Top-Down and Bottom-Up Reflective Equilibrium

The methodology I will employ is a version of the method that John Rawls employed in his Theory of Justice (1971) in which he attempted to attain "reflective equilibrium" between commonsense moral intuitions and more abstract ethical principles. One can use this approach in either at "top-down" fashion beginning with an abstract ethical principle and using it to predict and guide moral conclusions about a range of cases calling for moral judgment. Or, on can begin by describing a range of cases which evoke moral intuitions, and then attempt to frame a more abstract ethical principle which would account for that pattern of intuitions. This is the "bottom-up" approach. In either case, the goal of research is to try to bring our moral intuitions in line with our ethical principles so that they align with one another.

One can do this either by revising the moral principles when they conflict with strong moral intuitions about cases, or by setting aside certain of our moral intuitions when they conflict with what our ethical principles predict should be regarded as the morally correct judgment. One must approach this task with an open mind and be willing to regard at least some of one pre-analytic moral intuitions as fallible or illusory, and also be willing to revise or even abandon
one's proposed ethical principles when they are incapable of being squared with our robust moral intuitions. There is, of course, no reason why one cannot employ both top-down and bottom-up kinds of reasoning in this kind of endeavor and this is the way in which I shall proceed.

My aim in the next chapter is to present a moral principle, which I will call the Vulnerability-Care Principle (VCP) and to try to make the case that it is a plausible candidate for the status of a fundamental ethical principle. For the moment I only wish to convince my readers that the VCP is plausibly thought to be capable of accounting for a wide range of standard moral intuitions which normally conscientious moral observers have about a wide range of moral cases. If I am successful in this, it will not show that the VCP is true in any interesting sense. In order to gain further justification for accepting the VCP as a fundamental principle of ethics one needs also to supply a general rationale for why there should be such a principle of ethics, to show how accepting the VCP as basic helps to illuminate and explain certain moral issues for which we normally think that other ethical principles are more appropriate, and how it helps us to resolve conflicts and solve problems in ethical theory and applied ethics.

Like scientific theories, ethical theories can gain credence by demonstrating they are competitively supported by the available evidence and that they cohere with our considered beliefs in related domains of inquiry. So, for example, the theory of evolution in biology that assumes that extant species evolved over very long periods of time until they reached their present states, must cohere with theories in geology concerning the age of the earth. If the earth were in fact very much younger than is now generally believed, it would imply that either biological evolution would have to work much more quickly than is usually assumed, or that the theory of evolution is false or at least incomplete.

Normative ethical theories about our rights and responsibilities will ultimately have to cohere with theories in the social sciences and in psychology about the nature of social relations and human motivation. If the ethical theory that features the VCP or something like it turns out for one reason or another not to cohere with facts and reliable theories about these matters, then it would count against its feasibility as a fundamental principle of ethics.

So the attainment of a reflective equilibrium between ones ethical theory and a range of moral intuitions is only the first step in providing a rational justification for believing that the VCP, or any ethical principle, is indeed a fundamental principle in ethics. I am not suggesting that can provide such a justification at the present time, but am only attempting to present the VCP as a plausible candidate for this status.

Fortunately, there have been other thinkers who have explored much of the territory I plan to cover and whose guidance I will be relying on for much of what I will have to say about the VCP. In particular, Robert Goodin and Virginia Held have pioneered this approach to ethical theory, and, as will become obvious, I am greatly in their debt as concerns the task of demonstrating the plausibility of the VCP.

My specific contribution will be to attempt to go further than either of these authors and to show that, when properly understood, the VCP is able to provide an account of what we normally think of as the social responsibilities derived from human rights. If I am successful in showing how this is the case, then we will be in a theoretical position in which becomes possible to connect the discourse of human rights with the discourse of social responsibility, and to show how human rights are derived from social responsibilities rather than the other way around, as is normally assumed.

Having made the connection between social responsibilities and human rights, we will then be in a position to argue that the range of our social responsibilities is wider than only those that ground human rights, and extend the VCP to the bio-centric and intergenerational realms. If this is successful, then I believe that I will have succeeded in making the case that the VCP is a plausible candidate for a fundamental principle of ethics, one which when properly understood, can provide a common normative framework for a global ethics of the kind I envision.

A Meta-Ethical Digression: Moral Pluralism

Meta-ethics is a set of theories about doing ethics. Ethics, considered broadly, is a normative theory about the nature of the moral life or the moral realm, that is, the realm in which we talk about things like values, duties, rights, virtues, responsibility, blame, guilt, and a variety of other moral concepts. Normative ethics is that branch of ethics that attempts to explain morality, that is, roughly, to give an account of what it is moral agents owe to one another as members of a moral community. I say roughly, because as I will define the notion of a moral community, it will include as members moral patients who are not also moral agents to whom (or to which) moral agents owe moral responsibilities. Morality concerns what it is we should do, and normative ethical theory attempts to give a systematic answer to this question.

In meta-ethics there is a theoretical dispute between the partisans of a monistic approach, and those of a pluralistic approach. Monists hope to find a single, comprehensive ethical principle which is capable of explaining all of our considered moral judgments about moral matters.

The leading candidate for this status is the principle of utility, particularly that version championed by John Stuart Mill, known as the Greatest Happiness Principle. According to this theory, what is morally right for moral agents to do is to act so as to maximize that happiness (or well-being) of all of those (sentient) individuals who are affected by our actions in the long term counting each individuals interest in happiness as equal. This is sometimes referred to as the ethics of universal benevolence. The principle of utility has many variants and many defenders, so many, that I do not have time to review them here. I want to focus only on that group of utilitarians who join this principle of normative ethics to the assumption of theoretical monism, that is, the idea that there is only one fundamental principle of ethics. It is this idea that I want to reject.

In my view, the duty to maximize utility is an ordinary standing moral responsibility like the duty to prevent harm, to protect and care for the vulnerable, the duty to do justice, or the duty to respect another person's autonomy. The mistake that some utilitarians make is that they try to portray the maximization of utility as a kind of "master principle" that encapsulates all other moral considerations. But from a pluralistic, deontological point of view, like the one I prefer, utility maximization is only one normative principle among many others with which it may agree or conflict. While it would be theoretically "sweet" to have a "master principle" in ethics, just as it would be theoretically sweet to have a grand unified field theory in physics, I do not believe that any such theory is in the offing, at least as far as normative ethics is concerned.

Instead, on my view, there will be a plurality of fundamental principles of normative ethics that together describe and explain the moral intuitions that normal, morally sensitive individuals have over a wide range of cases and contexts. In some cases and in some contexts, utility provides a useful moral guide to what conscientious moral agents ought to do. But it is not the only guide to normative rightness and must give way in certain kinds of cases to moral considerations deriving from other fundamental moral principles.
One can appreciate the pluralism of normative ethical principles by focusing on the nature of the arguments that are commonly employed against utilitarianism when it is cast in the role of the master principle of ethics. One finds counter-intuitive examples in which considerations of utility conflict with those of justice, for instance, in the case of the drifter who can be framed for a crime he did not commit. Or arguments involving conflicts between the duty to maximize impartial utility and duties of care that arise because of special interpersonal relationships. Or cases in which the duty to respect personal autonomy runs up against a attempt to do what ones knows to be in another person's best interests. In all of these kinds of arguments the basic structure consists in noticing that the duty to maximize utility conflicts with some other kinds of moral obligations derived from some other moral principle.

But this problem is not unique to the theory of utility or other consequentialist theories in ethics. The same kind of argument can be used to draw attention to conflicts between justice and care or between justice and autonomy or between autonomy and care, and so on. The conclusion that one should draw then is that there is simply no master principle of morality that can be used to guide moral decision-making and evaluation in all cases. As Kwame Anthony Appiah has put it, "Anyone looking for decision procedures, a way of ranking values or a set of rules for choosing among them, such be warned that 'naturalized ethics' is never going to get us there. This isn't because of any crevasse between 'is' and 'ought'; it is because there's no there there. Normative theories, if they are sensible, do not offer algorithms for action." (Experiments in Ethics. Cambridge MA: Harvard University Press, 2008, p. 193).

The meta-ethical stance known as moral pluralism supposes that rather than a single table of values and a single master principle of morality, what we have is a plurality of values and a plurality of fundamental ethical principles. The standard objection to this view is that it lacks theoretical simplicity and offers no means by which to decide which duties shall take precedence when duties derived from independent principles conflict with one another in practical cases. But while theoretical simplicity may be an important value in the empirical sciences, its value in the moral sciences is overrated.

The reason for this is that in normative ethics what we are aiming for, in part, is a social consensus about what kinds of actions and policies ought to be generally accepted as morally right. In order to justify any particular proposed normative policy to a lot of moral agents one needs to find what John Rawls called a overlapping consensus, that is, everyone may not agree to endorse a particular course of action or policy for the same reason, but if there is enough convergence among everybody's own reasons, then we can say that the policy has strong support, even though everybody's back story about why they endorse the policy may be a different one. Most favored policies and practices are those that are supported by the convergence of a variety of independent reasons deriving from various sorts of moral and non-moral considerations. They have what in science is termed "consilience", that is, support from a number of independent lines of evidence or argument. Having a plurality of fundamental moral principles and values is what makes such multiple, independent but sometimes intersecting kinds of justifications possible, and thus it is what allows us to achieve a broad-based social consensus.

The current global consensus on human rights is a good example of this kind of "many-legged" justificatory strategy. Human rights norms and values are justified by a variety of different sorts of moral and practical considerations deriving from considerations of justice, utility, nonmalefiecence, vulnerability, dignity, equality, convention, as well as by religious or metaphysical and metaphysical beliefs. There is no such thing as the justification for human rights. Rather there are a set of partially adequate overlapping justifications for various particular rights as well as a general set of philosophical and political rationales for holding that certain rights should be regarded as belonging to persons as such, irrespective of their particular identities. (See Morton Winston, "Human Rights as Moral Rebellion and Social Construction." Journal of Human Rights, Vol. 6, No. 3 2007: 279-305. for a fuller account).

A plurality of principles also enables us to achieve a better fit between our principles and our moral intuitions over a wide variety of kinds of situations and issues calling for moral reflection and decision. No single moral principle can do the job of describing our actual patterns of moral judgment as well as a set of multiple moral principles. It is not just that normative ethics is just a "messy" field of inquiry that has not yet achieved its true paradigm -- the moral life is just too complex to be reduced to a single over-arching theory of what makes actions morally right, what makes some things morally valuable, and what the good life for human beings consists of.

To borrow some terms from linguistic theory, ethical theories must strive to attain both descriptive and explanatory adequacy. To attain descriptive adequacy they must provide a plausible account of why people's moral intuitions about cases or situations calling for moral judgment or evaluation are as they are. In order to achieve this, it is often necessary to hypothesize the existence of a variety of moral rules and higher-level ethical principles, a moral grammar, that correctly predicts how ordinary competent moral observers will respond to cases calling for moral evaluation. However, there are likely to be many descriptively adequate ethical theories in this sense. Theory choice in ethics, as in other sciences, is underdetermined by the empirical evidence. So one needs to find other considerations to motivate the choice among competing normative theories. One then resorts to looking as theoretical parsimony, explanatory power, fruitfulness, coherence with theories in related domains of inquiry, and so forth, in order to find additional factors that can be used to help determine the choice of theories. But theoretical parsimony or simplicity should not be traded off against descriptive adequacy, in general, but especially in ethics. Because normative rules and principles are developed in order to guide the ordinary moral decision-making of typical moral agents, it is better that they be practical and accurate.

So, then, by advancing the Vulnerability-Care Principle as a fundamental principle of normative ethics I am not suggesting it is a "master principle" that supplants other fundamental principles of ethics. Nor should my narrative about vulnerability, dependence, care, and responsibility be construed taking the place of a much richer moral vocabulary that also talks about rights, justice, virtue, utility, and other matters relevant to the moral life. As in the case of other fundamental moral principles found in normative ethics, the VCP must compete with and often conflict with the demands of other moral principles, and when it does so, its victory is not assured in advance.
But because the VCP is a relatively under-studied principle of ethics, one whose theoretical value and importance is not widely understood or appreciated, I think it worth emphasizing it in order to reveal its potential. In my view the VCP is not just as a normative principle that can be used in the private sphere of the family, where it finds it most natural home, but also in the public sphere whether it is often considered not to apply at all or to apply in only limited ways.

The burden of my argument will be to make a plausible case that the VCP is indeed a fundamental principle of normative ethics, not to claim that it is the only or the most important one. But I do wish to claim that the VCP and the associated concept of social responsibility derived from it do helps to account for a wide range of moral intuitions we have about our moral responsibilities, and that looking at some problems in normative ethics from the perspective of vulnerability and responsibility reveals some interesting insights about the relationship of the VCP to other moral concepts, in particular, the concept of human rights.

Parental Responsibility

Undoubtedly, the paradigm case for the vulnerability model is parental responsibility towards infants and young children. The human child is dependent upon its parent(s) (or other adult caregivers) for all of the most basic necessities for survival, such as food, clothing, shelter, safety, etc.: "Indeed, biologists remark that the most salient feature of the human infant is its severe and protracted vulnerability. Man is more helpless for more of his life than virtually any other species. Somebody must be assigned the special responsibility of looking after the young. Who that is will, of course, be a matter for social determination; typically it will be the biological parents, at least in the first instance; but sometime it will not.

Whoever is picked out, however, the more basic point remains that those special responsibilities flow fundamentally from the child's special vulnerabilities" (Goodin 1985, 33). Here, the idea of vulnerability seems to be extremely appealing, both in terms of the generation of moral responsibilities on the part of parents towards their children as well as the specific contents of the moral duties that follow from them. Human infants and young children, as a particular class of moral patients, are particularly vulnerable to various kinds of neglect and abuse. They can be harmed by the failure of caregivers to provide them with adequate nourishment, shelter, and protection from various sorts of risks and threats, which is why responsible parents "child-proof" their homes.

Young children can also be improved in various ways, for instance, by providing them with educational opportunities, training, and privileges of various kinds through which they can develop their capacities and talents. Their parental or other caregivers are normally believed to have strong moral responsibilities to protect them from harm and to do many things which directly or indirectly benefit those children who are under their care. In this respect it is important to note that the ethics of care extends Goodin's VP by adding to it duties to benefit or improve the subjects of care, rather than only to protect them from harm. The VCP differs from the VP in this important respect. Under the VCP the responsible parties to the vulnerability-care relationship have specific moral duties to benefit the objects of their responsibilities in specific ways. The VCP combines what are commonly thought of duties of beneficence with those of nonmalefience, where both sorts of duties are understood as applying to both acts and omissions.

Duties of nonmaleficence are often thought of as stronger than duties of beneficence, and duties to avoid directly causing others to be at risk of harm are generally thought to be stronger than duties to prevent risks and threats that one did not directly cause. It is a curious fact that the English language seems to lack a specific term that corresponds to "vulnerable" but which means susceptibility to be benefited or made better off than one is.

One might suggest that the term whose meaning is closest to this sense is "corrigible"; to be corrigible is to be susceptible to improvement or benefit, or at least, that is the sense in which I shall employ that term here. So we can also posit a parallel moral relationship of corrigibility:

The Corrigibility Relation: A is corrigible to B with respect to C because of D.

The notion of corrigibility will be useful for discussing what are commonly thought of as duties of beneficence. Given this these terminological stipulations, we can describe in general terms four classes of moral responsibilities to avoid or prevent harm to the vulnerable or to help the corrigible that moral agents can have:

(I) Responsibilities to avoid harming others.

(II) Responsibilities to prevent harm coming to others.

(III) Responsibilities to benefit others.

(IV) Responsibilities to avoid preventing benefits coming to others.

In (I) if B acts in certain ways B would make some moral patient A worse off than they would otherwise have been. In (II) if B refrains from acting in certain ways A would be made worse off than if B had acted in those ways. In (III), by acting B makes A better off than he would have been had B not acted, and in (IV) B would make some moral patient better off than they would otherwise be by not acting. It is also worth noting that in (I) and (III) the agent is the direct cause of the harm or benefit in question, while in (II) and (IV) they are the intervening or indirect cause of the benefit or harm.

Parental responsibilities being the paradigm case for the ethics of care and vulnerability combine all of these kinds of special duties. Parents, can of course, delegate or assign some of their responsibilities to care for their young children to others, e.g. teachers, family members, or day-care workers. It is important to note that responsibilities can in general be delegated or reassigned in this way, which is one reason why I prefer to use the term 'moral responsibility' rather than 'duty'. When a parent delegates or reassigns his or her parental responsibilities, say to a baby-sitter or teacher, the parent or primary caregiver retains a supervisory responsibility to see to it that her designees are capable of adequately discharging the kinds of responsibilities appropriate for those placed under their care.

That the vulnerability of human infants should play a role in shaping the ethical responsibilities of parents or other caregivers carries strong intuitive appeal. The Christian icon of the mother and child is universally understood as representing the special moral relationship of care and vulnerability that exists between mothers and their children. The VCP has a dominant role in understanding and explaining the sorts of special moral responsibilities that mothers and fathers, and perhaps other family members, have toward a certain class of moral patients, infants and young children, who because of their immaturity are specially vulnerable and dependent on others for their care and protection.

But perhaps this not so clear when we consider other types of family relationships. Goodin argues it is implausible to analyze parental responsibilities on the model of promises and contracts where the agent comes to acquire a particular responsibility as the result of his or her own voluntary choices. One generally does not choose ones children in the way one chooses friends, business associates, or others we deal with on a daily basis. But, of course, this does happen sometimes when children are adopted. Other cases in family relations, however, are far from obviously associated with the vulnerability-care model. So let's examine some more of Goodin's arguments for extending the VCP to other kinds of special moral relationships.

Promises and Contracts

Promises and contracts and other types of explicit agreements provide the paradigm cases for the voluntaristic model of self-assumed moral obligations. Promissory obligations arise exclusively from voluntary acts of consent, as can be seen from the fact that coerced contracts are almost universally unenforceable (43). But, against the dominant view, Goodin argues that promises are nothing more than a way to coordinate behavior by reducing the uncertainty that would prevail in the realm of human action without them. Promises are made so as to allow others to make firm plans of their own by being able to anticipate what it is that a particular agent is going to do.

By making a promise the promisor allows the promisee to behave and plan accordingly; “What makes promises special is not so much that they represent a voluntary act of will on your part, but rather that the expectations about your behavior thereby engendered form crucial components in the plans of others” (44). By inducing such expectations, however, one also makes the other party vulnerable to betrayal. After the words “I promise” are uttered, the promisee is now depending on the promisor to do what has been promised, and when that expectation is not fulfilled he has been betrayed and his trust violated.

In other words, Goodin is arguing that the moral force of special obligations deriving from promises and contracts depends less on the volitional aspect of the act of promise-making than the vulnerability it engenders in the promisee by his reliance on the promisor’s intention to do what had been promised. As Goodin writes, “Special obligations do arise out of our voluntary [self-assumed] commitments. But what makes those obligations morally binding, I argue, is the vulnerabilities that those commitments engender; and those vulnerabilities are only one of many forms of vulnerability to which we should morally respond. Thus, the vulnerability model is a more general one, capable of subsuming and transcending the model of self-assumed obligations” (36).

Although volitional acts of promise-making are often the catalyst in generating such obligations, the vulnerability engendered in promises and contracts is far more important in terms of deriving the content and force of these obligations. Goodin argues for the primacy of the vulnerability concern in our conventional thinking about contracts by making reference to the legal notion of reliance. There can be obligations in the sphere of promises and contracts that are based merely upon one’s reliance (without an explicit contract or promise) upon another. Here Goodin notes that, “When one realizes that another is or may come under a misapprehension as to the authority of his agent or the ownership of his property - a misapprehension for which he is not at fault his duty to give information is a duty of care.”

That reasonable reliance on the intentions of another creates an implied obligation reinforces the idea that what is of primary importance concerning contracts and promises is the fact that an agent has rendered himself vulnerable to the actions of another by forming the conviction that that agent will make good on his expressed intentions.

Most often, this is the result of a specific promise or contract. But, it can also result from expressed intentions, intentions that induce others to act upon them where the agent has made no effort to qualify his intentions (i.e. asserting that it is very likely that he will not do what he has indicated in passing, etc.) We can see in such instances that the voluntaristic model cannot account for this common sense notion of moral responsibility, unless one relies on some notion like that of a tacit promise. Clearly, we generally ascribe a type of responsibility to a moral agent B who allows another moral agent A to believe and plan according to intentions that have been expressed by B. On the self-assumed voluntaristic model such considerations can be defended only with some difficulty due to the fact that there seems to have been no voluntarily assumed obligation on the agent’s part, yet B is responsible to A nonetheless because A is vulnerable and is depending on B to do that which he is relying on him to do.

The idea of contracts as fundamental elements of ethical and political theories should be familiar to most of us. The Enlightenment philosophers thought of government as being founded on a “social contract” and contractarianism is still an important feature of contemporary accounts of justice, such as that of John Rawls. Contracts are also regarded as the very foundation of business relationships, and some authors even think of corporations and other nonhuman social and economic agents as nothing more than an “nexus of contracts.”

But, Goodin argues that when we look beneath the surface we can see that there are a number of constraints placed upon contractual dealings that reveal that they depend upon the notion of vulnerability. For instance, in business relationships the legal codes that express duties of employers to employees reflects the relative vulnerability of employees in these relationships. We can see the same considerations within the law constraining the relationship between businesses and their customers. Although no explicit contractual obligations exist between business organizations and consumer, the law does place a heavy burden on business to protect certain interests of vulnerable stakeholders such as consumers. One such example is the special liability of a seller of products for the physical harm to users or customers. Sellers of goods to the public have special responsibilities towards consumers of their products to take reasonable steps to ensure that those products are safe and will not harm the health of their customers. Business enterprises, and those who work for them, have a variety of other kinds of social responsibilities which are not derived from contractual relationships nor imposed by legal requirements. This is a topic which we will explore in some detail when we discuss the notion of corporate social responsibility.

If Goodin's account of the moral force of promises and contracts is correct, or at least partly so, then it represents a major insight about the nature of moral responsibilities. It implies that many of the sorts of moral obligations which we normally understand to be based on contractual arrangements are in fact really based on the vulnerability-care relationship. If this is true, then the VCP begins to emerge as a plausible candidate for the status of a fundamental principle of normative ethics, one capable of bridging the private realm of the family and friends, with the public realm of social and political relationships among strangers.

Professional Responsibilities

Goodin also defends the vulnerability model in the case of those special responsibilities found in professional relationships. We can look at the relationships between professionals, e.g., doctors, lawyers, teachers, etc., and their clients as being shaped by the relative vulnerability of those clients to the actions and choices of the persons occupying the professional role.

Professionals generally speaking have strong duties to protect the interests of their clients, and these responsibilities often entail both duties of nonmaleficence and duties of beneficence. There are three factors Goodin sees as pointing to the fact that the specific duties of this type of special relationship are governed by vulnerability. First, if the relationship were merely contractual, it could not fully explain the fact that contracts drawn between the two parties are fixed to standard form. There is very little permission to negotiate terms between the involved parties themselves alone. Second, while professionals are free to serve whomever they choose, they must respond to any request for assistance in an emergency. Third, terminating a contract is more difficult for a professional; for instance, in a therapeutic relationship significant notice must be given to client, no neglect can occur, and extended time must be given for the client to secure an alternative.

These factors point directly to the unequal relationship between professional and client. This inequality, Goodin claims, stems from the high degree of specialized knowledge possessed by the professional, and the fact that the client is attempting to secure a basic need (health, legal status, education, etc.) whereas the professional is not, Further, it is usually easier for a professional to find other clients whereas the same may not be true for the client in finding other professionals. Goodin states it most clearly in saying that "clients are and must necessarily be relying upon professionals to protect them in crucial ways" (66). This, once again, points to the vulnerability model's ability to better explain such responsibilities. Goodin states that it is trust rather than mere contract that shapes these kinds of relationships (67). In essence, the client, who is usually the weaker party to the relationship, depends upon the professional in ways that the latter does not depend on the former. The patient must trust her doctor to do what his professional responsibilities require whether or not he has given her an explicit promise to do so.

Relationships between professionals of various kinds and their patients or clients exhibit the general characteristics of the vulnerability-care relationship: they are asymmetrical in that the parties to these relationships are unequal in power to affect one another's interests; one party has special knowledge or skills that can affect the well-being of the more vulnerable party for good or for ill; and the vulnerable party is in some way specially dependent on the professionals' exercise of responsible care in order to secure some aspect of his or her interest or well-being that she cannot secure by self-help alone. In cases such as these, the VCP predicts that the dominant or stronger party to the relationship has special moral responsibilities to protect the interests of those who are vulnerable and are depending upon them for their care. This kind of moral responsibility arises because of the name of the relationship between the parties, and does not depend upon there being explicit or assumed consent.

From Responsibilities to Rights

Before moving on to refine and develop the theoretical framework that I have only sketched up to this point, I want to mention one other potentially interesting point about the theoretical significance of the VCP.

Thus far we have been looking at the VCP from the point of view of a moral agent who is a position to care for and protect others who are vulnerable and dependent upon them. In such cases, the VCP predicts that the more powerful agent(s) should acquire special moral responsibilities to protect the interests of the weaker and more vulnerable parties to the relationship, particularly when those vulnerable others cannot protect their own interests and are in some ways dependent on others for their care and protection. But what happens when we turn this situation around and look at the matter from the point of view of oneself being the vulnerable party?

In this case, the VCP prescribes that one should regard oneself as entitled to certain kinds of care and protection from others. If one finds oneself in a position of relative vulnerability and dependence, then according to the VCP, one has a moral claim to the care and protection of others. In some cases, these sorts of claims can be addressed to specific others who are in special relationships with oneself, such as ones parents, doctors, teachers, and so forth, while in other cases the moral claims generated by the VCP can be addressed to society at large. If I am specially vulnerable and am depending on others, then I can claim that society has a responsibility to organize some form of care and protection that benefits me.

This insight I believe, suggest a way in which rights can arise from responsibilities. Rights are, at least in part, moral claims that persons advance against other members of society that invoke the responsibility to protect and care for vulnerable others. Thinking that the moral responsibility to protect the vulnerable precedes the existence of a right to claim social protection against some standard threats to ones liberty or well-being, then, offers an explanation for human rights that does not rely on either God or nature nor merely on social conventions. Rights, on this view, derive from the moral responsibility to protect the vulnerable.

If there is a fundamental moral responsibility to protect vulnerable members of one's moral community, then the right to be protected and to claim protection as one's moral right is logically prior to the existence of any social contract among members of society which serves to legitimize these moral relationships. There do indeed come to be moral and legal conventions that we agree upon that recognize specific rights and responsibilities, but from the perspective of the ethics of responsibility, rights derive from considerations of relative power and vulnerability. This turns on its head the standard view of rights in which rights are regarded as theoretically fundamental and are seen as providing the moral and/or legal basis for various kinds of social responsibilities.

The idea that a having a right gives its holder the moral basis to make a claim to social protection against certain kinds of threats can then be seen as the flipside of the social responsibility to protect the vulnerable. In cases in which the right-holder is vulnerable to harm and is depending on other members of society to come to his aid when he is under various kinds of threat, the VCP predicts that other members of his moral community have a standing moral responsibility to protect him. If there is such a standing moral responsibility to protect the vulnerable members of the moral community, particularly when their survival, well-being or freedom is threatened in a serious way, then when I am threatened I can claim social protection. So my right to social protection derives from the social responsibility to protect the vulnerable borne by other members of my moral community.

The human rights violations and abuses that are considered to be 'standard threats' to human freedom and well-being represent lessons learned from historical experiences of oppression in which large numbers of persons have in fact been deprived of these goods and few if any other members of society attempted to protect them. Vulnerability to human rights abuses must then be understood in a particular rather than in a general and abstract sense to refer to specific kinds of harms that persons have been subjected to in our historical experience, for instance, the threat of arbitrary arrest and imprisonment, the threat of torture, the threat of religious persecution, the threat of economic deprivation, the threat of disenfranchisement, and so forth.

Human rights are designed to afford individuals some degree of social protection against these specific kinds of standard threats, particularly when the threats originate because of acts or omissions by governments. But rights are effective normative instruments only to the extent that other members of society respond to the moral claims made in their name. It is the regularity and reliability of the social response to the moral and legal claims that rights provide that determines whether or not human rights are operative or only aspirational.

In other words, without a set of institutions that discharges the social responsibility to protect the vulnerable, human rights provide only the moral basis for claims to social protection, not the protection itself. A menu is not the same thing as a meal. Which is why the most important challenge for the global human rights movement in the twenty-first century is to create and support rights-implementing institutions and thereby make human rights operational for all the people of the earth. In order for this to happen, more people must be made aware of their social responsibilities to protect others against human rights violations and abuses.

Legalism and Responsibility

A somewhat different approach to the limitations of rights-based moralities has been developed by John Ladd ["Legalism and Medical Ethics." Contemporary Issues in Biomedical Ethics. J. W. Davis et. al. Eds.. Clifton, NJ: Humana Press, 1978: 1‑35]. Ladd criticizes what he calls "legalism" in medical ethics and argues for moving away from a legalistic approach to one based on the notion of responsibility. He writes, "By `legalism' I shall mean: `the ethical attitude that holds moral conduct to be a matter of rule following, and moral relationships to consist of duties and rights determined by rules'"(1).

According to Ladd, legalism leads to the "almost exclusive use of the concept of rights as a category of analysis for problems in medical ethics....As a result,...our view of these problems and how to go about resolving them is unduly narrow and dogmatic, i.e. "legalistic" in the worst sense" (6). The legalistic approach is limited in the medical and other contexts because of its essential reliance on the notion of rights, and therefore, on the limitations of an exclusively "rights-based" approach to ethics. There are, he argues, three essential features of rights which make them particularly inappropriate in the medical context: "(1) the peremptory nature of rights, (2) the particular kind of interpersonal relationship implied in the appeal to rights, and (3) the ethical importance of distinguishing between the possession and the exercise of a right" (16).

First, "unlike other moral considerations, such as appeals to generosity, appeals to rights are appeals that are peremptory; to secure them it is usually permissible to use coercion, either in the form of legal action or in the form of self‑help" (16). Second, rights "represent a relationship between two persons (or parties): the right‑holder and the right‑ower. To have a right is to have a right against someone (or against anyone or everyone)" (17). Third, "one can possess a right only if one can choose not to exercise it" (17): "Strictly speaking, this condition requires that right‑holders (and right‑owers) be competent adults capable of self‑directed choice." Ladd argues that, "these three properties of rights show why it is sometimes quite inappropriate ethically to base medical decisions on the notion of rights alone. For neither medical advice nor patients' requests need be peremptory, that is, advanced as demands backed by force, nor does the doctor‑patient relationship need be an adversary one" (18). Generally speaking, "standing on one's rights is a last ditch stand, to be taken only after communication has broken down or when there was no hope for communication to begin with" (20). "Indeed, if one insists on a right, it will more than likely destroy the relationship altogether because it implies the absence of trust, which may be more important ethically. That is why the appeal to rights is sometimes inappropriate, improper, and immoral" (19).[i] A fourth feature of the notion of rights, acording to Ladd, is that there can be no rights without rules: "it is quite impossible to understand what is involved in asserting a right without examining a rule, or a possible rule, that it refers to or with which it is correlated in some way" (7).

But, rights and the rules they are based on, Ladd argues, are often mistakenly treated as context-independent guides to action: [I]n dealing with issues of medical ethics, and with other kinds of ethics that are concerned with problems of modern life, we must begin by recognizing the contextual character of rules, practices, and concepts. Legal answers are relevant in legal contexts, but may not be so in other contexts; by the same token, some moral categories may be appropriate in one context and quite inappropriate in other contexts. In general, then, we must take into account the context in which problems arise and in which our rules, principles, etc. are designed to operate before we can determine how and to what extent they are valid for the radically new kinds of situation that arise in modern medicine and modern life in general (4).

Ladd contrasts the rule-guided, context-independent, "right-based" approach to moral questions with an approach which he terms "the ethics of responsibility." Responsibility, unlike rights, operates contextually relying on the discretion of the moral agent to determine what is the morally "best" response to make in a given situation. While agents may have recourse to moral and legal "rules" in making these judgments, the ethics of responsibility acknowledges that it may not always be possible to bring any particular action‑situation completely under rules. Situations calling for moral judgments are often unique and require the moral agent to devise a "fitting" response, not just one that follows a rule.

In the medical context, "the unreflective acceptance of an ethics of rights in preference, say, to an ethics of responsibility, inevitably leads to moral confusion and irresponsibility...for the simple reason that the ethics of rights rests on the twin assumptions that, from a moral point of view, someone must have a right and that, in the ultimate analysis, rights relationships can only obtain between equals" (29). As an alternative, he suggests, "we need [an ethics] that covers such things as caring, as providing for another person's needs‑‑in more general terms, we need an ethics of giving and receiving" (22). The ethics of care and the VCP provide the basis for such an approach to normative ethics. It may also enable us to better understand the sorts of moral relationships that exists between "unequals".


Ladd's views are consonant with my own; the doctor-patient relationship is a special moral relationship characterized by the VCP as one in which the physician or caregiver has special responsibilities to protect her patient that cannot be reduced to or entirely derived from the patient's rights. These responsibilities are highly discretionary, but non-optional in the sense that the physician can be rightfully blamed or called to account for failing to fulfill them, even though they are also non-peremptory in that no one can demand that these responsibilities be fulfilled as their right.

The reflections in the last two sections, then serve to cement the proposal that there is a class of moral obligations, what I call moral responsibilities, that are distinct both from peremptory duties derived from the rights of others and from supererogations.

Having reviewed the meaning of the term "responsibility" we are now in a position to consider and respond to several objections to the kind of account I have been developing.

Uni-criterial vs. Multi-criterial Theories of Moral Status

According to Mary Anne Warren, "ascriptions of moral status serve to represent very general claims about the ways in which moral agents ought to conduct themselves towards entities of a particular sort" (Moral Status: obligations to persons and other living things. Oxford: Oxford University Press, 1997, p. 9). She also notes that "a second important feature of the concept of moral status is that the moral obligations that are implied by the ascription of moral status to an entity are obligations to that entity" (10). Having moral status, then, qualifies something as a possible object of the moral responsibilities and duties of moral agents. If the VCP is a fundamental principle of normative ethics, then it tells us that human moral agents have moral responsibilities towards things which have some kind of moral status, in particular, it tells us that we have moral responsibilities to protect them particularly when they are specially vulnerable and in some way dependent on our choices and actions.

Warren contrasts theories of moral status which are uni-criterial with those that are multi-criterial. A uni-criterial theory of moral status takes the view that there is some single criterion that we can use to determine what the moral status of a particular entity is. Various philosophers have proposed various uni-criterial theories of moral status. Kant's view that take rational agency as the sole criterion of intrinsic moral value, and hence moral status, is one example. Other people have proposed that genetic humanity is the proper criterion. Peter Singer, and other utilitarians have proposed that sentience, in particular, the capacity for pain and pleasure, functions as the sole criterion. Still others take other things such as self-consciousnsess or being the "subject of a life" as providing the criterion for moral status. Others, like Albert Schweitzer, propose that life or being alive is the criterion for having moral status.

But Warren argues that none of these uni-criterial theories of moral status can adequately account for the full range of our considered moral intuitions about cases calling for moral discrimination and judgment. Her view, which I find to be convincing is that "any satisfactory account of moral status must be a multi-criterial one, comprising a number of distinct but related principles" (20-21). In particular, Warren argues that:

(1) there is more than one valid criterion of moral status; (2) that there can be more than one type of moral status, with different types involving different obligations on the part of moral agents; and (3) that the criteria of moral status must include both certain intrinsic properties, including life, sentience, personhood, and certain relational properties, which sometimes include being part of a particular social or biological community. (21)

Like Warren, while I think that it would be nice to have a simpler theory of moral status, it is preferable to have one that is more descriptively adequate and which can account for a wider range of robust moral intuitions that competent moral judges have about cases calling for moral discrimination and judgment found in our current commonsense morality. Theories of moral status, however, should not be expected to satisfy everyone's opinions on these matters. Theories and principles are important in normative ethics because they help us correct our biases and guide our judgment in borderline cases. Theory choice in normative ethics is determined by a number of factors other than simplicity and descriptive adequacy alone, so it does not follow that just because a theory of moral status is "simple" or accounts for a number of current moral intuitions, it is correct. Ultimately, any theory of moral status will have to be discursively legitimized and translated into moral and in some cases legal norms that function to guide the judgment and action of large numbers of human moral agents. Philosophers, like Warren and myself, can only point the way towards such socially legitimized norms, we cannot by writng down our thoughts and reasons create them.

Warren does an excellent job in her book of giving philosophical arguments against the various uni-criterial theories of moral status. I am not going to spend a lot of time here rehearsing her arguments, but suggest that the interested reader consult her work. Since I agree with her general approach, I am going to save some time by jumping right into her own multi-criterial theory of moral status. It is, I believe, the best available account of its kind and so can provide a good starting place for further philosophical development.

The goal of this inquiry, recall, is to provide an account of what I am calling "moral patients", that is, those things which can function as the objects of our moral responsibilities. It is necessary to provide such an account in order to determine the boundaries of the moral community for which, I believe, we need an ethics of global responsibility.

Respect for Life Principle

According to the Respect for Life Principle all living things have a moral status and on account of that should not be killed or harmed without good reason. Moral agents, then, can have moral responsibilities towards other living things, for instance, responsibilities derived from the VCP to protect them from harm. But, of course, human moral agents kill and harm other living beings all the time. Should we feel guilty about this? What constitutes a "good reason" for killing, or harming, or otherwise failing to protect something that is alive?

For Warren, the Respect for Life Principle, "imputes no wrongdoing to those who harm living things when there are morally sound reasons for doing so," but the principle alone "does not explain what counts as a sufficiently good reason for harming a living thing" (149). But what counts as "good reason" is a function, at least in part, of that thing's moral status. It is also a function of the moral status of the things that are doing the harming, and their reasons for acting as they do.

In the natural world, biological organisms kill and otherwise harm other biological organisms all the time. The other day I came upon a black snake in my backyard whose slender body was engorged by what I surmised to be some small mice. The snake did nothing morally wrong in eating these mice because snakes are not moral agents and cannot be ascribed moral obligations.

And so it is with respect to all creatures great and small who prey upon one another for their survival. It is a matter of Darwinian survival of the fittest out there in the biosphere. Human moral categories just do not apply, except, of course, to us.

In her book, The Pine Island Paradox: Making Connections in a Disconnected World. (Minneapolis MN: Milkweed Editions, 2004), Kathleen Dean Moore recounts a story that illustrates this point beautifully.

In a wilderness Alaska research camp where Jon works summers, a baby pine marten turned up tangled in a fishnet in a storage shed. A pine marten is a large silky weasel, not a rodent, but close enough for this essay, I hope. And it's fierce -- the hiss, the flashing teeth, the predator's slashing speed. Hearing the baby squeal, the two young aquatic biologists struggled to disentangle it. (110)

Moore says that "what seems important to me about this story is that neither of the men wondered whether they should release the baby marten, whether they should risk the nasty bites to set it free. The only issue was how." The two men obviously felt that it was their moral responsibility to free the baby marten. They were not the cause of its plight, but they understood that without their help this creature was doomed, so they undertook to free it. She continues:

Philosophers say that you can't deduce an ought from an is. It's the old "is-ought" problem that has bedeviled Western philosophy since the eighteenth century, when David Hume explained that from a mere description -- this is the way the world is -- you can't infer a prescription -- this is the way the world ought to be. In one sense he's right: You can stare at the world as long as you want, examine it every which way in all its detail, and it will not reveal to you what ought to be. But that doesn't mean you can't infer what you ought to do, from a description of the way the world is. We do it all the time. Jon made that leap in a flash, the space of a synapse, so fast he was probably never aware of the jump: The marten is caught in the net; therefore, I ought to free it. (111)

Logically speaking the inference is invalid, unless one adds a missing premise, a premise that is a description "not about the world, but about the moral convictions of the person acting in the world" (112). What is this missing premise? Perhaps it is something like the VCP; it is the premise that says moral agents have moral responsibilities to protect vulnerable moral patients when some aspect of their good is dependent on their choices and actions. As Moore writes, "the moral impulse, the silent premise affirming one's own standing as a moral agent, the sharp knife against the tough strands of a net: this may be humanity's unique gift to the universe" (112).

But let's vary this case. Suppose that rather than a baby marten, one finds a Siberian tiger caught in the net. Here most of us would be more inclined to make a different judgment about what we ought to do. Let's assume that the marten and the tiger have the same moral status, they occupy a moral plateau on which we find wild sentient life forms who are not members of mixed moral communities. Based on their moral status alone it seems we should treat them equally. But the difference is that we place ourselves at much greater risk by attempting to free the tiger than we do when we attempt to free the marten. For this reason, most of us would not blame someone who does not try to free the tiger. Why not?

The answer is, I think, that we are also vulnerable living creatures and so we have moral responsibilities to protect ourselves from being killed or harmed. Our reluctance to place ourselves at risk of serious injury or death by means of a tiger mauling us is not just "speciesism" -- an irrational preference for our own kind. It is due to our having a higher moral standing than tigers. We humans occupy the highest moral plateau and thus it would be wrong for us to risk our lives in order to protect a tiger from losing its life.

We can vary the case again. Suppose that you are a hunter-gatherer in the Amazon rain forest and you have trapped a bird in your net. Is it morally permissible to take that animal, kill it, and feed it to your children?

Yes, of course it is, assuming that you and your human family have no other means of survival. Humans are also vulnerable living beings who must nourish themselves in order to survive. Since human beings occupy a higher moral plateau than birds, who are also vulnerable living things that must nourish themselves to survive, there is no wrongdoing in your capturing the bird and killing it for your dinner. But this does not imply that human beings may wantonly kill other living things.

Suppose that rather than a hungry hunter-gatherer, you are a weekend hiker in a national park whose ruck sack is stuffed with granola bars and other edible goodies. May you still trap birds in nets and kill them, say, for your amusement?

I think you may not, even though you can. You may not because trapping and killing birds for your amusement is not a good enough reason to do it. In this case, your standing moral responsibility to protect vulnerable living things becomes the overriding moral principle of action and it outweighs your desire to have some fun. Risking injury or need for nourishment provide good reasons for excusing moral agents from their standing responsibilities to protect vulnerable living things, but wanting to see what it is like to kill a bird for fun does not.

On this view, then, moral agents owe other vulnerable living beings certain responsibilities of care that require them, morally speaking, to avoid harming them, and to protect them from harm, unless there are good reasons for excusing them from such responsibilities. If one is excused from one's moral responsibility, then one does not wrong the bird in killing it. But if one is not excused, then what one does is wrong, not because it adversely affects some human interest, but for the sake of the bird itself, who is a moral patient whose status as a living, sentient creature gives it some claim on our moral concern.

This way of thinking of the matter may strike some people as strange. As Warren remarks, "the adoption of the Respect for Life principle seems to require something more akin to spiritual conversion than to a logical deduction" (151). Some people seem to have made this conversion. Albert Schweitzer, for instance, thought that we have a moral responsibility "without limit towards all that lives" (Quoted in Warren, p. 31). Someone who subscribes to this ethics of reverence for life, "tears no leaf from a tree, plucks no flower, takes care to crush no insect." Warren claims that Schweitzer's ethics, "permits no principled distinction between the moral status of plants and that of animals -- or humans." (33) It treats all living creatures as equally valuable, and as equally deserving of our concern and respect.

But Warren (and I) reject this form of radical biological egalitarianism. The reason, she says, is clear:

Any human society which treated ordinary acts of food preparation, personal hygiene, and medical care as the moral equivalents of mass homicide would jeopardize its own survival. For if, on the one hand, members of society adopted the view that the destruction of micro-organisms is as serious a crime as we consider homicide to be, and consistently sought to prevent such destruction, then the health of the human population would suffer severely. And if, on the other hand, they came to see homicide as no more serious a wrong than we consider the destruction of micro-organisms to be, then the society's survival would be threatened by uncontrolled intra-human violence. (37)

All men may be created equal, but all living things aren't. Or at least, all organisms do not have the same moral standing and stature as human moral agents do. Micro-organisms, plants, insects, birds, mammals, and human beings occupy different moral plateaus. Other things being equal, we should treat entities on each plateau impartially with due regard for their particular kind of moral standing. But we do not have to feel guilt or remorse about sacrificing living things with lesser moral stature for those with greater moral stature when there is a good reason for doing so.

But, if so, then doesn't this view boil down to the traditional homocentric theory of moral status where human interests always prevail over those of non-human animals? Neither Warren nor I think so. According to the traditional homocentric view, the value of other living things is only a derived value; their value depends upon their relationship to our interests. But, on the Respect for Life principle, other living things have intrinsic moral value "in themselves". That is, it is wrong to wantonly harm them for their sakes, not just for ours. As Warren writes,

It makes more than verbal difference whether we believe, on the one hand, that all living things have a claim to our consideration, however, modest; or, on the other hand, that plants and other non-sentient life forms should be protected only when they have demonstrable value to human beings. If we believe that the needless destruction of living things is a wrong against them, not just a possible wrong against other human beings, then we will be more likely to search for ways to reduce the needless killing that we do, individually and collectively. We will not be permanently content with methods of agriculture, animal husbandry, fishing, manufacturing, mining, transportation, energy production, forestry, recreation, flood control, and waste disposal that cause the needless destruction of harmless plants and animals. Respect for life may, therefore, substantially improve humanity's chances of surviving and flourishing into the deep future.(150)

This is a homocentric argument for a biocentric ethics. There is nothing wrong with it since it is, after all, for human moral agents that we fashion, critique, and re-create, our ethical philosophies, since we are the only things in the universe that we know of to which they apply. The reason why we should embrace a biocentric ethics now is that doing so will help our global civilization make a transition to a sustainable form of development. Sustainability means different things to different people, but in this context, it means a future in which life on earth, including human life, will continue to flourish. It is difficult to think of anything that could constitute a higher moral purpose than this.

Responsibilities to Oneself

If you are reading this you are most likely a moral agent. Since moral agents can also be moral patients, it would seem to follow on my theory that moral agents can have moral responsibilities towards themselves.I will call these self-regarding responsibilities.

Most people are willing to acknowledge that they owe it to themselves to promote their own happiness, to care for their physical health and emotional well being, and to realize their potential as human beings. In some systems of ethics, such responsibilities to oneself are taken as the primary moral duties which all persons have. Self-regarding responsibilities are sometimes thought to arise somehow directly from our nature as rational beings. A basic ethical theory known as ethical egoism proposes that all moral agents owe it to themselves to maximize their own self-interest.

But there is a problem with this view. Duties to benefit oneself are not ordinarily thought of as moral duties, but rather are considered to be matters of prudence. Many moral philosophers distinguish between the demands of morality, which concern what we owe to others, from those of prudence. When speaking about moral responsibilities we are concerned to understand how moral agents should treat others who have wills of their own. But in discussing prudence, one is speaking about how one manages ones own will. One's own will is not generally regarded as an alien will; it is supposed to be under ones own control. Morality is concerned with controlling wills that are alien to ones own will, and so the demands of prudence cannot be moral demands.

The way in which Goodin framed his Vulnerability Principle reflects this distinction between prudence and morality:

(VP): Moral agents acquire special responsibilities to protect the interests of others to the extent that those others are especially vulnerable or in some way dependent on their choices and actions. (Italics added)

As it is stated the VP excludes what I am calling self-regarding responsibilities. I think this is a fundamental problem with Goodin's statement of the vulnerability principle.

On the account I will prefer, there can be self regarding moral responsibilities, or responsibilities of "self care". Not all human individuals have such self regarding moral responsibilities; infants and young children, and persons suffering from various forms of physical or mental incapacity, for instance, do not have them or have them in a different way than do fully competent autonomous moral agents. So if there are self-regarding responsibilities it is necessary to specify how one comes to have them, when one is excused from them, and what specific kinds of duties they entail. Additionally, like other moral responsibilities, responsibilities to oneself, and the duties they entail, must always be balanced against responsibilities owed to others. However, one cannot lead a moral life and completely neglect the responsibilities which one owes to oneself.

So it will be necessary to reformulate the Vulnerability Principle so as to allow these kinds of self-regarding moral responsibilities to fall within its scope. I propose the following re-formulation.

(VCP) Moral agents acquire special responsibilities to care for and protect the interests of moral patients to the extent that those moral patients are especially vulnerable or in some way depending on the actions and choices of those moral agents for their care.

Since moral agents are also moral patients it follows that under the VCP moral agents can acquire special responsibilities to care for and protect their own interests, particularly when they are especially vulnerable or are depending on themselves for their own care or for the protection of their own interests. We can continue to refer to this subclass of moral responsibilities as prudential, understanding that they are now regarded as a special kind of moral responsibility. There are, I believe, situations in which we are vulnerable to ourselves and are also depending on ourselves to protect our own interests. Duties of prudence arise from the VCP, on this view, as a subspecies of the kinds of special moral responsibilities that derive from considerations of vulnerability and dependence to the extent that moral agents are vulnerable to and depending on themselves for their own care and protection.

Moral Gravity

The theory of moral status I have developed is designed to allow us to make moral judgments having to do with the treatment of members of various kinds of moral patients. A moral patient's particular grade of moral stature is proposed as a normative factor that determines, at least in part, whether or not certain kinds of acts or omissions regarding that individual are morally permissible, required, or forbidden.

I have illustrated how my notion of moral stature can be applied in several cases, e.g., human embryos, sentient fetuses, and infants, I have argued that while all of these kinds of moral patients have intrinsic value and are human beings, they have different grades of moral stature, and thus may ethically be treated differently. I have also illustrated how this account applies in cases of various kinds of non-human moral patients. Psychological organisms with greater moral stature, e.g. primates, cetaceans, elephants, etc., deserve greater respect than those with less developed or complex psychological capacities. Organisms that are not sentient, that have only the basic level of intrinsic moral standing, can be considered as moral patients, but the degree of respect they command is weaker than that of psychological organisms. Moral agents, particularly, autonomous moral agents, that is, persons like you and me, have the highest grade of moral stature. We are on a moral plateau that gives us both rights and responsibilities as members of the human moral community. Our responsibilities to protect vulnerable persons are weightier than comparable responsibilities to protect nonhumans on my view, other things being equal.

But it is necessary to introduce another normative factor to make this account better able to capture and explain our moral intuitions, a factor, that I will refer to as moral gravity. The intuitive idea of moral gravity is that some kinds of things that affect a moral patient's well-being, or interests, are more important than others. The notion of the "gravity" of different kinds of human rights violations can be used to illustrate this idea.

Amnesty International (an organization I have been associated with in various capacities for many years) has a mission statement in which it commits itself to oppose certain kinds of grave abuses of human rights. What are grave abuses? Well, torture, rape, murder, genocide, and arbitrary imprisonment are definitely on this list. What is not on the list of grave abuses of human rights that Amnesty International works on are, e.g. copyright infringement, infringements of personal privacy, denial of voting rights, and various kinds of labor abuses, for instance, requiring workers to work more than 60 hours a week. The international standards promulgated by the ILO (International Labor Organization) specify that the normal work week should be no more than 48 hours per week, with a maximum of 12 additional hours in voluntary overtime permitted. When workers are required to work for more than 60 hours per week, it is a human rights abuse. But it is not a grave abuse in Amnesty's view. There are probably several hundreds of millions of workers in the world who regularly are required to work more than 60 per week (think of young Wall Street lawyers in high-powered law firms, or medical interns at busy urban hospitals, not just low-wage workers in Asian sweatshops). In terms of the numbers of people whose human rights are affected, the abuse of overtime is a major human rights issue. But because the form of abuse is itself not considered "grave" it does not get much attention from human rights organizations like Amnesty International.

It is possible to think of moral gravity as a property of certain kinds of interests. The interests that have the greatest gravity are the ones most closely connected to an individual's survival, well-being, and freedom. These kinds of interests are sometimes called "ultimate interests" because they are typically the sorts of things that people regard as ultimately valuable or valuable for their own sake. They are interests like: not being killed, not being tortured, not having one's freedoms restricted, not having to be a slave, having enough to eat, having a safe place to live, and so forth. There are lots of other interests that people have that do not have much gravity as compared to these kinds of ultimate interests. For instance, I am interested in riding bicycles. I like it as a form of exercise and a form of transportation. I really enjoyed living in Copenhagen for a few months last fall because I got to ride my bike everyday on very safe bicycle paths along with lots other people who like me have a liking for biking. I would be unhappy if I could not ride my bicycle, but, as the saying goes, it wouldn't kill me not to. I could live without it.

How then does moral gravity affect our moral judgments? It seems that the straightforward answer to this question is that actions which threaten or harm an interest which has greater moral gravity to the moral patient concerned are more serious than those that threaten or harm an interest that has lesser moral gravity. The prohibition against killing other persons is the most serious moral prohibition because the interest that we have in staying alive is considered to be the ultimate interest which has the greatest moral gravity. If we cease be alive, nothing much else matters. The prohibition against theft of personal property is far less serious than the one against killing and other forms of physical assault against persons, e.g. rape, torture. That we have an interest in our property is clear, but whether it is an ultimate interest is doubtful. Like other Americans, I own a lot of things. But there is nothing that I own that I could not live without. If you steal my bike it harms me, but the interest involved in this case just isn't that grave. I can get another one.

But the gravity of the theft of a bicycle can be different for different people. Consider the classic film The Bicycle Thief in which the theft of a bicycle leads to a cascade of hardships for a whole family in post-war Rome. In this case, the bicycle was essential for a man's job that he needed to feed his family, and so the theft of his bicycle was a graver offense than if someone were to steal my bicycle. So this normative factor is going to be context-sensitive and it is therefore going to be difficult to say general things about it. Nevertheless, I think the intuition we have about it is sufficiently clear. It is possible to extend the idea of moral gravity to other kinds of moral patients.

Nonhuman animals which are sentient have different kinds of interests than do human beings, but it seems reasonable to suppose that they are also interested in some things ultimately such as avoiding pain and suffering, keeping their freedom, having enough to eat, having a safe place to live, and so forth. Non-psychological organisms, such as plants and bacteria, do not have interests, but may be said to have a "good of their own". A plant needs sunlight, water and soil in order to grow. These things constitute the good for plants such that depriving a plant of these things would kill it, while doing other things to it, like picking its fruit or pruning its branches, would not. It is possible to think that non-living moral patients, that is, those which have only derived moral status, for instance, works of art, also have goods of their own. It is better for a statue to be intact than to have parts missing or destroyed, for instance. The act of destroying a valuable work of art can be consider as a moral offense, but not nearly as grave as killing a sentient life form or a person because works of art only have derived moral status.

Assuming that the idea of moral gravity is sufficiently clear, we can now attempt to state the way in which moral gravity interacts with moral stature. All normative ethical theories require an account of the ways in which different normative factors interact with one another. This is as true for monistic theories, ones that propose a single fundamental moral principle, as well as for pluralistic theories, those that propose that there are several fundamental moral principles. Although I am a pluralist, I have been focusing attention on just one moral principle, the Vulnerability-Care Principle. But nevertheless, it is necessary to provide an account of how the different normative factors associated with the VCP interact with one another because, like other principles that apply generally, this principle can generate conflicts with itself. That is, we can have conflicting moral responsibilities to protect different vulnerable moral patients.

We can think of moral gravity and moral stature as two normative factors that combine to determine the weight of a moral responsibility to protect the vulnerable. In physics weight is a function of an object's mass times the force of the gravitational field that it is in, as given by Newton's Second Law:

F = ma: the net force on an object is equal to the mass of the object multiplied by its acceleration.

It would be nice if there was some way to quantify the notions of moral gravity and moral stature so that we would be able to calculate the precise weights of various kinds of moral responsibilities. But alas, I know of no such calculus and doubt that we can ever have one, because gravity and stature are not the only two normative factors and the VCP is not the only fundamental moral principle. But we can use the notion of moral weight analogously to the way we use and understand weight in physics.

In this case, the moral gravity of an interest or good is analogous to the A factor, and the moral stature of the patient is analogous to the M factor. The moral weight of a responsibility or obligation is then obtained by (figuratively) multiplying the A and the M. When ones does this one gets linear graphs that indicate that when one holds M (stature) constant, greater A (gravity) yields greater F (moral weight). Similarly, when one holds A (gravity) constant, greater M (stature) yields greater weight, and conversely in each case. The first relationship has been illustrated already with the examples of grave versus non-grave human rights abuses. The responsibility, say, not to kill another person comes out as weightier than the responsibility not to force people to work excessive overtime. The second relationship has been discussed earlier with respect to various duties towards non-human animals. For instance, I opined earlier that the duty not to kill a human person is weightier than a comparable duty not to kill a nonhuman psychological organism. I also argued that since not all human beings have the same moral stature, it is a less weighty offense to kill a human embryo than it is to kill an sentient fetus, which is less weighty an offense that to kill an infant or child who is assumed, according to the Human Rights Principle, to be a full member of the human moral community. Killing non-psychological organisms is not a very weighty matter because while they are moral patients and have some moral standing, their moral stature is slight compared to other kinds of moral patients.

Using this formula it is fairly easy to generate intuitive judgments about the weight of different moral responsibilities. One can test the theory by making up trolley car examples in which a moral agent (say, yourself) is in a forced choice situation in which you must choose between, say, killing a pet rabbit and killing a person. Tie them both to the tracks and let your moral intuitions tell you which way you would steer the trolley. If you would choose the kill the bunny, then your intuitions agree with what F=MA would predict. If not, then there is something wrong with my theory or something wrong with your moral intuitions, assuming, of course, that other things are equal, that is that there are no other morally relevant normative factors that might affect the judgment.

For instance, suppose that the person is a mass murderer and that if you don't kill him by running over him with the trolley he will carry out an evil plan in which 1000 innocent people will be killed. I bet you didn't think of that right off the bat, did you? This would, however, introduce another normative factor besides M and A and would complicate the interaction of these factors with our moral judgments. So let's just hold these other possibly relevant normative factors in abeyance for the time being and keep it simple.

When we do this, however, things still get complicated quickly when we vary the M and the A simultaneously. How would you decide a forced choice between say, depriving a psychological organism of its freedom, say to keep them in a cage for the amusement of tourists, versus not having a way for tourists to be amused by seeing animals in cages? Presumably, the interest that the animal has in its freedom is pretty grave, that is, it has a high A, while the interest that tourist have in being amused is not very grave at all, it has a low A. But tourists, assuming they are fully autonomous moral agents, have greater moral stature than nonhuman animals, so their interests, even their non-grave interests will count more.

So what is the answer you come to? Is it morally permissble to keep animals in cages at tourist attractions or isn't it? On the basis of these two normative factors alone, I would judge that it isn't. But, obviously even this simple case can get more complicated if you begin asking questions about, say, what happens to the animals if you close the zoo? and, What happens to the livelihoods of the people who own and operate the zoo? But, we have agreed to leave out these other normative factors. If we do this, then it would seem to follow that sometimes human interests ought to be sacrificed in order to protect the important interests of nonhuman animals.

This is precisely the result I wished to obtain. It is the result that follows from adopting a biocentric rather than a homocentric theory of moral status. Things get even more complicated when we vary the numbers of individual moral patients involved. Suppose that rather than just one chimpanzee on the trolley tracks we have 100 of them. And suppose on the other side we have one human person. While the lives of chimps are less intrinsically valuable than those of human persons, the lives of 100 chimps are collectively more valuable than the life of one person. If we had 100 people on one of the tracks and 1 on the other, I doubt many of you would choose to steer the trolley towards the 100. But what if there are 100 chimps, or dogs, or parrots? Should we still prefer to save one human life if doing so means ending 100 animal lives? Suppose it is not 100, but 1000 animal lives? Or pick an even higher number of your choice. If there is no threshold at which you would be willing to say it is morally preferable to sacrifice a human being to save some number of sentient animals, then my theory is not for you.

You want a theory that assigns zero weight to our moral responsibilities towards nonhuman animals and nature, that is, you want a homocentric ethics of the traditional kind. On this view, no matter how grave the interest or good of an nonhuman living thing is to its survival or well-being, its interests can never override any human interest, no matter how trivial. One gets this sort of view by assigning zero to the animal's moral stature for M factor. If you multiply zero times anything you get zero as the moral weight of the responsibility to protect. On the other hand, if you ascribe any non-zero number to an animal's M, their moral stature, and also assign some non-zero number to its A, or the gravity of its interest or good, then by adding more individuals we increase the moral weight of the responsibility to protect. At some point or another, the weight will be greater than those generated by conflicting human interests, and the latter will be out-weighed by the former. That is, as moral agents, we will be obliged to sacrifice some human interests in order to fulfill an even weightier responsibility to protect some vulnerable nonhuman organisms.

The principle of moral weight here operates to settle the conflict between two different moral responsibilities generated by the VCP. In general, the greater the weight of a moral responsibility, the harder it is to override it in favor of a conflicting moral responsibility. On my view, there are few if any non-overridable moral responsibilities; all or nearly all of them are defeasible under various circumstances in which combinations of normative factors yield great weight to some other moral responsibility requiring a different course of action. I do not think that moral stature and gravity are the only two normative factors that determine the weight of our moral obligations. There are other normative factors that can affect our "all things considered" evaluation of moral cases.

But I do think that these notions capture something important and useful about the way in which we ordinarily think about responsibilities to protect various categories of moral patients. The notions of moral stature and moral gravity can be part of a new philosophical vocabulary for making cross-species comparisons of our moral responsibilities. It can also, I will argue, be used along with other normative factors, to help explain other some moral intuitions we have about our responsibilities to one another as members of the human moral community.