the latter to humans. Of course, this is unwarranted. We cannot have two different criteria for animals and humans: either mental complexity matters or it does not; and either moral status comes in degrees, or it does not. It also has to be noted that if the threshold is applied, we are left without means with which to make decisions between two equal beings. The whole point of looking for a further model is in order to solve conflicts of equal interests, but this suggestion only makes differentiations between humans and other animals, while conflicts between humans are left unsolved. Furthermore, the model also fails in differentiation, as not all human beings qualify under this threshold test, unless the test is very low (in which case it would also include most animals). That is, any threshold that includes unfortunate humans would also include many animals.

Raymond Frey has a different approach to the matter: he understands the consequences (Frey 1980; 1996), and is actually seemingly ready to accept them. The power of the "Argument from marginal cases" (according to which we logically cannot use animals that are of the same mental capacity with unfortunate humans) in his mind depends on whether we accept the intuition behind it or not, and he apparently does not.12 The intuition in question is that of human equality, and not all human beings are necessarily equal. Logic demands that we apply the same moral criterion to humans and animals, and because of this, some animals may even be of greater value than some humans. Hence, there is no threshold, and intra-human consequences are accepted (Frey is willing to accept that in theory we may perform medical experiments also on marginal groups, if we are to carry them out on animals). The question is, whether or not we are ready to accept a morally hierarchical, elitist society, where the exploitation of the mentally poor is justified. Without further criticism of the hierarchical system, it is fair to claim that accepting it would be a moral mistake and that we still ought to refrain from consuming "idiotburgers"13 in fast food restaurants. Ultimately, the relation between mental capacity and interests remains unjustified, and it is because of this that we cannot use other animals any more than humans on the basis of mere mental complexity.

There are further problems connected to the quantitative side of the model. The main difficulty is the comparison of the mental complexity of different individuals, especially when we are talking on the level of species. First, it is difficult to draw the line at the required level of mental complexity: what amount of complexity has moral relevance?14 second, how are we to compare complexities of different kind? How could we, for example, make a qualitative decision between the mental life of a dog with all the different senses it is based on, and a human being? Ultimately we are expected to favor the human degree and type of complexity (see Pluhar 1995, 288-94).15 Because of this, the model has been claimed to be speciesist: the type of mental complexity it emphasizes is humancentered.16 James Anderson has tried to answer this difficulty with a "swine-plus view," and claims that there is a possibility for a non-speciesist comparison of capacities. Whereas pigs are merely pigs, humans are this and something more, for humans include both animal nature, and the extra that is humanity. (Anderson 1993). However, the argument is flawed simply because humans are not swine-plus anymore than pigs are humanplus or dogs are snail-plus: we are not on the highest ladder of evolution having absorbed the capacities of other animals, and therefore master of both "swinehood" and humanity, but are our own creatures. Anderson proceeds to claim that the human capacity to understand her own good may form a basis for prioritization. This is a common argument, as the claim often is that it is the human capacity to analyze her interests that sets them apart. However, the relevance of this capacity remains uncertain. Why does it matter that John can analyze his desire to not feel pain, if both John and Scruffy have the same desire? If experiences are at the heart of interests-we have interests, when we experience for instance pain or joy-why would the intellectual capacity for analyses bear relevance?

The model should not be abandoned completely, however. The problem with it is that mental complexity is connected to interest prioritization in a general sense (i.e., the more complex beings should be given priority in general), and even to moral value. Rather, we should simply recognize in the Singerian manner that mental complexity can have an effect on specific interests. Only when a given capacity concretely affects the nature of the interest (i.e., brings about its existence, or alters its strength) should it have relevance to interest conflicts.17 Despite this positive element, the model as a whole is inadequate. Although in some circumstances mental capacities can have an effect, in the majority of cases they simply are not relevant.

The Special Relations Model

The former three models are largely part of traditional ethics in that they emphasize such notions as categoricality, objectivity, and universality (see Sheer 1991). The three remaining ones, on the other hand, look for answers from plurality, the personal, and contextuality. The basic difference is between theory and practice-where as the former models emphasize theory, the latter ones underline practice.

One argument is that rather than search for neutrality, we should give more room in ethics for partiality and the personal (Plumwood 1991; Nussbaum 1990). The special relations model follows this argument, and concentrates on the identity of the interest holder. Instead of emphasizing more or less objective factors such as rights, interests, or mental capacities, it underlines the subjective by bringing forth the concepts of kinship and attachment. As it comes to human-animal interest conflicts, the claim is that we have special relations with other human beings, which lead into favoring humans-being human gives ground for special bond and prioritization. Just as a mother would save her own child from a burning house before someone else's child, so are we entitled to favor other human beings over members of other species.

The model has two versions: one based on kinship, and another emphasizing attachment.18 As it comes to the first option, kinship is defended as a "natural inclination," which gives grounds for moral decisions-we "naturally" favor our own kind. This is the argument of those opposed to the individual value of animals (see Becker 1983; Carruthers 1992, 54-56; Benson 1978), but it is also defended by some of the proponents of such value. One philosopher of the latter position is Mary Midgley, who claims that although only one consideration is interest conflicts, species do matter. She argues that, contrary to the traditional claims of animal ethics, to favor one's own species does not necessarily imply speciesism, but is actually a crucial part of our moral inclinations. This is because of the elements of kinship and nearness. She claims that: "In cases of real, sharp, life-or-death competition, we can indeed owe special, overriding duties based on kinship and other forms of social nearness" (Midgley 1983, 23), and that: "The natural preference for one's own species does exist. It is not, like race-prejudice, a product of culture. It is found in all human cultures, and in cases of real competition it tends to operate very strongly" (104). Furthermore: "In handling species conflicts, the notion of simply rejecting discrimination as speciesist looks like a seductively simple guide (-) I am suggesting that it is not simple, and that we must resist the seduction" (103). The claim then is that (1) kinship is a morally relevant factor, and (2) it is part of our biological nature and thus universal, and (3) this factor gives non-speciesist reasons for prioritizing human beings in certain circumstances.

An obvious problem here is the danger of committing the naturalistic fallacy, which confronts any claim that a biological fact could by itself lead to a moral conclusion. Therefore, simply arguing that since we favor we ought to favor strikes one as rather odd. What is needed is a reason as to why biological inclinations should bear moral importance. It should be pointed out that feelings of kinship are also found in conflicts between human beings: we tend to favor those more like us in regard to color, background, gender, and so forth. As seen, Midgley tries to solve this difficulty by claiming that intra-human favoritism is cultural, where as species favoritism is universal and hence somehow more objective and natural. However, this reason is not sufficient, as it remains unclear whether, for instance, racism is always and entirely a cultural construct, and not something that is partly based on a biological tendency (for instance, in the form of fear of previously unaccounted things and beings). Furthermore, species favoritism does carry cultural elements, even to a degree of being at times entirely a cultural construction (for instance the Western dualism between humans and other animals, and the ideas of animals as mechanistic, instinctual "others" are surely culturally loaded). Both intra- and inter-species favoritisms can be biological and cultural, and a categorical distinction is not possible. If we do maintain the distinction, we are easily at fault of dualisms between "natural" and "cultural," human and animal. Also circularity is a danger, for we are relying on a presumption that species is a morally acceptable boundary. What is relevant is to question why the biological or cultural ramifications ought to have moral importance, not to seek to justify inter-species favoritism on the basis of vague, universal notions.19

Midgley rests partly on the idea that species matter in relation to actual treatment of animals, where as knowledge of color is not relevant in treatment of humans. To put it simple: we cannot feed rabbits meat or sustain a lion on a diet of parsley, but we can feed all humans the same. Midgley argues: "Race in humans is not a significant grouping at all, but species in animals certainly is" (Midgley 1983, 98). Differences in nature have to be taken into account in order to fully respect animals. However, the argument is weak. Midgley is mistaken in claiming that human qualities are not "significant." Quite the opposite: we need to acknowledge cultural, and even biological differences in order to treat individuals with respect (we cannot demand devout Muslim women to wear short skirts). Furthermore, there is a logical flaw here, as two claims are conflated: (1) species is significant to respectful treatment, and (2) species entitles prioritization. Respectful treatment has little to do with general groupbased prioritizations: the fact that respectful treatment of a vegan implies she is not fed lamb chops does not mean that veganism may also be used as a basis for prioritization. The fact that species matter with regards to respectful treatment does not mean that species matter with regards to prioritizations. Finally, there is a difference between consideration and treatment. Ethics concentrates on the former, as for instance equality demands that all interests are considered equally-however, the actual treatment of different beings may not be similar. That is, what is crucial is that the interest of both humans and rabbits for food is considered equally, even if they in actuality are fed different things. Therefore, Midgley is mistaken in thinking that differences in actual treatment carry moral implications in this context (DeGrazia 1996, 63).

There are three further arguments for "kinship" favoritism. First of all, it has been claimed that ethical decisions have to be reasonable to those making them, and that hence humans can favor other humans (Fjellstrôm 2002). However, it is unclear what is meant by "reasonable." If rational egoism is meant by it, we could also favor our own sex and social class in order to bring benefits for ourselves, or refrain from giving preference to any humans (after all, in the light of natural recourses, there are too many of us). If the term points toward something that simply is "in accordance with reason" (Oxford definition), it remains unclear why prioritizing animals would fall outside the term. If practicality is meant by it, again the question is, why would giving importance to animal interests be impractical, and moreover, why would impracticality be a reason to refrain from moral change (freeing slaves may have been impractical). second, we can talk of a "natural right" to favoritism on evolutionary grounds. For instance, Baird Callicott has advocated biological "communitarianism" and argued that every species has a natural right to favor itself in the name of self-protection (Callicott 1989). Again intra-human consequences and the naturalistic fallacy pose problems. Furthermore, favoring members of one's own species may not always be self-protective, as the good of the species in general may suffer from overpopulation, and so forth. Third, Peter Wenz has suggested that it is the amount of interaction between humans that leads into prioritization. According to him, interaction with other animals cannot compare to that between humans, and hence humans are to be favored especially in conflicts of equal interests (Wenz 1988). Again problems emerge. It seems anthropocentric to argue that interaction with other animals is lacking in comparison. Also intra-human consequences raise their ugly heads again, for surely there are huge differences in interaction between different people. Finally, Wenz acknowledges that one-sided interaction is enough: having an impact on other people's lives leads into special obligations (hence obligations towards developing countries). We clearly have a huge impact on the lives of other animals already because we consume tens of billions of them each year-surely also this should lead to special obligations (see also Pluhar 1995).

The ultimate reason why the account fails is the fact that it actually rests on generalizations, and abstract characteristics, such as similarity. To use generalizations and abstract characteristics as grounds for moral prioritization is bound to lead to group-orientated ethics and to the neglect of individuals, for it is the general and the abstraction rather than the individual herself that is important. What matters instead of John is humanity, gender, or middle class, and what matters instead of Scruffy is species and breed. Instead of these factors, special relations ought to concentrate on the personal. As Midgley claims: "We are bond-forming creatures, not abstract intellects" (Midgley 1983, 102). A mother would not be entitled to save a child of her own color instead of a child of another simply on the basis of an abstract notion of similarity, but is morally justified in saving her own child on the basis of personal connections. Here we come to the other possibility of special relations, attachment. Special relations deal with personal attachments, not generalizations, or abstract factors such as being "alike." This means two things. First, the model does not offer justification for prioritizing species, and second, it includes animals, toward whom we also may have attachments.

Attachment is an important factor in morality, but instead of abstract entities, it concerns personal experiences. Because of this, the special relations model does not offer the primary tool for the analysis of interest conflicts, simply because interest conflicts do not always concern beings we have personal attachment to. Furthermore, even when personal attachment is of importance, other factors may come first: "Nearness alone can never have a walk over" (Midgley 1983, 21).20 Therefore, although attachment is relevant in certain situations, it needs to be reconciled with other considerations. To give room for the personal does not mean that all efforts of impartiality ought to be abandoned; rather, the two should be seen as co-existing.

The Multi-Criteria Model

A further option is a multi-criteria model, which gives moral value different degrees of importance. Instead of interests, rights, or personal attachment, it is value that decides interest conflicts, as the model argues for different criteria of value, each leading to a different type of value. According to Mary Anne Warren,21 it is important to note the different bases on which values can be built. Sentience (Singer) or being the subject of a life (Regan) are not the only criteria, and when taken to be such, lead ultimately to difficulties in decision-making, and further more, to absurd results. Besides the categorical understanding of value, she also abandons the mental complexity model, for in it many entities we consider valuable are left outside the moral sphere. What we need is not a single, quantitative criterion, but many qualitative criteria with quantitative features. Quite simply, we need to acknowledge various types of value, which again can be gradual. Warren offers six principles, all of which refer to different kinds of value, based on the "intrinsic properties" (Warren 1997, 182)22 of the beings in question. Loosely, the different bases of value can be listed as follows (in no specific order): (1) life, (2) sentience, (3) moral agency, (4) humanity, (5) ecology, and (6) community. Therefore, value is not only grounded on sentience, but also, for example, on being a living entity, or having importance in the eco-system.

These values can be arranged into a hierarchical system, which guides decision-making. A bacteria that is valuable merely because of being organic, is over-ridden in conflicts when the other party is not only organic, but also sentient. Similarly, a being that is not only organic and sentient, but also a moral agent, has again stronger claims. Again, we should favor an ecologically important tree over a "common" tree, and a human being over a moral agent; we should also favor a sentient animal inside our community over a wild sentient animal. The more of these categories a being has, and / or the more important the category is, the more her/his interests have importance. Value also gets different degrees; for instance, the level of sentience affects the value of a sentient being (Warren 1997,148-77; 225). The model then suggests that we make decisions on the bases of the (1) type of value, (2) amount of types of value, and (3) degree of value.

The strength of Warren's account lies in recognizing the pluralism of values. She points out that animal ethics rarely concentrates on the value of non-sentient entities, and rightly assesses that also they can be considered valuable. However, she presents animal ethics in a simplified light, for animal ethics has not claimed that sentience or consciousness is the sole basis of value, leaving species and eco-systems in the category of the morally meaningless.23 The point is that on the individual level it is consciousness (experiencing the world as something) that is crucial. Similarly as we cannot criticize the notion of human rights for not including the value of ecological systems, we cannot criticize animal ethics for concentrating on the individual criterion. There are many sources of value, but when it comes to animals, the individual aspects-such as consciousness-are given priority.

This is not the only difficulty. The more important one is found in the idea of a hierarchical system, and especially the way moral agency and humanity are prioritized. Warren's multi-criteria model states what the other models deem as speciesist, and claims that human beings do have higher value than other animals, and that in an interest conflict, humanity and moral agency come first. In this way Warren is inclined toward a traditional understanding of human value, which places a hierarchical and categorical difference between the value of humans and the value of other animals-an understanding that can be heavily criticized. Warren follows a traditional line of thinking very closely. She first states that the view according to which "moral agency logically entails the possession of full moral status" is "useful," and claims moral agents to have "full and equal basic moral rights" (Warren 1997, 156-7). She then acknowledges that some human beings might be left outside this criterion, and goes on to claim that: "Human beings who are capable of sentience but not of moral agency have the same moral rights as do moral agents" (164). This type of an argument is plagued with difficulties and unwarranted assumptions (for instance, why do moral agency or humanity matter?).24 Warren does not take into account the problems revealed by decades of animal ethics. She offers only traditional types of premises for the importance of moral agency (seemingly emphasizing the origins of morality and reciprocity) without properly justifying their relevance. Furthermore, like most arguments emphasizing moral agency, she ends up in difficulties in tackling the issue of marginal cases.

She refers to "the social, psychological, and biological realities of human existence that require that basic rights not be restricted to human beings who are capable of moral agency" (Warren 1997,164). Regarding infants, she states that: "Fortunately, instinct, reason and culture jointly ensure" that they have similar value as moral agents, and regarding the mentally disabled she talks of "empathy" as the source of value (165-6). No thorough accounts are given of what exactly do these factors consist of, why they bear moral relevance, and why would they not include animals. What exactly are the "social and psychological realities" and why do they matter morally? And why do they not include also animals? Surely, especially today, social factors and our capacity to identify with animals lead us to seeing individual value in them too. Furthermore, what is exactly meant by "instinct, reason and culture," and what is the moral relevance of "instinct" and "culture" (are we to assume that if a sexist would have both the instinct and the culture to abuse women, he would be morally entitled to do so)? If "reason" is to offer guidance here, exactly why would it not also apply to animals? And yet a step further: why does not "empathy" include animals? (Midgley, for one, has offered convincing arguments for accepting empathy as a source of value in relation to animals.) It seems that ultimately Warren is trying to justify the intuition that all and only human beings have full moral value. She treats this like a given assumption, at one point simply justifying it as "common-sensical" and as something that hence must be accepted (88). At the same time, she does not sufficiently acknowledge criticism pointed toward arguments emphasizing moral agency and humanity.

Other problems stem from complexity and vagueness. The model does not state clearly how pluralism of values can be used in practical situations. Plurality is a positive element, but here it seems to lead to not only plurality of choice, but even actual confusion. First, we are not told exactly what matters in the weighing process: is it the number of different values that matters, the degree of them, or the type? Although some guidance is given (for instance, as it comes to prioritizing humanity and moral agency), it is not clear how we should decide: is an extremely rare tree more important than an animal that is both sentient, and part of the community? second, it remains unclear whether the weight of interests is taken into account. Warren implies that it is an important aspect, and that the other factors mainly influence the conflicts between equal interests. At the same time, however, she condemns interest-based thinking as something that leads to absurd consequences, and emphasizes relativity (for instance, cultural considerations can over-step the importance of interests). Third, how is all this to be done in the midst of practical situations? If we are to consider the type of value, the number of types, the degree of value, and finally the weight of interests, decision-making becomes a complicated, head-scratching activity.25 It seems that there are two conflicting problems here: on the one hand, the theoretical basis remains obscure; on the other hand, the model is very categorical. Although it is supposed to resist simple categorization and better match practicality, it actually offers a more detailed categorization that is quite difficult to handle on the practical level. When these two are combined, it is as if we have a series of very specific language rules to go together with only a vague system of grammar. Therefore, despite the positive elements of introducing pluralism, and converging different types of value together, the model faces difficulties. Warren is right in underlining many sources of value, and the model offers important aspects to, for instance, ecological conflicts. However, it suffers from anthropocentrism, and complexity.

The Contextual Model

What has been established so far is that rights and especially interests are an important part of conflict resolution; that mental capacities can, if relevant, affect the weight of a specific interest; and that attachment is to be taken into account in personal issues. What has also been established, is that something more is needed to solve conflicts of equal interests, and that this something ought to be less abstract, and more focused on the practical level. A way to combine these conclusions is to search for a model that (1) takes many factors into account, and (2) allows room for the practical.

One obvious alternative is offered by pluralism, which is characterized by the claim that we have to take the practical level into account when making moral decisions, rather than just apply an abstract theory to it. Because practical situations are multiple, so are our values, principles, and even theories.26 Pluralism (or what Peter Wenz, 1993, calls "extreme pluralism") criticizes theory monism-instead of one theory being the basis for decision-making, we have numerous theories to choose from. Pluralism differs from the multi-criteria model as it comes to both the number of options, and theory; whereas the latter emphasizes a limited pluralism of principles and values, and the existence of a theory (giving basis, for instance, to a hierarchy of values), the former argues for unlimited pluralism, and a lack of theory (see also Warren 1997, 20).

The obvious difficulty with the pluralistic approach is relativity, for without some type of theory moral decisions become arbitrary (see Callicott 1990; Tannsjo 1993; Little 2000). This criticism has been contested especially in recent years, and different alternatives have been offered to replace theory in order to fight off relativism. The so-called traditionalists have argued tradition to be the suitable replacement for theory (see MacIntyre 1988), and more specifically in environmental ethics, ethical attitudes such as the Heideggerian "openness to the world" (Schalow 2000), intuition (Domsky 2001), or simply taste (Brennan 1992) have been offered. However, the strength of these alternatives remains dubious, for the threat of relativism is still very much present simply because we have many different traditions, attitudes, intuitions, and tastes.27 However, this does not mean that pluralism ought to be abandoned altogether, for although monism of theory seems to be necessary, pluralism of principles may still be the option that is needed to take practicality into account. A way to combine a singular theory with pluralism of principles is offered by what Wenz calls "moderate pluralism."

Here "contextualism" is advocated, for it emphasizes not only (1) theory and (2) pluralism of principles, but also (3) the particularities of specific situations. Contextualism differs from pluralism in underlining the meaning of theory, and from the multi-criteria model in giving room for the possibility of unlimited amount of principles; it also differs from both in giving special importance to the context.28 The importance of context has been endorsed by the particularists (such as Bernard Williams, John McDowell, Martha Nussbaum, and Jonathan Dancy), who claim that we have to look at moral problems through the particularities of each given situation, rather than through abstraction, generalizations, and categorizations/codifications; instead of principles determining practicality, practicality determines the use of principle. The importance of contextual considerations is in the fact that detailed knowledge of situations affects their moral nature; what we know affects how we value, and only after knowing the relevant aspects of specific situations, can we see the moral picture emerge. Broadly stated, contextualism argues that there are many principles that rest on a theory, and from which we choose on the basis of the context. The role of contextualism in animal ethics and interest conflicts is two-fold: it enables giving preference to certain principles, and hence does not ascend to complete pluralism, and it also enables making decisions in conflicts of equal interests. The claim here is that there are principles based on theory of value that ought to be prioritized, whereas in dead-lock situations the context should play a role.

First, we need a theory of value. Practicality should be an elemental part of such a theory, both as it comes to construction/origin, and as it comes to content. This means that the theory should be based on human nature (and especially epistemology), rather than doctrines perceived to be entirely objective; furthermore, it ought to be recognizable and functional in practicality, rather than be abstract and rest on distanced application. Such a theory is implied to different degrees in many theories of animal ethics, and is at the root of "experientialism."29 Broadly stated, it claims that experience is the source of any valuing process-we value because we experience. Values ought to be approached from the point of view of experience: what is fundamentally important to experiencing beings? One simple answer is welfare, which in a very elemental sense is a primary value to us: we value as a result of experiencing, and value that which serves our welfare. Furthermore, individual value is based upon the capacity to experience (understood here as the Nagelian phenomenal consciousness): all those that can experience ought to be given individual consideration. Therefore, the claim is that it is experiences, and more specifically welfare, that is the basis for deciding between different principles, and that the only criterion take into account in interest conflicts is the capacity to experience. To put it bluntly, in interest conflicts we ought to seek to best serve the individual welfare of those involved, humans, cows, and giraffes included.

Principles, on the other hand, can assume different meanings depending on the level on which they are used. There are principles of theory, and principles of practicality, which both are needed to resolve conflicts: principles without theory have no basis, and principles without practicality remain abstract. On the level of theory, principles are general, and point out what duties can be derived from the basic value, welfare. The principles of theory include respect for the welfare of others, and equality between experiencing beings. On the practical level, principles follow a more particular form, and articulate how the principles of theory should be understood in practical context (by stipulating exactly what respect for welfare includes). The principles of practicality concentrate on the specifics of welfare, and can give room for various approaches, ranging from utilitarianism to rights theory and virtue ethics. They usually fall into three categories: (1) those concerning the nature of welfare (the nature of the interests involved, the nature of the beings in question), (2) those concerning the responsibility over welfare (personal responsibilities, personal attachments), and (3) those concerning the fair distribution of welfare (number of those involved, the necessity of choice, the probability of gaining the desired outcome).30 The list might sound clinical, but the idea simply is that interest conflicts are affected by the nature of welfare involved, the personal responsibility of those involved, and matters such as the numbers involved. The list includes the best aspects of the previous models, as rights, weight of interests, and personal attachment, are taken into account. Especially important are the weight of interests and the role of rights, as giving preference to the more elemental interests and respecting rights are prima facie duties that can only be overlooked in special circumstances.

Finally, as it comes to the context, matters that affect welfare are of importance. It is not just any knowledge that ought to be taken into account, but things that have a direct impact on the welfare of those involved. Here identification and empathy are important tools. Martha Nussbaum (1990, 1995) has underlined the relevance of seeing things from the point of view of others, and argued that morality needs to take the perspective of those that it involves. In order to grasp welfare and experience, we need to identify with other beings. Therefore, the context ought to be approached through the point of view of the beings whose interests are in conflict, and by taking into account matters that directly involve their welfare.31

The model then makes a differentiation into: (1) theory, (2) principles of theory and principles of practice, and (3) context. We try to pursue a basic value by recognizing that there are many possibilities for doing so depending on our knowledge of each situation. The categorization here may seem overtly abstract, but the idea is quite simple, and something that we often tend to follow in everyday life: moral problems have to be approached from many angles, taking each situation into account.

To give a concrete example, think of the permissibility of killing an animal. Respect for welfare and the principle of equality require that the welfare of the animal is to be taken into account despite her species. The rights principle dictates that the animal has a right for life, and the interest principle claims that we cannot take a life of another unless this is necessary for our own primary interests. This means that we cannot kill a cow unless to ensure our own survival, and that keeping production animals in order to merely gain milk, meat, or leather lacks justification if there are other means of fulfilling our own needs. However, when it is necessary for our own primary interests to kill an animal, justification can be found-for instance, if there are absolutely no alternative manners of gaining nutrition, we can shoot an elk in order to survive. Also, when the primary welfare of the animal herself requires euthanasia, killing her is not morally corrupt. The context can affect the situation further. We can favor those close to us instead of strange animals, and hence shoot a bear in order to protect a child or a dog if the harm is immediate, and when facing a compulsory decision between different animals, we ought to favor the one with more mental ability, with the ability to live longer, with more ties of her own, and so forth. We also have special responsibility for those animals that we have affected, and have to go to lengths in order to ensure the welfare of domesticated animals, or animals whose ecosystems we have damaged. Furthermore, we ought to take into account the numbers involved. Can we kill millions of animals in order to feed pets, or carry out animal experimentation that rarely leads to concrete benefits,32 or hampers the rights of animals, and usually deals with illnesses that we have induced ourselves through our lifestyles?

There are three sources of difficulty here. The first is that the model is committing the act of codification with all its different categories. The answer to this criticism is that the model acts as a compromise between theory and practicality instead of reducing morality into the latter, and by doing so necessarily leaves room for a certain amount of categories. As argued earlier in relation to the interest model, even though overt categorization is to be avoided, categories are still needed in order to refrain from reducing morals into mere chaotic practice. The second criticism is the theory-dependency: does the model not emphasize too much theory at the expense of the practical level? The answer to this is that a basic value and a handful of prima facie principles that revolve around individual welfare are necessary in ethics dealing with individuals: respect of individual welfare should not be just one consideration among many, but the basic approach to moral problems. The third criticism is the vast amount of principles of practicality. As seen, the multi-criteria model was criticized for being both too complex and too vague, and the same criticism could be levied here. Fortunately there is a difference. Whereas the multi-criteria model fails to offer a clear theoretical basis for choosing between a number of options, such a basis is offered here.

Of course, there is still much room for complexity. However, it does not necessarily pose a great problem. Certain amount of complexity is both necessary and desirable, for surely morality is not only a collection of legislative rules, but also something that requires effort and gives room for reflection.

CONCLUSION

Interest conflicts cannot be solved by a model that ignores practical considerations. What is needed, is a model that takes both theory and practice into account. Contextualism offers such a model. It emphasizes theory, and the pluralism of principles in specific contexts. From the point of view of animal ethics, the main claim is that all experiencing beings must be considered equally, and that their welfare is the basis for decision-making: interest conflicts ought to be solved so as to best serve the welfare of those involved in each context. This has radical implications for the use of animals, as many practices ranging from dairy farming to animal experimentation are difficult to justify.