Full Title Name:  Animal Rights Theory and Utilitarianism: Relative Normative Guidance

I. Introduction

In Practical Ethics, Peter Singer argues that ethics is not "an ideal system which is all very noble in theory but no good in practice." [FN1] Singer identifies deontological approaches, such as rights approaches, to ethics as impractical and as having to "rescue" themselves from their inapplicability to real-world moral issues through the introduction of "complexities" such as formulating very detailed rules or establishing ranking structures for rules. He argues that utilitarianism does not start with rules but with goals and thus has greater normative specificity because actions are prescribed or proscribed based on "the extent to which they further these goals." [FN2] Utilitarianism, Singer argues, is "untouched by the complexities" required to make deontological moral theories--including rights theory--applicable in concrete moral situations. [FN3] According to Singer, "[t]he classical utilitarian regards an action as right if it produces as much or more of an increase in the happiness of all affected by it than any alternative action, and wrong if it does not." [FN4]

Singer's views about the nature of rights theory have had a profound impact on the animal rights movement. In the past five or so years, an increasing number of animal advocates have eschewed rights theory for precisely the reason that rights theory is supposedly incapable of providing determinate normative guidance. These animal advocates express concern that rights theory demands the immediate abolition of animal exploitation, and that immediate abolition is simply unrealistic. Instead, these advocates support the pursuit of incremental welfarist reform as a "realistic" means of reducing suffering and eventually achieving abolition. For example, Ingrid Newkirk of People for the Ethical Treatment of Animals (PETA) ostensibly endorses a rights position and ultimately seeks the abolition of animal exploitation, but she argues that "total victory, like checkmate, cannot be achieved in one move," and that we must endorse the moral orthodoxy of animal welfare as involving necessary "steps in the direction" of animal rights. [FN5] Newkirk argues that animal welfare facilitates a "springboard into animal rights." [FN6] Similarly, Animal Rights International's Henry Spira maintains that animal rights theory requires an "all or nothing" approach, and that "[i]f you push for all or nothing, what you get is nothing." [FN7] I refer to this position as "new welfarism," and its proponents, as the "new welfarists." [FN8] Animal welfare theory is very much like utilitarianism in that both permit all animal interests to be traded away as long as the requisite aggregation of consequences so indicates. [FN9]

At least some scholars come to much the same conclusion about the supposedly unrealistic nature of animal rights theory--and the supposedly realistic nature of animal welfare reforms. For example, Bernard Rollin believes that incremental change, in the form of welfarist reform, is the only realistic approach. Rollin claims that in the United States, "we have never had a social and moral revolution that was not incremental." [FN10] In the context of discussing animal experimentation, he argues that, although he endorses the rights view, that view is "utopian and socially and psychologically impossible in our culture." [FN11] As a result, Rollin endorses incremental change based on welfarist reform that would supposedly ensure that the human benefit of exploiting animals "clearly outweighs the pain and suffering experienced by the experimental animals." [FN12] Political scientist Robert Garner claims to be "more convinced by the protection afforded to both humans and animals by rights" [FN13] than alternative views, but he endorses the welfarist view that "any significant human interest outweighs any [sum of] significant non-human interests" because his book "is primarily a book about practical politics." [FN14] Garner argues throughout his book that incremental welfarist reform is the only "practical" way to achieve greater protection for animals.
In this Article, I will explore Singer's view of normative guidance relevant to the human/animal relationship provided by deontological theory, both as an absolute matter and relative to Singer's utilitarian theory. Parts II and III examine Singer's utilitarian theory and the theory of rights presented by Tom Regan in The Case for Animal Rights. Part IV discusses the notion of nonhuman personhood, a notion central to animal rights theory. Part V proposes a theory concerning three components of moral theory. Part VI concludes that whatever indeterminacy may exist with respect to the application of rights theory as a general matter, rights theory provides clear normative guidance concerning the human/animal relationship, and that this guidance is far more determinate than that provided by Singer's utilitarian theory.

II. Singer's Utilitarian Theory

Singer is an act utilitarian who believes that it is the consequences of the contemplated act that matter, and not the consequences of following a more generalized rule. There are, of course, differing views of which consequences are relevant. For classical utilitarians, such as Jeremy Bentham and John Stuart Mill, pleasure alone was intrinsically valuable and pain alone was intrinsically not valuable. Singer, however, claims to subscribe to a modified form of utilitarianism, known as "preference" or "interest" utilitarianism, which provides that what is intrinsically valuable is what "furthers the interests of those affected." [FN15] Those interests include the desires and preferences of those who are affected. Pleasure and pain matter because they are part of what humans and nonhumans desire or prefer or seek to avoid. In Animal Liberation, Singer argues that in assessing the consequences of our actions, it is necessary to take the interests of animals seriously and to weigh any adverse affect on those interests from human actions as part of the consequences of those actions. Humans have failed to do this, Singer argues, because of a species bias, or speciesism, that results in a systematic devaluation of animal interests. Singer claims that speciesism is no more morally defensible than racism, sexism, or other forms of discrimination that arbitrarily exclude humans from the scope of moral concern. [FN16] When people seek to justify the horrific ways animals are treated, they invariably point to supposed animal "defects," such as the inability of animals to use human language or to reason as intricately as do humans. But there are severely retarded humans who cannot speak or reason (or, at least, can do so no better than many nonhumans), and most of us would be appalled if those humans were used in experiments, or for food or clothing. Singer maintains that the only way to justify our present level of animal exploitation is to maintain that species differences alone justify that exploitation. But that is no different, Singer argues, from saying that differences in race or sex alone justify the differential treatment of otherwise similarly situated persons.

Singer's approach is clearly more favorable toward animals than classical animal welfare, which accorded little weight to animal interests. It is important, however, to understand that Singer's theory is not a theory of animal rights. For Singer, the rightness or wrongness of conduct is determined by consequences, and not by any appeal to right. If violating a rightholder's right in a particular case will produce more desirable consequences than respecting that right, then Singer is committed to violating the right. For example, Singer opposes most animal experimentation, only because he thinks that most animal experiments produce benefits that are insufficient to justify the animal suffering that results. But he does not--and cannot--oppose all animal experimentation because if a particular animal use would, for example, lead directly to a cure for a disease that affected many humans, Singer would be committed to approving that animal use. Indeed, Singer has acknowledged that under some circumstances, it would be permissible to use nonconsenting humans in experiments if the benefits for all affected outweighed the detriment to the humans used in the experiment. [FN17]

Singer's theory does not concern rights since Singer does not believe that animals or humans have rights. Indeed, Singer himself refers to his theory as one of "animal liberation" and states that claims of right are "irrelevant." "The language of rights is a convenient political shorthand. It is even more valuable in the era of thirty-second TV news clips." [FN18] It is easy to understand why Singer rejects rights in light of his view that only the consequences (understood in terms of the preference satisfaction of those affected) of acts matter. A right is generally regarded as "a moral trump card that cannot be disputed." [FN19] A right serves as a type of protection that cannot be sacrificed even if the consequences of doing so would be very desirable. Rights, or at least most rights, are not thought to be absolute, but at least some rights provide strong prima facie protection and cannot be compromised without the most compelling reasons. For example, overall social happiness might be increased if I were used without my consent in an experiment, the goal and likely outcome of which would result in a cure for cancer. Nevertheless, I have a moral and legal right not to have my interests in my life or liberty traded away in order to secure that admittedly desirable result.
In determining the consequences of actions, Singer argues that we must accord equal consideration to equal interests. Singer's notion of equal consideration does not mean that animals receive equal treatment, and it does not preclude the morality of a decision to exploit a human or nonhuman. As long as an animal's interests receive equitable consideration (consideration untainted by the speciesism that discounts animal interests simply because they are the interests of a supposed "inferior"), Singer's equality principle is satisfied. But this notion of equality is consistent with animal exploitation if the consequences justify that exploitation and if the decision to exploit is not based on species discrimination. Indeed, Singer acknowledges that he "would never deny that we are justified in using animals for human goals, because as a consequentialist, [he] must also hold that in appropriate circumstances we are justified in using humans to achieve human goals (or the goal of assisting animals)." [FN20] Singer claims not to be "the kind of moral absolutist who holds that the end can never justify the means," and he has denied arguing that "no animal experimentation is ever of use to humans" or that "all animal experimentation involves suffering." [FN21] Garner has noted that Singer does "talk as if the killing of animals for food and their use for experimental purposes should be morally condemned per se because the infliction of pain means that they lead miserable lives" and that "[s]uch a view could be taken to mean that he thinks they have a right to have pain inflicted on them[,] [but] Singer is clear . . .that he is not an advocate of rights." [FN22]

Perhaps the clearest indication of the difference between Singer's view and the rights position is expressed by Singer himself in the second edition of Animal Liberation. Singer argues that many nonhumans, and this class apparently includes food animals, are incapable of "having desires for the future" or a "continuous mental existence." [FN23] These cognitive characteristics assume "an understanding of what it is to exist over a period of time" and Singer doubts that most animals used for food have such an understanding. [FN24] This supposed lack of future desire and continuous mental existence are generally irrelevant when the issue involves pain or suffering alone (although I will argue later that Singer appears to contradict himself on this point and allows for individual capacities to affect assessments of pain and suffering). Singer believes that these characteristics become relevant, however, when the issue involves killing an animal in a painless or relatively painless manner. Singer expresses "doubts" on the issue, but he concludes that "it is not easy to explain why the loss to the animal killed is not, from an impartial point of view, made good by the creation of a new animal who will lead an equally pleasant life." [FN25] Singer maintains that it may be morally justified to continue "to eat free- range animals (of a species incapable of having desires for the future), who have [had] a pleasant existence in a social group suited to their behavioral needs, and are then killed quickly and without pain." [FN26] Singer states that he "can respect conscientious people who take care to eat only meat that comes from such animals." [FN27]

I do not plan to discuss the various criticisms made of Singer's theory; however, there is one aspect of his utilitarianism that requires comment. There is no doubt that: 1) Singer regards most animal experimentation as without merit; 2) he would eliminate factory farming; and 3) we ought, for the most part, to be vegetarians because although it may be morally permissible to eat animals, as a practical matter, the circumstances surrounding their rearing and killing will morally preclude eating them. These views, however, are based on Singer's empirical assessments of the consequences of particular acts in light of his theory that individual acts ought to further the interests or preferences of those affected. Like all such empirical assessments, the consequences of the acts may be evaluated differently by different people. For example, Singer thinks that the negative consequences for the animals involved in factory farming outweigh the benefits, but as Regan points out, "[t]he animal industry is big business," and although "[i]t is uncertain exactly how many people are involved in it, directly or indirectly, . . . the number must easily run into the many tens of thousands." [FN28] Those involved in animal agriculture "have a stake in the animal industry as rudimentary and important as having a job, feeding a family, or laying aside money for their children's education or their own retirement." [FN29]

Similarly, philosopher R.G. Frey, who is critical of Singer's utilitarianism and of rights theory, presents a lengthy list of "practical considerations  that must be taken into account" in evaluating Singer's claim that animal agriculture, and especially the practices involved in intensive agriculture, are not justified under Singer's theory of preference utilitarianism. Frey's list includes negative consequences that would befall those directly involved in the raising and killing of animals, such as farmers and slaughtering operations; those involved indirectly in various enterprises such as: food retailers, fast food restaurants, and the dairy industry; the pet food industry; the pharmaceutical industry; the leather goods and wool industries; agricultural and veterinary research incidental to agriculture, the publication of books about animal agriculture, the advertisement of products of animal agriculture, and so forth. [FN30] Frey is correct that the collapse of factory farming would have a profound impact on the international economy. This is not to say that these negative consequences would not necessarily outweigh the animal interests involved in not experiencing pain and suffering incidental to intensive agriculture; it only says that if the issue hinges on the aggregation of consequences, it is unclear whether it would be morally right under Singer's view to abolish factory farming. What is clear is that given Singer's view that the rightness or wrongness of action is determined by the consequences it has for the interests of all affected, he simply "cannot say that the interests of those humans involved in. . . . [factory farming], those whose quality of life presently is bound up in it, are irrelevant." [FN31] The problem is that once the preference satisfaction of everyone involved in factory farming (humans and nonhumans) is deemed relevant and counted equitably, Singer's assumed result appears to be much more controversial than he recognizes.

III. Rights Theory

In The Case for Animal Rights, Tom Regan argues that the rights position regards as morally unacceptable any institutionalized exploitation of nonhumans. Regan unambiguously and without equivocation condemns the use of animals for food, hunting, trapping, testing, education, and research. Regan believes that humans and nonhumans are subjects-of-a-life that have equal inherent value. That is, agents and patients are conscious, possess a complex awareness, and have a psychophysical identity over time. Agents and patients may be harmed or benefited and have a welfare in that their experiential life fares well or ill for them, independently of utility that they have for others or the interest that others have in them. Inherent value theory holds that the individual has a distinct moral value that is separate from any intrinsic values and that the attribution of equal inherent value to both moral agents and relevantly similar moral patients is required because both agents and patients are subjects-of-a-life.

Regan argues further that the respect principle requires that we treat those individuals who have inherent value in ways that respect their inherent value. The respect principle states simply that no individual with equal inherent value may be treated solely as a means to an end in order to maximize the aggregate of desirable consequences. Regan's respect principle shares important theoretical similarities and differences with the notion articulated by Immanuel Kant that we treat other persons as ends in themselves and never merely as means to ends. Rational agents, Kant argued, have value in themselves independent of their value to others. Regan's contribution to this notion is his use of the subject-of-a-life criterion to identify in a nonarbitrary and intelligible way a similarity that holds between moral agents and patients and that gives rise to a direct duty to the latter. [FN32] Being a subject-of-a-life is a sufficient condition for having inherent value, but is also a criterion that allows for the intelligible and nonarbitrary attribution of equal inherent value to agents and patients, including nonhuman animals. Regan stresses that there is no nonarbitrary way to separate moral agents from moral patients, and that there is no way to differentiate human moral patients from nonhuman moral patients without relying on some form of species bias or speciesism. Regan argues that institutionalized animal exploitation (such as the use of animals for food, experiments, clothing, and entertainment) violates the respect principle by treating all animal interests as tradable as long as the aggregation of consequences are justified.

Although Regan's theory represents an important contribution that differs qualitatively from Singer's theory of animal liberation, there is a sense in which any coherent and non-speciesist theory of animal rights must rule out all forms of institutional exploitation. As Henry Shue has argued in the context of human rights, there is a logical distinction between what Shue calls "basic" rights and "non-basic" rights. According to Shue, a basic right is not a right that is "more valuable or intrinsically more satisfying to enjoy than some other rights." [FN33] Rather, a right is a basic right when "any attempt to enjoy any other right by sacrificing the basic right would be quite literally self-defeating, cutting the ground from beneath itself." [FN34] Shue states that "non-basic rights may be sacrificed, if necessary, in order to secure the basic right. But the protection of a basic right may not be sacrificed in order to secure the enjoyment of a non-basic right." [FN35] The reason is that a basic right "cannot be sacrificed successfully. If the right sacrificed is indeed basic, then no right for which it might be sacrificed can actually be enjoyed in the absence of the basic right. The sacrifice would prove self- defeating." [FN36] Shue emphasizes that basic rights are a prerequisite to the enjoyment and exercise of non-basic rights, and that the possession of non- basic rights in the absence of basic rights is nothing more than the possession of rights "in some merely legalistic or otherwise abstract sense compatible with being unable to make any use of the substance of the right." [FN37]

Although Shue identifies several basic rights, the most important of these is the "basic right to physical security--a right that is basic not to be subjected to murder, torture, mayhem, rape, or assault." [FN38] While acknowledging that it is not unusual in a given society that some members of at least one ethnic group receive less physical protection than others, Shue argues that "few, if any, people would be prepared to defend in principle the contention that anyone lacks a basic right to physical security." [FN39] If a person does not enjoy the basic right to security and may be murdered at will by any other person, then it is difficult to understand what other rights that person might enjoy. Most of the time, discussions about rights occur in the context of discussion of human rights, and these discussions do not concern whether we should be able to kill and eat people, or whether we should be able to use people in experiments to which they have not given their informed consent, or whether we should be able to use people in rodeos, or exhibit people in zoos. It is assumed--at least under the law of most countries and at least in the moral views of most people--that people have certain rights, or, at least, that they have certain interests that cannot be compromised irrespective of consequence.
Shue is most certainly correct to note that we always assume that humans have basic rights to physical security, whether or not there are social differences in terms of the actual distribution of these rights. In other words, recognition of the basic right to physical security is a right as a matter of law irrespective of whether the state enforces this right in an even-handed manner. In the case of animals, however, the situation is precisely the opposite. We talk informally about the rights of animals, but animals do not have the basic legal right of physical security and they cannot possess it as a matter of law. Because animals are regarded as the property of their human owners, they can be killed for food, used in experiments, and exploited in numerous other ways simply because the owner of the animal regards it as a "benefit" to do so. If animals are to have any rights at all (other than merely legalistic or abstract ones to which Shue refers), they must have certain basic rights that would then necessarily protect them from being used for food, clothing, or experiments. [FN40]

IV. Animal Rights and Personhood

Our treatment of nonhuman animals reflects a distinction that we make between humans, whom we regard as persons, and nonhumans, whom we regard as things. Although we may regard some animals as having certain "interests," we regard all of those interests to be tradable and dependent on our judgment that the sacrifice of the interest(s) will benefit us. This trade is generally permissible even when the animal interest involved is significant and the human interest is admittedly trivial, as is the case of the use of animals for "entertainment" purposes such as pigeon shoots, rodeos, or circuses. Animals are not persons in either moral theory or under the law; they are property in that they exist solely as means to human ends. [FN41] They have no interests that cannot be sacrificed, even when the "benefit" to be gained by humans is mere amusement at the cost of great pain or death to the animal. That is precisely what it means to be property.

"Persons" are precisely those beings who have interests that cannot be traded merely for consequential reasons alone. Some of these persons, such as corporations, are de jure persons in that their personhood exists solely because they are creations of a legal system. But what is common to every person is that persons have at least some interests, although not necessarily all the same interests, that are protected (by moral theory or law or both) even if trading away those interests will produce consequences that are deemed to be desirable. All "persons" must have at least one interest that is protected from being sacrificed merely for consequential purposes; the interest in continued existence, without which all other interests would be meaningless. This is Shue's concept of the basic right of physical security.

Animal rights theory generally seeks to move at least some nonhumans from the "thing" side of the "person/thing" dualism over to the "person" side. There are at least two reasons in support of this move. First, those who support animal exploitation argue that animals are qualitatively different from humans and so animals can be kept on the "thing" side of the "person/thing" dualism; animal rights advocates argue that there is no such distinction because at least some nonhumans will possess the supposedly "exclusive" characteristic. Nor is it enough to argue that species difference alone is morally relevant; after all, to rely on species alone as morally relevant is to assume a distinction that needs to be proved by those who hold such a view. And, it is morally indistinguishable from using race, sex, sexual orientation, or ability to determine membership in the moral community of persons. In other words, there is no reason to exclude animals from a progressive concept of personhood. [FN42]

Second, is another related, more "positive" reason to view animals as persons. Although there will undoubtedly be borderline cases, it is clear that at least some animals possess the characteristics that we normally associate with personhood. For example, in The Case for Animal Rights, Tom Regan argues that theoretical and empirical considerations indicate that at least some animals (normal mammals of at least one-year of age) possess beliefs, desires, memory, perception, intention, self-consciousness, and a sense of the future. The attribution of at least several of these mental states reveals that it is perfectly sensible to regard certain nonhumans as psychophysical individuals who "fare well or ill during the course of their life, and the life of some animals is, on balance, experientially better than the life of others." [FN43] Because animals have desires, beliefs, and the ability to act in pursuit of their goals, they may also be said to have preference autonomy, an important characteristic for the attribution of rights.

A common misconception is that animal advocates argue that animals should have the same rights as humans. As far as I am aware, no rights advocate maintains this view. Moreover, the criticism itself indicates a fundamental confusion about rights theory. In many ways, animal rights theory is about the inclusion of nonhumans on the "person" side of the "person/thing" dualism. This matter of inclusion is to be distinguished from the matter of the scope of any rights that animals may have once we move them from one side to the other. I have elsewhere used the example of human slavery to illustrate this point. [FN44] Although human slaves in the United States were regarded under the law as "persons" for purposes of criminal liability, they were, for virtually all other purposes, both de jure and de facto "things." This status as a "thing" is a logical consequence of the institution of human slavery which treated all slave interests--including Shue's basic right of physical security--as tradable as long as there were perceived benefits for slaveowners. Slaves had no rights of association, slave families were routinely broken up, and slaves could be killed or tortured for what was essentially the pleasure or amusement of slave owners.

To include slaves in the class of persons--moving them from the "thing" side of the "person/thing" dualism to the "person" side--is not necessarily to say anything about the scope of particular rights that the liberated slaves may have had. Indeed, the move entailed the exclusion of only one sort of exploitation: the institutionalized commoditization of human beings in which their basic right of physical security, the prerequisite for their having rights at all, was violated by others for consequential reasons. Other considerations governed the scope of rights that these "new" persons may have had. For example, the abolition of human slavery only began, and did not end, a discussion about what additional rights--other than the right not to be slaves--should be accorded to former slaves.
Similarly, when we move at least some nonhumans from the "thing" side over to the "person" side, we have said nothing about the scope of rights that they will have. All we have done--through the inclusion of animals on the "person" side--is to recognize that species alone is an insufficient justification for treating nonhumans as "things." Species may be significant when we determine the scope of rights. For example, it would be absurd to discuss the rights of animals to drive or to vote or the right of an animal to get a scholarship to attend college. But the inability of nonhumans to adhere to rules of the road, choose intelligently among political rivals, or do calculus are all irrelevant to the basic notion of personhood. After all, we accept that some adult humans will be unable to perform basic functions but we still do not place them on the "thing" side of the "person/thing" distinction. We may very legitimately award a math scholarship to Jane rather than Simon based on Jane's superior mathematical ability. As long as Simon has had a fair opportunity to develop his mathematical abilities, using Jane's "intelligence" as a criterion for determining the distribution of the particular resource in question (educational benefits) is fair. But Jane's greater intelligence does not justify Jane treating Simon as her slave or otherwise placing Simon on the "thing" side of the equation.

There is, however, one sense in which including animals as members of the class of "persons" is very different from including additional humans within that class. If we acknowledge that Simon is not a "thing," the protection we have given Simon is at the same time quite significant (after all, the basic right to physical security is a prerequisite to all other rights), but also the bare minimum needed to distinguish Simon from being a thing. Saying that Simon is included in the class of persons says nothing about the scope of rights that he may have other than to say that we will protect Simon's right to be a person in that we will at least recognize de jure that Simon's basic right to physical security will be protected from being traded away for consequential reasons.

If, however, we recognize that animals are not "things," (that their basic right to physical security cannot be sacrificed merely because we think the consequences justify the sacrifice), then we can no longer justify the institutionalized exploitation of animals for food, experiments, clothing, or entertainment. These forms of institutionalized exploitation necessarily assume that animals are things whose interests are contingent on human desires. Once we recognize that animals are not "things," we can no longer justify the use of animals in experiments any more than we could justify the use of humans. We have at least de jure ruled out the institutional use of coerced humans in biomedical experiments. And, although many people will tolerate the payment of low wages to workers, few would similarly tolerate human slavery.

A primary result of according personhood status to at least some nonhumans would be to require the abolition of institutionalized animal exploitation. Once we recognize that animals are no longer "things," then we can no longer treat them as beings whose fundamental interests in their own lives may be sacrificed because we enjoy the taste of meat, or because we enjoy shooting pigeons, or because we enjoy the feel or look of fur or leather. That is, according personhood status to animals does not mean that we simply get more serious about whether a particular form of slaughter to produce meat is more "humane," or that we take animal interests more seriously in determining whether a particular experiment involving animals is "necessary." It means that we accept that the use of animals for food or science or entertainment or clothing represent forms of institutionalized exploitation that are logically inconsistent with the personhood of animals.

V. The Three Components of Moral Theory

To evaluate Singer's claims about the normative indeterminacy of rights theory, I will identify three separate normative components, or levels, of moral theory, and I will explore the relative normative guidance of the utilitarian and deontological approaches with respect to each component.

The first component is what the theory ideally seeks. That is, what state of affairs would the theory want to achieve were all other things equal. The second component provides normative guidance to the individual, on a personal level, in terms of what theory ideally requires. I will use the term, "micro- level" when dealing with this component of moral theory. The third component addresses incremental change. Do these theories, which propose ideal moral states, have prescriptions for how to effect incremental change in order to achieve the state of affairs for animals that would be required under the ideal state? The inquiry here is whether rights theory provides a prescription for how the individual can help move law and social policy in the direction of the ideal state of affairs (the abolition of all institutionalized exploitation) in addition to providing more specific and personal normative guidance to the individual (the subject of the second component). I will use the term "macro- level" when dealing with this component of moral theory.
An example may help put this in perspective. Assume that my overall goal is to achieve a completely pacifist world in which there is no violence. On the ideal level, my theory requires that there be no or substantially no acts of violence. On a micro-level, this theory may require that I not respond violently to others irrespective of provocation. On a macro-level, the theory may prescribe that I try to implement the ideal state of affairs (a peaceful world) through legislation that eliminates various forms of violence (such as laws that forbid the manufacture of guns). These are three very different aspects of moral theory.

A. Relative Normative Guidance: The Ideal and Micro Components of Moral Theory

The first component of moral theory--the ideal level--requires that we ask what the theory envisions as the ideal state that would be achieved if the theory under consideration were accepted. For Regan, the answer is quite clear; Regan's theory is a theory of abolition, and not regulation, of institutionalized animal exploitation. Regan objects to the treatment of animals exclusively as means to ends; to put the matter in legal terms, Regan objects to the property status of animals that allows all of their interests, including their basic interest in physical security that is a prerequisite to the meaningful recognition of other interests, to be bargained away as long as there is some sort of human "benefit" involved. This would necessitate complete abolition of those forms of animal exploitation that are dependent upon the status of animals as human property. These activities would include using animals for food, experiments, product testing, clothing, entertainment, or in any other way that fails to respect the inherent value of the animal.

Similarly, rights theory is reasonably clear concerning the "micro" component of moral theory. Just as rights theory condemns the institutionalized exploitation of nonhumans as a matter of social practice, it also condemns at least the direct participation in animal exploitation. After all, if a person advocates the abolition of human slavery because the institution of slavery is unjust, that person would presumably also conclude that ownership by the individual master of human slaves was also violative of the rights of that owner's slaves since slaves can only be subjected to the institution of slavery by being owned by someone. Similarly, the individual participates directly in the exploitative institutions by eating meat or dairy products, wearing animals, or using them in experiments. These institutions cannot exist without individual moral agents who choose to participate directly in the institutionalized exploitation.

Difficult moral questions will remain. It is impossible to avoid participation in institutionalized animal exploitation completely since virtually every aspect of our lives is involved in some way with the institutionalized exploitation of animals. So, the rights advocate is faced with decisions, for example, as to whether to use drugs that have been tested on animals, just as opponents of human slavery are faced with the decision to travel upon roads in the southern United States, many of which were laid originally with slave labor. But the rejection of institutionalized animal exploitation does resolve many of the moral questions that confront us. If animal rights means anything, it means that, as a society and as individuals, we can no longer countenance the institutionalized killing of animals for food as a matter of individual moral choice, any more than we can justify performing experiments ourselves, or wearing clothing made from animal skins or pelts. [FN45]

Singer's long-term goal is not the achievement of animal rights, or necessarily even the abolition of all animal exploitation. Singer's theory of animal liberation requires that we reject speciesism, which would, for example, prevent the use of animals in experiments in those situations in which we would not use humans who had the same interests at stake. But beyond this rejection of species bias, and the use of a theory of act utilitarianism that would treat animal interests seriously, Singer's theory of  animal liberation provides little normative guidance concerning issues of animal suffering and the killing of animals.

On one level, both Singer's equal-consideration theory and Regan's rights theory can be said to represent an "all or nothing" approach in that both theories describe ideal states that are far removed from the present reality of the human/animal relationship. Neither ideal state will be realized without a profound change in the current state of affairs, and that change is most unlikely to happen overnight. On the level of "ideal" theory, then, both theories describe "utopian" states that are far removed from the world in which we presently live.

On another level, however, Regan's theory provides a rather vivid description of the ideal state of affairs whereas Singer's does not. The clarity of the ideal state is important because that clarity will help to inform a more definite theory about how the individual ought to behave on the micro- and macro-levels of moral decision. It is easy to identify the practices to which Regan objects given that his target is the institutionalized exploitation of animals. To the extent that there is any lack of clarity, Regan's overall prescription that we stop using animals exclusively as means to human ends, and that we recognize that some animals are subjects-of-a-life, would eliminate the overwhelming portion of what Regan regards as activity that violates the rights of animals. There may, of course, be some "hard cases," but under Regan's theory, institutionalized animal exploitation can never be justified irrespective of consequences, just as human slavery is rejected as morally repulsive by most people, irrespective of any beneficial consequences that would occur were we to enslave humans.

Singer's utilitarian theory is different from traditional animal welfare in that Singer regards the long-term goal as animal "liberation," which is Singer's shorthand for a state of affairs that would accord equal consideration to the equal interest of animals. So, in this sense, Singer's long-term goal is arguably more progressive than the traditional welfarist approach as long as everyone agrees how to describe competing interests, and also agrees how to weigh those interests in light of the assessment of consequences--and agreement about such matters is not easy to achieve. But Singer's theory is similar to animal welfare because it requires that we balance the interests of humans against the interests of animals under circumstances that threaten to compromise the assessment of animal interests in any event.

There are at least six aspects of Singer's theory that portend great normative uncertainty at any level of application. For present purposes, however, I am concerned primarily with the ideal and micro-levels of moral theory. It is my view that certain aspects of Singer's theory render his view at the ideal level to be far more unclear than that offered by Regan. Additionally, Singer provides very little guidance on the micro-level to aid the individual in resolving those human/animal conflicts that are presented as part of everyday life in a society where certain sentient beings are treated as the property of others. I stress that the purpose of this discussion is not to present and analyze critiques of utilitarianism in  general, or even Singer's utilitarian theory of animal liberation in particular. Rather, I am responding to Singer's claim that rights theory is incapable of providing concrete normative guidance relative to the supposed clarity that Singer claims for his view over rights theory.

First, as I mentioned above, Singer's utilitarian theory requires some sort of empirical description of the consequences of acts. But it is often difficult to predict these consequences under the best of circumstances. For example, Singer's long-term goal is to ensure that equal human and nonhuman interests receive equal consideration in a balancing process that is as free of speciesism as is possible. Even if animal interests were taken seriously, as they would be in Singer's ideal framework, assessments of consequences of actions--especially actions that purport to effect systemic changes, such as legislation--are very difficult to assess before or after the fact. [FN46]

Second, Singer's theory requires that we make inter-species comparisons of pain and suffering. That is, in order to maintain that the equal interests of animals and humans ought to be treated equally, Singer's theory needs some notion of how we can measure (however imprecisely) inter-species experience. For example, he observes correctly that a slap that would cause virtually no pain to a horse may very well cause considerable pain to a human infant. "But there must be some kind of blow--I don't know exactly what it would be, but perhaps a blow with a heavy stick--that would cause the horse as much pain as we cause a baby by slapping it with our hand." [FN47] The difficulties with making such assessments are obvious, it is difficult to compare pain intensity when we are concerned only with humans who can give detailed verbal reports of the sensation that they are experiencing--it becomes virtually impossible to make even imprecise assessments when animals are involved.

Third, and related to the problem of inter-species comparisons of pain and suffering, is Singer's analytic framework. Although the framework requires that we reject speciesism, Singer acknowledges that species differences may very well affect our assessment of these various interests. [FN48] In some instances, these differences will be obvious and their use will not be controversial. For example, no one (as far as I know) maintains that because of differences in the type of intelligence that exists between a human and a dog, scholarships for higher education ought to be given to dogs. But in many cases in which there is a purported conflict between animal and human interests, the differences may not be as obvious and their use may be far more controversial. For example, even if we can ascertain what type of blow, when delivered to a horse, will cause the same amount of pain as a sharp slap will cause a human infant, the question still remains as to whose interest in pain avoidance should be sacrificed in the case of conflict. Singer claims that pain is pain irrespective of "whatever other capacities, beyond the capacity to feel pain, the being may have," but those capacities may very well be relevant to an assessment of suffering and to the ultimate determination as to whose interests should be protected in the case of conflict. So, there can be considerable controversy as to whether the horse's mental capacities, which differ from those of the human, will result in more overall suffering by the horse, who may be terrified to a considerable degree for a short period of time as the result of the blow, or the human, who may not only experience the pain, but who may experience anxiety over a longer period of time, or who, as a result of different mental capacities, may anticipate another blow or be more distressed by the blow because of memories of physical abuse suffered earlier. Singer could, of course, reply that any interest balancing requires that competing interests be characterized as accurately as possible and that accurate characterization requires taking account of individual characteristics. This is, of course, one reason why utilitarianism is such a difficult theory to apply in the real world, even when animal interests are not included in the calculus. When they are included, there is a tendency, as Singer's own work shows, to evaluate the characteristics of individuals by reference to species differences. This approach both invites and facilitates introduction of humanocentric notions about animal consciousness. In any event, even if the individual characteristics and capacities of particular animals or species could be ascertained with some degree of empirical certainty, it would still be virtually impossible to ever apply this framework in concrete circumstances given the inexhaustible differences among individuals.

Fourth, when Singer turns from pain and suffering to the morality of killing animals, he again explicitly allows for consideration of individual capacities. He concludes that a "rejection of speciesism does not imply that all lives are of equal worth" because

[w]hile self-awareness, the capacity to think ahead and have hopes and aspirations for the future, the capacity for meaningful relations with others and so on are not relevant to the question of inflicting pain--since pain is pain, whatever other capacities, beyond the capacity to feel pain, the being may have--these capacities are relevant to the question of taking life. [FN49]

It is precisely this view that leads Singer to conclude that it may be morally acceptable to eat animals who have been raised under intensive- agricultural conditions, as long as they are slaughtered humanely because, according to Singer, "it is not easy to explain why the loss to the animal killed is not, from an impartial point of view, made good by the creation of a new animal who will lead an equally pleasant life." [FN50] Once again, Singer's rejection of speciesism is tempered by his competing view that there are species differences concerning such matters as self-awareness, that most animals used for food purposes "cannot grasp that [they have lives] in the sense that requires an understanding of what it is to exist over a period of time," and that these capacity-differences are relevant to moral assessments about killing.

Fifth, as the preceding points make clear, Singer's rejection of speciesism when "cashed out" is really quite formalistic and is almost impossible to apply in concrete circumstances because of the difficulty of assessing inter-species pain and suffering in the absence of considering species differences, which, when applied to make relative assessments of pain and suffering, and for the purposes of determining the morality of killing animals, make any practical application virtually impossible. Part of the problem here is that there is a tension between Singer's rejection of speciesism and his utilitarian theory. Indeed, Lawrence Finsen and Susan Finsen argue that although Singer defends a utilitarianism theory, he "presents an important objection to the current treatment of animals that is not based on a utilitarian calculation but expressed in terms of demanding that we avoid speciesism." [FN51]

Singer's own rejection of speciesism may not be justifiable in light of utilitarian moral theory, which is why this rejection is so carefully qualified by considerations of capacity in the assessment of overall interests in avoiding pain and suffering, and in the assessment of the morality of killing animals. In any event, to the extent that Singer accepts a nonconsequential element (the rejection of speciesism irrespective of consequences) in his theory, there is an inevitable tension with his overall view that even speciesism can be morally acceptable if the aggregation of consequences so indicates. This confusion and uncertainty, and the resultant tension between rejecting speciesism but purporting to judge the morality of acts based solely on consequences, makes Singer's theory even more difficult to understand and to apply.

Sixth, although Singer is an act utilitarian, it is not ever clear whether, on the micro-level of moral decisionmaking, Singer requires an application of his utilitarian theory--or whether he argues for something else. It is unclear whether Singer believes that the individual moral agent should pursue the action that will have the best overall consequential effect, or whether he requires only that the agent seek to educe suffering and minimize pain. As I argue below, the reduction of suffering--and not that moral agents should assess what action will most reduce suffering--is certainly what Singer advocates on the macro-level of social and legal change.
In sum, Singer's principle of equal consideration for equal interests may sound simple, but it is not at all clear what it requires at the ideal level, and practical application on the micro-level is almost impossible because of uncertainty and controversy surrounding the assessment of consequences, the characterization of competing interests, and the weighing of those interests. But even if the uncertainty was reduced, and the controversy diminished, the question of animal use would still have to be evaluated on a case-by-case basis. And herein lies what is perhaps the most important difference between rights theory and welfare theory for purposes of applying either to concrete situations. Singer may be correct to say that rights theory in general can become complicated in light of complex rule formulations and ranking structures to govern rights conflicts, but Regan's rights theory provides relatively clear and unambiguous normative direction at the long-term level and on the level of personal moral choice as that choice involves the institutionalized exploitation of animals. Regan argues that his long-term goal is the abolition of the institutionalized exploitation of animals and that if we accept that animals have at least the basic right not to be treated exclusively as means to human ends, then certain animal uses, such as the eating of animals, the use of animals in experiments, or the killing of animals to make clothes, must be morally unjustified.

This is not to say that rights theory does not leave many questions unresolved, even at the level of long-term theory. For example, if we assume that animals have the rights that Regan attributes to them, there may be a conflict between human and animal rights, such as when humans seek to build housing for other humans that is needed but that will displace nonhumans. In such cases, rights theory may become more complicated because criteria would need to be devised to decide what to do when rights conflict. But, for the most part, the overwhelming instances of animal exploitation are ruled out from the start in Regan's theory, where, under Singer's view, they are all ruled in unless Singer can demonstrate that the aggregation of consequences indicates otherwise. Indeed, even if we started with the presumption that most animal exploitation will also be ruled out under Singer's theory as a prima facie or initial matter, whether that particular type or instance of animal use should be allowed (because it maximizes overall utility) is still open to discussion because its initial exclusion may not be justified under Singer's own theory. Singer cannot have such a bright line because, as an act utilitarian, he is precluded from arguing that institutionalized exploitation is always wrong because it violates the interest of animals in not being regarded as property.

B. Relative Normative Guidance: The Macro Component of Moral Theory

There is one more level of moral theory remaining against which we must assess the claim that animal rights theory is "utopian," "unrealistic," or "absolutist." That inquiry requires that we examine the macro aspects of these various theories to determine what each prescribes for an advocate to achieve the ideal state of affairs for animals.

I have been unable to find a single instance in which animal rights advocates support the notion that there is any possibility of immediate action that will result in the immediate abolition of all institutionalized exploitation. The only way that such an effort could succeed is if huge numbers of people were willing to rise up in what would probably be a very violent confrontation given the large numbers of people who are involved in institutionalized exploitation and the capital that they control. But, if there were a sufficient number of people to make such a scenario even remotely likely, I suspect that the confrontation would be unnecessary because that number of people (and it would have to be a most considerable number) would be able to effect dramatic changes in the treatment of animals through political means and would not have to resort to such a violent revolution.
Rights advocates must necessarily accept some theory of incremental change if they are going to pursue social and legal change that impels motion toward the ideal state of the abolition of institutionalized exploitation. They have no choice but to do so. Even if the rights advocate eschews all incremental legislative or judicial change, the advocate is relegated to educational efforts, protests, boycotts, or demonstrations that can, under the best of circumstances, move toward the ideal goal of rights theory only through the incremental increase in the number of people who accept that animal exploitation should be abolished and who eschew, at least, direct animal exploitation on a micro-level.

Based on the structural defects of animal welfare and the legal and political institutions that enforce some version of animal welfare based on the property status of animals, there are probably some compelling reasons for an animal rights advocate to spend her limited time and resources on these more indirect incremental changes through various forms of education, protest, and boycotts. The primary reason is that judicial or legislative change sought by formal "campaigns" requires some sort of "insider" status as discussed by Robert Garner. [FN52] Once an animal advocacy group decides to pursue activity other than public education, or, more precisely, once the group decides that it wants to have an affect on legislation or regulatory policy, it becomes necessary to decide whether to seek "insider status" in order to "achieve access to government" and "to influence policy makers." [FN53] Garner states that it "is easy to see why insider status is regarded so highly. Access to government gives groups an opportunity to influence policy development at the formulation stage, thereby avoiding the difficult and often fruitless task of reacting against government proposals" which "are unlikely to change fundamentally" once they are formulated. [FN54] Garner recognizes that this "insider" status may be used to marginalize animal advocates through, for example, the creation of government advisory bodies that do little if anything but that give the mistaken impression that animal concerns are being taken seriously. Nevertheless, he holds to the view that "insider status can allow pressure groups to have a significant input into the formulation of public policy. This insider status, however, is largely dependent upon a group being perceived by government as moderate and respectable." [FN55] Garner observes that although moderation and respectability are relative terms, "it is clear that the radical demands of the 'rights' faction of the animal protection movement are not regarded as acceptable enough" to give rights advocates "insider" status. [FN56] Garner argues that insider status is necessary for animal advocates to be effective, yet states explicitly and repeatedly that despite the fact that moderate animal welfarists have enjoyed insider status, "the animal protection movement has made relatively little progress in influencing decision makers." [FN57]

There is nothing in rights theory that necessarily precludes the animal advocate from pursuing incremental legislative or judicial change, but I do not think that we can speak meaningfully of legal rights for animals as long as animals are regarded as property. To put the matter in the context of my earlier discussion of basic rights, as long as animals are property, then their basic rights, or those rights that are a prerequisite for the enjoyment of other, non-basic rights, can be sacrificed as long as some socially recognized "benefit" is found to exist. As long as we can kill animals for food, or use them in experiments, or imprison them for their entire lives in cages so that we can be amused at zoos, or maim them for our amusement in rodeos, or shoot them for fun at yearly pigeon shoots, then, to say that animals have rights is, as Shue observed, using "rights" "in some merely legalistic or otherwise abstract sense compatible with being unable to make any use of the substance of the right." [FN58] Basic rights are a prerequisite to the enjoyment of non-basic-rights, and the possession of non-basic-rights in the absence of basic rights is meaningless.

My critics will respond that every movement achieves rights incrementally. For example, Henry Spira "notes that in social movements, progress is made incrementally, through continual reform. 'If you push for all or nothing, what you get is nothing."' [FN59] Spira attempts to compare incremental progress made in other social movements to incremental progress made toward the abolition of animal exploitation. This attempt fails for the reason that no other situation--with the exception of slavery--is comparable with respect to the baseline protection afforded to animals. When we talk about incremental progress made in other social movements, we are talking about rightholders who seek a greater scope of rights protection. A "reform" in another area-- improved labor conditions for factory workers, for example--operates in the context of actors who already have basic rights that are sought to be extended. Put simply, we do not just arbitrarily kill and eat factory workers. Although interests may be balanced, some interests, such as the right of the worker not to be arbitrarily killed by the boss, cannot be traded away because those interests simply are not on the table. But because animal interests are treated in a completely instrumental manner, and all animal interests may be sacrificed if animal owners decide that there is a benefit in doing so, then the animal will virtually always be on the short end of the stick because we will almost always presume that property owners are the best judges of whether a particular use of their property, including their animal property, will be a "benefit" to them.

To put the matter another way, once we have persons who are at least holders of basic rights, it makes sense to talk about making incremental reforms in rights. But the basic right not to be treated as property is a right that does not and cannot admit of degrees, at least in this sense. Indeed, the issue is not whether we achieve animal rights incrementally, but whether we can incrementally eradicate the property status of animals because, in a sense, we are really only taking about one right--the right not to be treated as property. A recognition of the validity of that one right would compel the conclusion that institutionalized animal exploitation violates principles of justice that could be tolerated only as long as animals are classified as property, which gives humans license to ignore the basic similarities between humans and nonhumans that are relevant for attribution of the status of being a subject-of-a-life. But rights theory does not really concern the particular rights that animals have; rather, it asks whether animals should be in the class of rightholders as an initial matter. Answering this question in the affirmative does not commit the rights advocate to particular animal rights beyond the right to respectful treatment, which precludes institutionalized exploitation, but does not address much beyond that basic right not to be regarded as property, or, put in Regan's language, not to be treated exclusively as a means to an end.

Despite my view that it does not make sense to talk about animals having rights in a society in which they are regarded as property, my reservation is related to the notion that under the animal welfare paradigm that currently regulates the human/animal relationship, any animal interests that are recognized will almost always be subject to being sacrificed in the face of even trivial human interests. The only way that this will change is if the characterization of animals as property changes and moves closer to personhood--which is another way of saying that animals cannot have any non- basic rights until they get the basic right of not being regarded exclusively as means to human ends. The question becomes whether there is a way that this right--the right not to be regarded as property--can be achieved incrementally in a manner that is consistent with animal rights theory. This question can be rephrased as whether there is any way to change incrementally the legal status of animals that is consistent with rights theory. One thing that the rights advocate cannot do, and remain consistent with rights theory, is use welfare reforms to achieve this goal incrementally because such reforms, which necessarily assume the legitimacy of the property status of animals, only reinforce the property characterization and cannot create rights in animals.

I have elsewhere argued that incremental change is arguably consistent with rights theory as long as the incremental change represents a prohibition of some significant form of institutionalized exploitation, and when the prohibition recognizes that nonhumans have at least some interests (outside of those that must be recognized in order to exploit the animals) that cannot be traded away irrespective of the consequences for human beings. [FN60] For example, a law that prohibited the use of nonhumans for particular types of experiments (such as drug addiction experiments), irrespective of the expected benefit for humans, would qualify as a prohibition that represented that animals had interests that could not be traded away irrespective of the expected consequences for humans. Although animals would still otherwise be regarded as "things" with no right of physical security, they would at least enjoy some deontological protection for interests that are themselves part of that basic right. This sort of protection is certainly more significant than a norm that prohibits nothing but "inhumane" treatment understood primarily as that conduct that exceeds what is necessary to ensure the economically efficient exploitation of the animal in light of the type of institutionalized exploitation that is involved. Moreover, it is important that animal advocates not suggest or support alternative, and supposedly more "humane" forms of exploitation as "substitutes" for the exploitation to which the advocates object in the first instance.

In formulating these criteria, I have relied on only two aspects of rights theory. My reason for proceeding in this manner is to try to keep my criteria as uncomplicated and uncontroversial as possible. The first aspect on which I focus is that rights theory seeks the abolition of the institutionalized exploitation of animal subjects-of-a-life, which involves treating animals exclusively as means to ends. Put in legal language, rights theory seeks the eradication of the property status of nonhumans. This aspect of rights theory reflects that animals have interests other than merely being protected from pain and suffering, and that animals have an interest in not being part of institutionalized exploitation that causes the pain and suffering in the first place. The second aspect is that, in seeking this long-term goal, the rights advocate cannot endorse the sacrifice of fundamental interests of some animals today in the hope that some animals tomorrow will no longer be treated as the property of human owners. All subjects-of-a-life have equal inherent value, and it violates the respect principle to ignore the inherent value of any such being because some other beings would "benefit" from ignoring that value. Indeed, such a trade-off is a defining characteristic of the utilitarianism that Regan rejects.
Singer's view of incremental change is ostensibly more simple, but again, this simplicity is deceptive. Singer argues that animal advocates should support "any legislation that reduces the suffering of animals or enables them to meet their needs more fully." [FN61] The problem is that no one, including the animal exploiters themselves, disagrees with the view that we ought to minimize or reduce suffering. I have also argued elsewhere, that as a direct result of confusion on this point, animal advocates frequently find themselves in the position of supporting the same proposals that are advocated by institutional animal exploiters. For example, in recent years, animal "rights" advocates have joined with the American Meat Institute to promote guidelines for more "humane" slaughter. [FN62]

Singer would, of course, object and argue that he has a very definite understanding of what sorts of action will "reduce" or "minimize" suffering. But that is precisely the problem. Without any sort of theoretical criteria to delimit what incremental changes, which supposedly "minimize" or "reduce" suffering, are desirable, the welfarists are incapable of distinguishing their program for strategic change from that of the exploiters themselves, all of whom agree that animals ought to be treated "humanely." [FN63]

The problem can be illustrated clearly with the following example. Assume that animal advocates criticize experiments in which animals are burned at a high temperature without the benefit of anesthesia for five minutes. In response to the protest, the experimenter proposes a "compromise" under which the experiment will be conducted, but will be limited in duration to four minutes and fifty-nine seconds instead of five minutes. There is arguably a reduction of suffering. There is even more arguably a reduction of suffering if the proposed compromise duration is four minutes and thirty seconds. I would suspect that Singer (and most animal advocates) would not find this to be an acceptable position despite endorsing the view that "any" measure that reduces suffering is acceptable.

Singer desires as a long-term goal the treatment of animals that would be dictated by a theory of act utilitarianism informed by the principle of equal consideration for equal interests. That is, if Singer were able to construct his ideal moral world for animals, animals would be treated in such a way that their treatment would maximize the pleasure and preference satisfaction for all beings who are affected. In making these determinations, animal interests would receive as much consideration as the equal interests of human beings. Putting aside the problems that I identified about knowing how such a principle would translate in real-world terms, we can identify two separate elements that constitute Singer's position: 1) endorsement of the principle of act utility, according to which individual acts (and not classes of acts) are to be tested against the principle of utility; and 2) endorsement of the principle of equality, according to which the equal interests of beings are accorded equal consideration without reference to considerations of race, sex, or species.
It is difficult to understand how Singer relates these notions to his view that animal advocates ought to support any measure that they think will reduce suffering. Both aspects of Singer's theory are conspicuously absent. Singer does not seem to subject any particular incremental measure to any analysis using either aspect. He does not, for example, require that any particular incremental measure reduce suffering more than possible alternatives. He does not even urge that as a proactive measure, animal advocates should assess the competing options and pick the one that will reduce suffering the most. Part of the problem is related to the fact that it is difficult to know what the consequences of various options will be if the primary or sole concern is the reduction of animal suffering. After all, whether the federal Animal Welfare Act reduces animal suffering is anyone's guess, and the consequences of that law in terms of reducing animal suffering could be debated forever. Recognizing these problems, Singer urges that we simply support "any" measure that "reduces suffering." But that is the same as providing no guidance on incremental measures, or at least no guidance that serves to differentiate the incremental measures that should be supported by animal advocates from the welfarist reforms that are supported by animal exploiters.

Similarly, although Singer's major contribution is his argument against speciesism (or in favor of according equal interest to equal considerations without species bias), he nowhere requires that this portion of his theory be applied to incremental change on a macro-level. Singer does not maintain that incremental changes have to be those that are untainted by species bias--and, if he is going to support "any" measure that he thinks will "reduce" animal suffering, he cannot use the criterion. For example, many new welfarists regarded the 1985 amendments to the Animal Welfare Act as representing incremental change in the direction of animal rights. [FN64] But those amendments explicitly assumed that it was morally acceptable to use animals in experiments under conditions in which similarly situated humans could not be so used. There was, however, no consideration of this by the new welfarists; nor, as far as I am aware, was any such consideration urged by Singer.

VI. Conclusion

Singer argues that as a general matter, rights theory possesses weak normative force and is incapable (or more incapable than utilitarianism) of proving specific normative guidance in concrete situations. This is incorrect. For Regan, rights theory requires the abolition of institutionalized animal exploitation and, in practical terms, this would mean that we would no longer eat animals, or use them in experiments, for clothing, or for entertainment. Whether humans or other nonhumans would benefit from the institutionalized exploitation of animals is not relevant, because the respect principle simply rules such considerations out as a result of the equal inherent value possessed by all rightholders. For Singer, whether institutionalized exploitation would be abolished or modified, and if the latter, in what ways, would be open to question on a case-by-case basis, because Singer, as an act utilitarian, is committed to applying the principle of utility on a case-by-case basis. Although he rejects speciesism, and requires that equal interests be given equal consideration, we saw that such a principle provides very little normative guidance as to how animals (or anyone else) ought to be treated.
Moreover, the clear normative guidance provided by rights theory concerning the abolition of institutionalized exploitation provides concrete normative guidance on the level of individual moral choice. And on the third level of theory--the macro-level--rights theory allows for incremental change. Such change, however, should not be regarded as the incremental achievement of rights as a general matter. Part of the confusion that plagues the modern animal protection movement is connected to the failure to realize that rights theory has at its core the rejection of the property status of animals. In this light, the issue of incremental change is understood as the incremental eradication of this property status.

[FNa]. Copyright 1996 by Gary L. Francione. All rights reserved. This paper was originally presented at a meeting of the American Philosophical Association, Eastern Division, December 29, 1996. Some of the material herein appears in Gary L. Francione, Rain Without Thunder: The Ideology of the Animal Rights Movement (1996); Gary L. Francione, Ecofeminism and Animal Rights, 18 Women's Rts. L. Rep. 95 (1996) (book review); and in Gary L. Francione, Animal Rights and Animal Welfare, 48 Rutgers L. Rev. 397 (1996). The author acknowledges assistance from Anna Charlton, Priscilla Cohn, and Drucilla Cornell.

This article is dedicated to my nonhuman friend, Emma, whose personhood is completely clear in my mind.

[FNaa]. Professor of Law and Nicholas deB. Katzenbach Scholar of Law and Philosophy, Rutgers University School of Law--Newark. Professor Francione is also faculty director of the Rutgers Animal Rights Law Center.

[FN1]. Peter Singer, Practical Ethics 2 (1979).

[FN2]. Id. at 3.

[FN3]. Id.

[FN4]. Id.

[FN5]. Ingrid Newkirk, Total Victory, Like Checkmate, Cannot Be Achieved in One Move, Animals' Agenda, Jan./Feb. 1992, at 44.

[FN6]. Id. at 45.

[FN7]. Mark Harris, The Threat From Within, Vegetarian Times, Feb. 1995, at 70 (quoting Henry Spira).

[FN8]. See generally Gary L. Francione, Rain Without Thunder: The Ideology of the Animal Rights Movement (1996) [hereinafter Francione, Rain Without Thunder]; Gary L. Francione, Animal Rights and Animal Welfare, 48 Rutgers L. Rev. 397 (1996).

[FN9]. For a discussion of the relationship between Singer's utilitarianism and animal welfare theory, see Francione, Rain Without Thunder, supra note 8, at 54-62. Singer may respond that, as a utilitarian, he believes
that all human interests are able to be traded away for consequential reasons alone. Even if Singer would maintain this position, the humanocentric evaluation of consequences is likely to be evaluated differently when nonhumans are involved, for a number of reasons, including that it is humans who do the evaluating.

[FN10]. Bernard E. Rollin, Animal Rights and Human Morality 12 (rev. ed. 1992).

[FN11]. Id. at 137.

[FN12]. Id. at 140.

[FN13]. Robert Garner, Animals, Politics, and Morality 34 (1993).

[FN14]. Id. (emphasis added).

[FN15]. Singer, Practical Ethics, supra note 1, at 12-13.

[FN16]. See generally Peter Singer, Animal Liberation 1-23 (2d ed. 1990). The term "speciesism" was first coined by British psychologist Richard Ryder.

[FN17]. Peter Singer, Ethics and Animals, 13 Behavioral & Brain Sci. 45, 46 (1990).

[FN18]. Singer, Animal Liberation, supra note 16, at 8.

[FN19]. James M. Jasper & Dorothy Nelkin, The Animal Rights Crusade 5 (1993).

[FN20]. Singer, Ethics & Animals, supra note 17, at 46.

[FN21]. Id. (emphasis in original).

[FN22]. Garner, supra note 13, at 27.

[FN23]. Singer, Animal Liberation, supra note 16, at 228.

[FN24]. Id. at 229.

[FN25]. Id. The context of Singer's comments involves an examination of the argument that meat eaters actually do animals a favor by causing them
(directly or through consumption demand) to come into existence in the first place. Singer acknowledges that although in the first edition of Animal Liberation, he rejected this view as "nonsense," id. at 228, he is now uncertain about its validity and concludes that it is difficult to deny that bringing a being into the world confers a benefit on that being as long as the being has a pleasant life. This leads him to the view that it may be morally permissible to eat animals who have been raised and slaughtered humanely.

[FN26]. Id. at 229-30.

[FN27]. Id. at 230. Singer argues that if a being does have desires for the future or a continuous mental existence, then it would be wrong to kill that being even if the killing were painless. The problem with this view is that it is inconsistent with Singer's utilitarian theory. The fact that x may have future desires may count against killing x because the frustration of x's future desires is a negative consequence for a preference utilitarian like Singer. But Singer cannot maintain that there is any absolute rule against killing such a being because the aggregation of consequences may militate in favor of such killing.

[FN28]. Tom Regan, The Case for Animal Rights 221 (1983).

[FN29]. Id.

[FN30]. See R.G. Frey, Rights, Killing, & Suffering 197-203 (1983).

[FN31]. Regan, supra note 28, at 222.

[FN32]. The respect principle is a type of Kantian "transcendental" principle that Kant regarded as unifying moral judgments.

[FN33]. Henry Shue, Basic Rights 20 (1980).

[FN34]. Id. at 19.

[FN35]. Id.

[FN36]. Id.

[FN37]. Id. at 20.

[FN38]. Id.

[FN39]. Id. at 21.

[FN40]. I do not wish to give the impression that Shue argues that animals ought to have basic rights since his book does not even address the question of animal rights.

[FN41]. For a discussion of the status of animals as property, see Gary L. Francione, Animals, Property, and the Law (1995). See also Gary L. Francione, Animals as Property, 2 Anml. Law (i) (1996); Gary L. Francione, Animal, Property and Legal Welfarism: "Unnecessary" Suffering and the "Humane" Treatment of Animals, 46 Rutgers L. Rev. 721 (1994).

[FN42]. This is essentially the approach employed in James Rachels, Created from Animals: The Moral Implications of Darwinism (1990).

[FN43]. Regan, supra note 28, at 82.

[FN44]. See Francione, Rain Without Thunder, supra note 8, at 179-80; Francione, Animals, Property, and the Law, supra note 41, at 110-12.

[FN45]. It is often argued that clothing made of nonanimal products, such as synthetics, may have unintended, but nevertheless serious, consequences for humans and animals alike. That may very well be, but there would be no difference in that circumstance from other situations in which unintended harm occurs. Although our use of synthetics may have deleterious but completely unintended consequences for the environment that adversely impact humans, this would not support the view that there is no difference between pollution that indirectly kills five people, and choosing five people at random for use in making products, such as clothing. Again, this reflects a view that "personhood" establishes certain limits, irrespective of consequential considerations.

[FN46]. For example, some argue that there has been a significant reduction in the use of animals as the partial result of welfarist legal reform and political pressure; others disagree, citing the unreliability of the data used, the analysis of that data, and the lack of empirical evidence that would establish any sort of causal link between the decline (if there is one in fact) and welfarist reform. Indeed, Frey and Singer are both utilitarians, and they disagree over the consequences of abolishing factory farming.

[FN47]. Singer, Animal Liberation, supra note 16, at 15.

[FN48]. Some scholars have accused Regan of the same problem based on his discussion of the following hypothetical: five survivors--four normal adults and one normal dog--are on a lifeboat. There is room in the boat only for four, and one of the occupants must be thrown overboard. Regan maintains that his rights theory provides an answer to the problem. Although death is a harm for the dog, Regan argues, death would be a qualitatively greater loss, and, accordingly, a greater harm, for any of the humans: "To throw any one of the humans overboard, to face certain death, would be to make that individual worse-off (i.e. would cause that individual a greater harm) than the harm that would be done to the dog if the animal was thrown overboard." Regan, supra note 28, at 324 (emphasis in original). It would, on Regan's view, be morally obligatory to kill the dog. Further, Regan claims even if the choice is between a million dogs and one person, it would still be obligatory under rights theory to throw the dogs overboard.

For criticism of this view, see S. F. Sapontzis, Morals, Reason, and Animals 219 (1987). See also Peter Carruthers, The Animals Issue 9 (1992). Ironically, one of Regan's most vocal critics on this point is Singer, who claims that a "theory that tells us that all subjects-of-a-life (including dogs) have equal inherent value [cannot] be reconciled with the intuition that it is the dog that must be sacrificed." Peter Singer, Ten Years of
Animal Liberation, 31 The New York Review of Books 46, 49 (1985).

To the extent that Regan allows for the resolution of this hypothetical problem by appealing to certain characteristics of the dog that Regan disallowed when he argued that all subjects-of-a-life have equal inherent value, his resolution is inconsistent with his general theory. But Regan's discussion of the lifeboat example is irrelevant to his general theory that animals ought not to be regarded exclusively as means to human ends, and, even if Regan is incorrect, the error does not affect his general theory. The lifeboat example explicitly assumes the absence of any institutionalized exploitation and the example can, therefore, not be used to support the view that rights theory could provide support for, say using animals to find a cure for cancer. Moreover, the lifeboat hypothetical deals explicitly with a "post-rights" situation; that is, the hypothetical concerns the content of rights that animals would have were they no longer regarded as the property of humans. As such, the hypothetical does not concern Regan's theory of basic rights. See Gary L. Francione, Comparable Harm and Equal Inherent Value: The Problem of the Routine Subordination of the Equally Virtuous, Between the Species (forthcoming 1997).

[FN49]. Singer, Animal Liberation, supra note 16, at 20.

[FN50]. Id. at 229.

[FN51]. Lawrence Finsen & Susan Finsen, The Animal Rights Movement in America: From Compassion to Respect 186 (1994).

[FN52]. See Garner, supra note 13.

[FN53]. Id. at 193.

[FN54]. Id.

[FN55]. Id. at 207-08.

[FN56]. Id. at 208.

[FN57]. Id. at 211.

[FN58]. Shue, supra note 33, at 20.

[FN59]. Harris, supra note 7, at 70 (quoting from Henry Spira).

[FN60]. See generally Francione, Rain Without Thunder, supra note 8; Francione, Animal Rights and Animal Welfare, supra note 8.

[FN61]. Kim W. Stallwood, A Conversation with Peter Singer, Animals' Agenda, Mar./Apr., 1994 at 27.

[FN62]. See generally Francione, Rain Without Thunder, supra note 8.

[FN63]. For example, Kenneth Shapiro, an animal welfarist who was has served as president of Animals' Agenda, and as editor of the Journal of Applied Animal Welfare Science, promotes the use of a six-step "pain scale" by experimenters to evaluate the invasiveness of their research. At the lower end of the scale, level 2, represents "laboratory experiments and field studies involving mild pain/distress and no long-term harm," includes "frequent blood sampling," "intramuscular injection, skin scrapping," "negative reinforcement" such as "mild electric shock" and "brief cold water immersion, "food deprivations" that do not result in more than a 10% weight loss, "water deprivation" slightly exceeding particular species' requirements (e.g., deprivation in rats of less than 18 hours), and "[p]rocedures involving anesthetized animals with mild post-operative pain/distress and no long-term harm." Kenneth J. Shapiro & Peter B. Field, A New Invasiveness Scale: Its Role in Reducing Animal Distress, 2 Humane Innovations and Alternatives in Animal Experimentation 43 (1988).

[FN64]. For a discussion of the 1985 Amendments to the Animal Welfare Act, see Francione, Animals, Property, and the Law, supra note 41 at 195-99. For a general discussion of the Animal Welfare Act, see id. at 185-207.